TSUserDocs 2.2

USER DOCUMENTATION
Torrent Suite 2.2
Revision Date May 2012
For Research Use Only. Not intended for any animal or human therapeutic or diagnostic use.
Information in this document is subject to change without notice. LIFE TECHNOLOGIES AND/OR ITS AFFILIATE(S) DISCLAIM ALL WARRANTIES WITH RESPECT TO THIS DOCUMENT, EXPRESSED OR IMPLIED, INCLUDING BUT NOT LIMITED TO THOSE OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. IN NO EVENT SHALL LIFE TECHNOLOGIES AND/OR ITS AFFILIATE(S) BE LIABLE, WHETHER IN CONTRACT, TORT, WARRANTY, OR UNDER ANY STATUTE OR ON ANY OTHER BASIS FOR SPECIAL, INCIDENTAL, INDIRECT, PUNITIVE, MULTIPLE OR CONSEQUENTIAL DAMAGES IN CONNECTION WITH OR ARISING FROM THIS DOCUMENT, INCLUDING BUT NOT LIMITED TO THE USE THEREOF. LIMITED USE LABEL LICENSE No right to resell this product or any of its components is conveyed expressly, by implication, or by estoppel. For information on obtaining additional rights, please contact outlicensing@lifetech.com or Out Licensing, Life Technologies, 5791 Van Allen Way, Carlsbad, California 92008. NOTICE TO PURCHASER: DISCLAIMER OF LICENSE Purchase of this software product alone does not imply any license under any process, instrument or other apparatus, system, composition, reagent or kit rights under patent claims owned or otherwise controlled by Applied Biosystems, either expressly, or by estoppel. TRADEMARKS The trademarks mentioned herein are the property of Life Technologies Corporation or their respective owners. Windows and Internet Explorer are registered trademarks of Microsoft Corporation in the United States and other countries. Mozilla is a trademark of Mozilla Foundation Corporation. 2012 Life Technologies Corporation. All rights reserved. Document Revision Date May 2012 Rev. A
USER DOCUMENTATION
Torrent Suite 2.2
Revision Date May 2012
For Research Use Only. Not intended for any animal or human therapeutic or diagnostic use.
Information in this document is subject to change without notice. LIFE TECHNOLOGIES AND/OR ITS AFFILIATE(S) DISCLAIM ALL WARRANTIES WITH RESPECT TO THIS DOCUMENT, EXPRESSED OR IMPLIED, INCLUDING BUT NOT LIMITED TO THOSE OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. IN NO EVENT SHALL LIFE TECHNOLOGIES AND/OR ITS AFFILIATE(S) BE LIABLE, WHETHER IN CONTRACT, TORT, WARRANTY, OR UNDER ANY STATUTE OR ON ANY OTHER BASIS FOR SPECIAL, INCIDENTAL, INDIRECT, PUNITIVE, MULTIPLE OR CONSEQUENTIAL DAMAGES IN CONNECTION WITH OR ARISING FROM THIS DOCUMENT, INCLUDING BUT NOT LIMITED TO THE USE THEREOF. LIMITED USE LABEL LICENSE No right to resell this product or any of its components is conveyed expressly, by implication, or by estoppel. For information on obtaining additional rights, please contact outlicensing@lifetech.com or Out Licensing, Life Technologies, 5791 Van Allen Way, Carlsbad, California 92008. NOTICE TO PURCHASER: DISCLAIMER OF LICENSE Purchase of this software product alone does not imply any license under any process, instrument or other apparatus, system, composition, reagent or kit rights under patent claims owned or otherwise controlled by Applied Biosystems, either expressly, or by estoppel. TRADEMARKS The trademarks mentioned herein are the property of Life Technologies Corporation or their respective owners. Windows and Internet Explorer are registered trademarks of Microsoft Corporation in the United States and other countries. Mozilla is a trademark of Mozilla Foundation Corporation. 2012 Life Technologies Corporation. All rights reserved. Document Revision Date May 2012 Rev. A
Version 2.2
Technical Notes and Whitepapers

Torrent Suite Technical Notes and Whitepapers
Welcome to the Torrent Suite User Documentation Technical Notes and Whitepapers Page. Contents Technical Note - Analysis Pipeline Technical Note - Filtering and Trimming Technical Note - Quality Score Technical Note - TMAP Alignment White Paper - Torrent Variant Detection Algorithms
Technical Note - Analysis Pipeline

Technical Note
Torrent Server Analysis Pipeline
The Analysis Pipeline is the primary software used to process raw Personal Genome Machine (PGM) sequencer acquisition data to produce read files containing high quality bases. This document provides an overview of the software modules and concepts applied at various stages of the data processing. Introduction
The Torrent Server Analysis Pipeline, the "Pipeline", processes raw acquisition data from a PGM sequencer run, and outputs base calls in both SFF and FASTQ file formats. The Torrent Browser provides a web interface to the process, including many metrics, graphs, and reporting features derived from the Pipeline results. The following figure shows a conceptual overview of the Pipeline:
The following sections provide additional, detailed information about the Analysis Pipeline modules that work together to convert the raw acquisition data into high-quality base calls and reads. Background
- Page 3 Copyright 2012 Life Technologies Corporation
Version 2.2
Although the focus of this document is on the analysis pipeline itself, a few concepts are presented here for review. During a PGM sequencer run, for each nucleotide flow, one acquisition file is generated. Each acquisition file contains the raw signal measurement in each well of the chip for the given nucleotide flow. So each acquisition flow, for an Ion 314 chip, contains roughly 1.5 million separate incorporation events. A series of such acquisition files represents the roughly 1.5 million possible reads. The analysis pipeline converts these raw signal measurements into incorporation measures, and ultimately into base calls for each read. The raw measurements are the conversion of the raw pH value in each well into a voltage converted into a digital representation of that voltage. This measurement over the entire chip occurs many times per second. Analysis Pipeline Modules
The following figure shows the high level modules within the analysis pipeline. Each module accepts specific inputs and produces specific outputs, described in more detail, below.
DAT Processing
The DAT processing module deals directly with raw acquisition files (acq_*.dat). The following figure shows an Ion 314 chip heat map, which indicates percent loading across the physical surface:
DAT processing performs the following functions: Raw acquisition data loading into memory.
Version 2.2
This includes decompressing the data files. Data is compressed when it is streamed off the PGM sequencer: An optional dynamic frame rate compression mode whereby various portions of the incorporation event throughout the nucleotide flow duration are captured at different frame rates. The variable frame rate permits biologically specific events to be captured with high resolution while averaging multiple frames where appropriate, giving a reduction in overall file size. A compression approach may use a keyframe/delta compression technique whereby an initial value is stored, followed by the changes in that initial value, rather than storing the actual value each time. This results in a nearly 2x reduction in file size. Raw measurement offset corrections. Raw acquisition data are stored using the values output by the chip. Each well has its own reference value, and to compare well to well, the two wells may use a common reference. The offset correction takes the average of the first few frames within each acquisition file and subtracts that value from each well, thus permitting well measurements to have a common reference value. Pinned well identification. Due to the nature of the chip output voltages, a range of values exists for any given chip. The PGM sequencer calibration code brings the majority of wells within range of the hardware analog to digital converters. The output is a distribution of values centered around the center voltage. Wells that reside outside a selected distribution are considered "pinned" (functional wells outside of the range of the Analog to Digital Converter (ADC)). These pinned wells typically represent fewer than one percent of the total available wells. Excluded Wells. Various flow cell configurations often make tradeoffs on flow velocity profiles and chip coverage areas. For the Ion 314 chip, for example, a percentage of the wells are covered by the flow cell and are not fluidically addressable. A mask is loaded, per chip type, to mark those wells as excluded so they do not complicate down-stream processing of the chip. Default Sequencing Keys and Flow Order The next part of the classification module involves identifying particle wells as test fragments (TF) or library fragments. To understand this part of the algorithm, the concept of "keys" and flow order must first be understood. The following table shows the default library and TF sequencing keys in both base-space and flow-space for the default, TACG, flow order: Base-space Library Key TF Key TCAG ATCG Flow order TACG vector 1010010X 0100101X
The "X" in the vector column indicates that the value is at least a one, but could in fact be higher because the next base after the key could also match, thus creating a longer homopolymer.
Version 2.2
Classification
Classification is the process of determining the contents of each well on the chip. The primary purpose of classification is to determine which wells contain particles with template for sequencing and which wells are empty for use in background estimation during signal processing. The figure below illustrates the nested classes of classification.
A well can either be empty, contain a particle or be flagged as an outlier. Wells are flagged as outliers if they do not appear to be responding appropriately to the nucleotide, are not fitting the model well or are wells marked empty that appear to be buffering. For wells containing a particle, the process additionally determines if the particle is: A library particle. A test fragment (control) particle. A dud particle (a particle that does not produce a sufficiently strong sequence signal). It is important to note at this point that classification is not simply processing the entire chip as one group. As is mentioned for other pipeline modules, the chip is processed in smaller regions. An Ion 314 chip, for example, is divided into on the order of 50x50 well regions, resulting in about 625 total regions. This enables processing many small regions in parallel and taking advantage of multi-core/multi-process compute nodes that have such capabilities. More importantly, regions in the chip are processed that contain fluidically similar wells, where smaller regions tend to be relatively homogeneous and facilitate comparing wells to each other within a region. Key Classification Each well is checked to see if it is producing sequence that matches any one of the keys. A model is fit and the key with the best fit is assigned to that well. If no key has a high enough score then the well is not assigned to the live classification and will not be processed downstream. With the default keys and flow order, above, there are two vectors in flow space:
Nuc Flow: TACGTACG Library: 1010010 Test Frag: 0100101
Each key is tested to see if it fits the expected behavior of similar signals in the 1mer flows and no signal in the 0mer flows. Note, as mentioned above for these keys, the last 'G' flow is not used in the model fitting. There could be
Version 2.2
another G following in the template sequence which would mean that in some cases the incorporation may be a 2mer, 3mer, etc. and not fit the model. A key must have at least two 1mer flows and two 0mer flows to be valid for sequencing. Mask The output of the classification module is a mask object (and file as bfmask.bin) that contains bit flags for each well, indicating the contents of each well.
Signal Processing
The signal processing module focuses on wells containing particles, which is now using the mask information in addition to the raw acquisition data. The following figure shows a conceptual representation of the signal measured in a well during an incorporation event.
This signal has additional components, referred to as "the background". This background part of the signal is present during each flow and can vary over time, across the chip, and during an acquisition. The signal processing problem can be thought of as having two parts: The first part involves deriving the background signal that would have been measured in a given well, had no incorporation occurred. The second part involves subtracting (or fitting) the background signal and examining (or fitting to) the remaining signal. The output of the incorporation fitting model produces the estimate of the incorporation at each nucleotide flow for each well. At this point, the output from the analysis pipeline is a raw incorporation signal measure per well, per flow, stored as a 1.wells file.
Basecalling
The basecaller module performs signal normalizations, phase and droop estimations, signal corrections, and declares bases for each flow of each well. It outputs non-incorporation events in addition to incorporation events. The SFF output file stores all such calls. The following figure shows basecaller phase correction:
Version 2.2
Droop estimation Observed signal droop can be attributed to DNA polymerase loss that can occur during a sequencing run. Typically, such loss is experienced only during incorporation events, and typical values are in the 0.1 to 0.2% range over the course of a run. By averaging groups of reads in a region together and averaging their measured signals, an exponential can be fit to the resulting curve, and the rate of signal loss over time is extracted. Thus, an estimate of the polymerase loss during incorporation events can be determined. Phase estimation Once the droop has been established, it is used in the phase model as a constant for each read. Droop estimates can still vary across the chip, but is assumed fixed for each processed region. The model then fits the carry-forward and incomplete extension parameters for each read, over a limited number of flows and excluding certain flows. The output from this fit is an estimate of the carry-forward (CF) and incomplete extension (IE) for each well. The values are averaged over small regions to reduce errors and noise in the fit. The output CF and IE values are used as inputs to the solver. Solver The solver uses the measured incorporation signal, the phase estimates, and the droop estimates to determine the most likely base sequence in the well. The solver achieves this by repeatedly constructing predicted signals for partial base sequences and evaluating their fit to the measured signal. The predicted signals are generated by a highly efficient dynamic programming algorithm. These signals are further used by the solver to perform a branch-and-bound search through the tree of partial base sequences. The solver keeps a list of promising partial sequences. At each step, it selects the most promising sequence and extends it by one base into four new longer partial sequences. These new candidates are checked against a set of fit criteria, and those that pass them are added to the list. The list size is limited and the excess sequences are discarded based on a fit heuristic. When none of the entries on the list yield a better fit than the best sequence examined thus far, the tree search concludes. Normalization The solver expects that in absence of phasing and droop, the measured signal for flows incorporating a 1-mer is very close to 1, and for non-incorporating flows very close to zero. In practice, the measured incorporation signal can deviate from this assumption by an additive offset and a multiplicative scaling that slowly varies with the flow number and is different for each read. This is addressed by normalizing the signal before it is passed to the solver. Before the first execution of the solver, a key normalization preliminarily scales the measurements by a constant factor. The factor is selected to bring the signal of the known expected 1-mers produced by sequencing through the key as close to unity as possible. Once the solver has guessed a sequence, the corresponding predicted signal is leveraged for adaptive normalization. Adaptive normalization compares the predicted signal to the measured incorporation signal to fit the flow-varying additive and multiplicative terms, and subsequently generates an improved normalized signal with the distortion removed (or at least reduced). Adaptive normalization is invoked iteratively with the solver for several rounds, allowing the normalized signal and the predicted signal to converge towards the best solution.
Version 2.2
Read Filtering and Trimming
Read filtering involves generation of various quality metrics for each base call, quality metrics for each read, and using those metrics to filter out low quality reads and trim reads back to acceptable quality levels. Additionally, adapter trimming is performed on the ends of each read. See: Technical Note - Filtering and Trimming Technical Note - Quality Score Per-base Quality Scoring Per-base quality scores are assigned to each read, and written to the SFF file along with the read itself. The per-base quality scores are assigned by calculating various metrics for the read and using those metrics as lookup indexes into a pre-defined quality lookup table established through prior system training. The metrics often involve estimates of accuracy at the current base, flow, and earlier or later bases or flows for each read. Filtering Filtering reads involves calculating an overall quality metric representing the basecaller capability to accurately correct and base call the raw signals. Reads that have calculated low quality are not written to the SFF file. Low quality reads are typically mixed reads or very low copy-count reads that produce low quality sequence such that they do not fit the expected incorporation model. Quality and Adapter Trimming Using processing algorithms, a read can be trimmed back until an acceptable level of quality persists for the entire read. A read is also examined to determine if the B-Primer adapter sequence or a portion thereof can be identified with high confidence within the read. If a matching sequence is found, the read is trimmed to the base immediately prior to the start of the adapter. Trimmed SFF file The output of this stage is an SFF file that contains only the filtered reads and, for each read, the adapter and quality trimming markers are set appropriately. This file is used to generate the FASTQ file.
Alignment QC
Version 2.2
The alignment QC stage involves alignment of reads produced by the pipeline to a reference genome, and extracting metrics from those alignments. Currently, the TMAP aligner is used to align reads. As inputs to the aligner, some or all of the reads produced in the pipeline are used, along with the reference genome and index files. The output is a BAM file. This output BAM file is processed to extract various quality metrics, including estimates of Q20 bases, estimates of number of reads at or above 100Q20. The results of the Alignment QC module are viewable using the Torrent Browser web interface in the Reports summary page. The key output component is the BAM file. Data Files Produced
See Quickstart On Dataflow for a listing of the files output at each stage or module, along with their expected size for typical runs.
Technical Note - Filtering and Trimming

Technical Note
Filtering and Trimming
The Ion Torrent system applies some quality checks to sequence reads before writing them out. Reads are tested to see if they are possibly the result of mixed DNA templates on an Ion Sphere Particle in a given well, or if they are generally of low signal quality. The 3' ends of reads are scanned for matches to the adapter sequence and for regions of trailing low quality. Only the high-quality portions of reads passing through these checks are written out to the SFF and FASTQ files. Introduction When processing data from the Personal Genome Machine (PGM), the identification and removal of low-quality or uncertain base calls is an important consideration for delivering accurate results. In the Torrent Suite analysis software, low-quality bases are removed from the output by: Filtering out entire reads. Trimming low-quality 3' ends of reads. The operations described in this document are applied as post-processing operations after the initial base calls have been estimated. Read Filtering The goal of read filtering is to remove reads estimated to contain low-quality base calls. The read filter is targeted at removal of reads that are derived from wells with non-clonal DNA template populations, such as mixtures of different DNA templates.
Removal of Mixed-template Reads
An Ion Sphere Particle is clonal if all of its DNA fragments are cloned from a single original template. All the fragments on such a bead are identical, and they respond in unison as each nucleotide is flowed in turn across the chip. Using the standard PGM flow order, about 44% of all nucleotide flows yield a positive signal from the chip.
Version 2.2
The data from the chip are normalized so that the 1-mers in the key yield a unit signal. 1-mers in the insert also yield a signal value of 1, 2-mers yield a signal of 2, 3-mers 3, and so on. Data for a clonal ISP from a 260 flow PGM run are shown below. Slighly more than half of the flows cluster around zero, and the rest are positive. The positive flows cluster around integer values. This pattern is most clearly seen for the earliest flows, before phase effects cause some templates to respond out of sync with others.
What happens if an ISP carries fragments derived from two distinct templates? The next figure shows data from such a bead. This bead harbors two distinct populations of clones. On some flows, only one of the populations responds with an incorporation. Since only half the templates are incorporating, a 1-mer now yields a signal value of about 0.5 instead of 1.0; a 2-mer yields a signal of 1.0, instead of 2.0. One strand may incorporate a 1-mer, and the other a 2-mer, yielding a signal of 1.5. A signal of 0 is still possible, if both populations fail to incorporate. But such 0-signal flows are less frequent than in the clonal case, accounting for only about 19% of flows. In summary, polycl onal beads exhibit more non-zero flows, and the signals no longer cluster exclusively around integers.
It is also possible for emPCR to generate beads carrying fragments derived from a large number of library templates. Such beads are referred to as super-mixed. The signal from a super-mixed bead is shown in the next figure. Now almost all flows have non-zero signal, and the signal does not cluster around integer values at all.
Version 2.2
These kinds of patterns can be used to identify polyclonal beads in software. Torrent Suite does this by computing two scores for each well. The first score is simply the percentage of flows that return a non-zero signal. Because of the possibility of noise in the data, only flows with a signal value greater than 0.25 are counted as positive. This score is computed for each bead for flows 12 through 72, and is referred to as percent positive flows (PPF). The second score captures the degree to which the signal clusters around integer values. For each flow, the difference between the signal and the nearest integer value is computed. This distance is squared, and the squared values are summed over flows 12 through 72. This score is low if the signal clusters around integer values, and higher otherwise. This score is referred to as the sum of squares (SSQ). The next figure shows a scatter plot of PPF vs. SSQ for a sample of 200,000 beads from a PGM run. Three distinct populations are evident in this plot. The high density group below and to the left of the other groups has the lowest values for PPF and SSQ. Analysis of a large number of PGM runs shows that beads from this cluster are typically clonal, and map well to identifiable reference sequence. The more diffuse group at the far right has the highest PPF and SSQ scores. This group includes super-mixed beads, and other beads that fail to produce high quality sequence. The third group has intermediate PPF and high SSQ, and lies between the other two groups. This group is typically composed of mixed-template beads that fail to map to identifiable reference sequence. In this PGM run, about 74% of beads fall in the clonal cluster, 21% in the mixed cluster, and 6% in the super-mixed cluster.
Version 2.2
Trimming The goal of read trimming is the removal of undesired base calls at the 3' end of a read. Such undesired base calls come from: High-quality base calls on reads that have read all the way through the library insert into the adapter. Low-quality base calls at the 3' end of the read. Three filters are applied to trim reads, targeted at each of the two categories of undesirable 3' bases. Each filter is applied separately and the read length is taken as the shortest of the three. If the resulting trimmed length is shorter than the minimum read length, the read is filtered out entirely. Otherwise the read is written to the SFF and the FASTQ files. As long as trimming does not reduce the read length below the minimum designated allowable length (four bases), the flow values and base calls for the entire read are written into the SFF, regardless of the trimmed length. The trimming information is represented in the designated fields in the SFF file. If the read length is reduced below the minimum length, the read is not included in the SFF and FASTQ files.
Removal of Adapter Sequence
Trailing adapter sequence is trimmed out by searching the read for candidate matches to the known adapter sequence in flow-space. The basecaller works by reversing the effects of CF and IE, producing a phase-corrected ionogram that is stored in the SFF file. If a read extends into the adapter, the 3' end of this phase-corrected ionogram exhibits a pattern that is characteristic of the adapter sequence. Adapter trimming is accomplished by testing each candidate position in the phase-corrected ionogram to see how well it matches the pattern expected for the adapter. The test computes the Euclidean distance between the phase-corrected ionogram at the tested position and the predicted ionogram for the adapter. If this distance falls below a fixed threshold, the corresponding position (translated from flow-space back to read-space) is recorded as the adapter trim location; otherwise, the read is considered not to have a match to the adapter. The threshold used is a Euclidean distance of 4. Note that the software has a default expectation that the adapter sequence starts with ATCACCGACTGCCCATAGAGAG GCTGAGAC. To override this setting, see Configure Global Configs to modify the Args to use for SFF trim field.
Removal of lower-quality 3' Ends with Low Quality Scores
The distribution of quality scores within Ion Torrent reads is such that the highest quality calls tend to occur at the start of the read where phase errors are smallest in magnitude. For reads that run into low-quality bases before reaching the end of the template, it is helpful to trim away the lower-quality calls at the 3' end. The quality trimming is performed using the per-base quality scores (see Technical Note - Quality Score). The approach is to scan along the bases of the read computing a moving average in a fixed-length window, and to set the read trim point to just before the earliest (5'-most) base at which the moving average of the per-base quality scores drops below a fixed threshold. The window size is set to 30 bases, and the threshold below which the trimming occurs is a quality score of nine.
Trimming based on High-Residual Ionogram Values
Phase-corrected ionogram values for 1-mers range from 50-149 and for 2-mers range from 150-249. We define an high-residual 1-mer as one with an intensity value between 50-59 or 140-149 and an high-residual 2-mer as one with an intensity value between 150-59 or 240-249. We require that the fraction of high-residual 1-mers and 2-mers
Version 2.2
does not exceed 3% within any read saved to SFF and FASTQ. Reads that do not satisfy this quality condition over the entire length, have their bases at the 3 end trimmed away, until the remaining bases satisfy it. If no amount of trimming can reduce the fraction of high-residual 1- and 2-mers below 3%, or the trimmed read is shorter than 4 bases, the read is omitted from SFF and FASTQ completely. Contents Introduction Read Filtering Removal of Mixed-template Reads Removal of Lower Quality Reads Trimming Removal of Adapter Sequence Removal of lower-quality 3' Ends
FTTechNoteTOC
Contents
Introduction Read Filtering Removal of Mixed-template Reads Removal of Lower Quality Reads Trimming Removal of Adapter Sequence Removal of lower-quality 3' Ends
Technical Note - Quality Score

Technical Note
The Per-Base Quality Score System
Overview The Ion Torrent per-base quality score system uses a Phred-like method to predict the probability of correct base call. The prediction is based on the quality of the base incorporation signal that was used for generating the base calls. The Personal Genome Machine (PGM) quality score system uses a set of 6 predictors whose values are correlated with the probability of a base miscall. A Phred lookup table is used for converting the values of predictors to error probabilities. The lookup table is generated by training on a representative data set in customer configuration. The lookup table is re-trained for each PGM software release and is shipped as part of the software package. Quality scores are published in the SFF, FASTQ, and SAM files. Quality Score Predictors Torrent software uses the following six predictors that are correlated with empirical base call quality:
Version 2.2
P1
Penalty Residual: A penalty based on the difference between predicted and actual flow values. Computed by the base caller. Local noise: Noise (defined as the maximum absolute difference between the flow value and the nearest integer) in the immediate neighborhood (plus/minus 1 base) of the given base. Signal separation: A separation between 0-mer and 1-mer signals at the start of the read, as measured by the mean and standard deviation of all 0-mers and 1-mers in the read: P3 = -(m1 - m0 - s1 s0)/m1, where m0 and m1 are mean 0-mer and 1-mer signals, and s0 and s1 are mean standard deviations of 0-mer and 1-mer signals. Multiple incorporations: For multiple incorporations of the same nucleotide in one flow, the last base in the incorporation order is assigned a value equivalent to the total number of incorporations. All other bases in the sequence of the multiple incorporations are assigned the value 1. Penalty Miscall: The increase in penalty residual (see P1) if the second best candidate sequence is used by the base caller. Measures certainty about the emitted sequence as compared to the alternative sequences. Environment noise: The average signal noise (defined as the absolute difference between the flow value and the nearest integer) in the neighborhood (plus/minus 5 bases) of the given base.
P2
P3
P4
P5
P6
Ion Torrent does not provide support for customer-generated Phred tables.
Lookup Tables The six quality predictors are calculated for each base. Other predictors (not described here) are computed from the corrected flow values generated by the base caller. The corresponding per-base quality value is located by finding the first line in the lookup table for which all six calculated predictors are less than or equal to the predictor values in the table. This process occurs automatically as part of the standard analysis. The Phred lookup tables are stored in the /opt/ion/config directory on Torrent Server. The Torrent Server supports separate phred tables for each type of chip (Ion 314 chip, Ion 316 chip, or Ion 318 chip), named phr edTable.txt_314, phredTable.txt_316, and phredTable.txt_318 respectively. Results
Version 2.2
The per-base quality along with all other read information is written to the Standard Flowgram Format (SFF) file, using the WriteData method from the sff class. The positive integer quality scores per-base are listed as the last data line for each read in the SFF file, under the Quality scores heading. The quality scores are on a phred -10*l og_10(error rate) scale. In addition to the SFF file, Ion Torrent also saves the results in FASTQ and Sequence Alignment/Map (SAM) files: SFF file The positive integer per-base quality scores are under the 'Quality Scores' heading. The per-base quality scores from 0 to 93 are represented by ASCII characters 33-126. The per-base quality scores are reported in the QUAL field.
FASTQ file
BAM file
References 1. Brockman et al. (2008): "Quality scores and SNP detection in sequencing-by-synthesis systems." Genome Res. 18: 763-770. 2. Ewing B, Hillier L, Wendl MC, Green P. (1998): "Base-calling of automated sequencer traces using phred. I. Accuracy assessment." Genome Res. 8(3):175-185. 3. Ewing B, Green P. (1998): "Base-calling of automated sequencer traces using phred. II. Error probabilities." Genome Res. 8(3):186-194.
Technical Note - TMAP Alignment

Technical Note
TMAP: The Torrent Mapping Alignment Program for Ion Torrent
A new mapping software program, called Torrent Mapping Alignment Program (TMAP), was developed to meet Ion Torrent data mapping challenges. Background Sequence alignment is a critical component of any project utilizing next generation sequencing technologies. There are many options for alignment software, each optimized for specific sequencing platforms and downstream applications. The Ion Torrent data have particular qualities requiring special consideration during the alignment process: Reads generated by the PGM Sequencer are variable length and are expected to increase over time. The principal error mode associated with Ion data relates to miscalling homopolymer lengths and results in insertion or deletion errors during alignment and post-processing. TMAP is part of Torrent Suite on Torrent Server and is used for alignment in the primary analysis pipeline. The TMAP source code is available on the Torrent Dev community Web site under the GPLv2 license, and can be compiled on any standard *nix system for use outside the Torrent Server primary analysis pipeline.
Version 2.2
Implementation TMAP integrates three popular alignment algorithms: BWA-short (Li and Durbin, 2009) BWA-long (Li and Durbin, 2010) SSAHA (Ning et al, 2001) These provide a fast and accurate aligner. The overall alignment strategy identifies a list of Candidate Mapping Locations (CMLs) using all three algorithms. These CMLs are then aligned using the Smith Waterman algorithm (Smith and Waterman, 1981). The resulting alignments are aggregated to find the best mapping(s), and a user-defined parameter determines if all alignments, a subset of alignments, or a random best alignment is reported. TMAP creates an efficient index using the compressed suffix array to quickly and compactly index and query the genome reference. Index creation occurs for both the forward and reverse references, which benefits some alignment algorithms, resulting in a final index size of 4.7GB for a human hg19 reference and taking less than four hours to create. For these evaluations, a Mac Pro running Mac OS X (10.6.6) with a dual 6-core Intel Xeon Processors (2.66GHz) and 32GB of 1066 MHz DDR RAM was used. Up to 24-threads could be used because of the hyper-threading capability of these processors.
TMAP implements a two-stage mapping approach to maintain sensitivity and specificity while significantly reducing runtime. In two-stage mapping, reads that do not align during the first stage are passed to the second stage with a new set of algorithms and/or parameters. The implementation also supports easy integration of other mapping algorithms that can utilize the TMAP index. Advantages TMAP has key advantages over other alignment tools: The re-implemented versions of BWA-short, BWA-long, and SSAHA are significantly optimized to deal with varied length reads and Ion Torrent-specific error profiles. Therefore, TMAP results are expected to perform significantly better when compared against the original algorithms. Because TMAP combines the CMLs for all three methods and identifies the best alignment, the final accuracy and specificity benefit from the advantages of each individual algorithm. TMAP has been optimized for performance and, because the TMAP index is shared between all algorithms, the overall performance (both CPU load and RAM utilization) is comparable to other popular alignment software. The performance when combining multiple algorithms is equal to or better than the total performance of running each algorithm, separately. With a framework that quickly integrates other algorithms, TMAP is a versatile tool for fast and accurate mapping of Ion Torrent data. References Li, H, Durbin, R (2009). Fast and accurate short read alignment with Burrows-Wheeler transform. Bioinformati cs, 25, 14:1754-60. Li, H, Durbin, R (2010). Fast and accurate long-read alignment with Burrows-Wheeler transform. Bioinformati cs, 26, 5:589-95. Ning, Z, Cox, AJ, Mullikin, JC (2001). SSAHA: a fast search method for large DNA databases. Genome Res., 11, 10:1725-9. Smith, T. F. and Waterman, M. S. (1981). Identification of common molecular subsequences. Journal of
Version 2.2
Molecular Biology, 147(1), 195 197.
White Paper - Torrent Variant Detection Algorithms

White Paper
Torrent Variant Detection Algorithms

This white paper describes algorithms developed by Ion Torrent for the detection of SNPs and small indels on the Ion Torrent PGM platform. There are two core algorithms: one for SNP detection (diBayes), and one for detection of insertions and deletions (Variant Hunter).
SNP Detection - DiBayes Algorithm

The SNP Detection (DiBayes) algorithm takes as input a BAM file of aligned reads (for example, a BAM file produced by TMAP, the Torrent aligner), and tiles the genome (or subset of the genome defined by a BED file), evaluating evidence for a variant by looking at a pileup of the reads. Our SNP detection algorithm includes filters by read, by base, and by genome position, and it includes statistical modeling of data that passes the filters to evaluate the probability that a SNP exists at a candidate genome position. If the position passes the filters and there is sufficient evidence, it is called as a SNP and assigned a genotype and a p-value. Two methods based on statistical modeling are implemented for base-space SNP prediction: Bayesian and frequentist. The Bayesian method is currently only implemented for germ-line SNP detection and is most useful when coverage is low to medium (between 10-50). The frequentist approach is used for positions with higher coverage and can be used to detect low frequency variants for example, it is used by default for positions with filtered coverage greater than 60x. Some common filtering occurs before the algorithm. Common filters are as follows: A read-level filter removes reads with low mapping QV. A position-level filter checks each position of the reference to see if there are enough reads mapped to an alternative allele. This is controlled by allele ratio (a user parameter). A base-level filter removes bases of reads with poor base QV. Once a position passes the filter and reads covering the position are collected, we choose one of the two algorithms to determine the genome type and p-value(s) of prediction.
Version 2.2
Bayesian Method When coverage is below a threshold (for example, 60x), we use the Bayesian method. From the set of reads covering the position, we have the number of reads supporting each of the four alleles. We use the most common allele, mca, and second most common allele, sca. The Bayesian algorithm checks three hypotheses: No SNP (ref), homozygous SNP (if mca is not a reference allele), and heterozygous SNP (mca/sca), by calculating P(ref|R), P(Het|R), and P(Hom|R), where R is the set of the reads, and each of the posterior probabilities is calculated using Bayes' theorem: P(H|R)=P(R|H)P(H)/P(R), where P(H) is the prior and P(R) is a constant, and P(R|H) is easily calculated using the base QV to estimate the chance that a base is correct.
Version 2.2
If P=P(ref|R)+P(Hom|R)+P(Het|R), then the likelihood of each prediction is P(H|R)/P, and the hypothesis with the highest likelihood is called. Frequentist Method When coverage is high, we can use the frequentist method, which effectively estimates P(ref|R). If the genome position does not have an alternative allele (i.e., it has no SNP), we do this by estimating the expected number of reads matched to an alternative allele. Using the QV of all reads, we estimate the error rate of the bases covering this position. This error rate times the coverage depth is the number of non-reference reads expected. From the set of reads, we know the actual number of reads matching the alternative allele. A Poisson distribution is used to estimate the likelihood of this happening by chance. Low-frequency variants can be detected by this method. The variants are reported in a VCF file.
Indel Detection - Variant Hunter Algorithm

As with the SNP Detection (DiBayes) algorithm, the Variant Hunter algorithm starts with a BAM file of aligned reads and tiles the genome (or subset of the genome defined by a BED file), evaluating evidence for a variant by looking at a pileup of the reads. The Variant Hunter algorithm calls variants with a score based on a flow-based realignment performed on TMAP aligned reads. The flow-based representation allows for better variant calling by combining simple intensity differences (an error mode of the Ion Torrent system) and very unlikely flow order indels into a single model. In more detail, a sequence deviation in a read that had only a single marginal flow intensity change supporting a variant would be weak evidence, while one that required change in the flow order (unlikely to occur by chance) and/or multiple strong intensity changes would be much stronger evidence of a variant. The software is designed to pick up multi- or non-positional indel events by representing deviations as any sequence of detected differences between the read query and the target reference in a flow-space alignment. From this alignment, it calculates sequence deviations on a read-by-read basis. With the deviations found in each read, it standardizes the representation of the deviations by merging adjacent deviations and representing them in the leftmost position. Differences found for base b, where the intensity is higher than i and the reference is higher than i /100 - 2, are ignored. The software then groups the variants of multiple reads together by position and produces a score based on flow-space alignment characteristics. Alignment characteristics for the score include intensity differences, missing flow bases, and added flow bases, all compared with what would be expected from a reference target sequence. An additional characteristic is how close the flow(s) that specifies the sequence deviation is (or are) to either the start or the end flow for the read. This first score serves as a first stage of filtering on a variant sequence level (even for those found at the same genomic position). Specially, this first level score is done on a per-read and variant basis by the following formula where the actual coefficients are derived from heuristic training.
where coefficients are for different events in the flow alignment, currently set to coefdeletion = 10, coefinsertion = 10,
Version 2.2
coefintensity = 1/10 nf is the number of flows for indels in the flow alignment dflows is the sum of the square distances between read and reference. Then the overall score of the variant is
where |Scores| is the number of scores (or reads) that support the prediction. A second, Bayesian score is calculated on the entire call (see the next section of this document). The overall strategy is to make the score more representative of the likelihood of the indel event. Currently, SNPs and indels are both determined by this approach to allow for choice between nearby SNPs and indels. Heuristic knowledge, as it is amassed, will enable greater refinement of the scoring method. Finally, for heterozygous and rare variant detection, the algorithm determines the rightmost representation, and with fully aligned reads determines the number of reads that fully span the variant, thus not falsely identifying spanning reads because of sequence similarity. The actual call is then made by using a simple frequency range. Positions containing evidence of the position of the variant are passed forward to a statistical algorithm that models the probability of a variant. If the p-value is below a predefined threshold, and if the variant is an insertion or a deletion, the variant is called as an indel and is assigned a genotype and a p-value. For protocols that include enrichment, only variants in or near enriched regions are called.
Bayesian Rescorer Algorithm The Bayesian Rescorer algorithm is used after the Variant Hunter algorithm for postprocessing. For each plausible prediction made using the Variant Hunter algorithm, there is a list of reads supporting the prediction which include flow intensity data. The Bayesian Rescorer algorithm uses these to calculate a likelihood score for the prediction. The rescorer method is currently a combination of the Bayesian and frequentist methods. This method has two steps: The first step is a Bayesian posterior probability calculation at read level. For each read r, we calculate P(r|H0) and P(r|H1), where H0 is always the null hypothesis that there is no variant at this position, and H1 is the predicted variant. The calculation needs to model only sequencing errors because it is conditioned on the true reference for the read being known, but we use both the reference context surrounding the genome position as well as the neighboring flow signals of the read. From Bayesian approach, the value Log(P(r|H0))-log(P(r|H1)) estimates the log of the probability that there is no indel (H0 is true) but we get the actual read r by chance, in other word it is similar to "error rate" at the indel position. Averaging log(P(r|H0))-log(P(r|H1)) over all the read, we get an average "error rate" of all the reads which is crucial in frequentist algorithm. The second step is a Poisson step, using the frequentist approach. With the average error rate from above, we can estimate the expected number of reads that may support hypothesis H1. From the actual number of reads, the algorithm uses a Poisson distribution to estimate the likelihood of the prediction.
Validation
To validate our algorithms we sequence samples with known variants. Some of the samples have fully-known sequence. We sequence samples using the Ion AmpliSeq Cancer Panel and we sequence samples using TargetSeq exome enrichment. We count the number of true positives in the sample and count the numbers of
Version 2.2
false positives and false negatives. Note that if the sample isnt perfectly characterized, some of the variants reported as false positives may be true positives. Exome Validation for SNP Detection We sequenced sample NA12878 to an average depth of 95x using our TargetSeq Exome Enrichment Protocol. We observe 25,334 SNPs, of which 93.66% are in dbSNP. In a high quality sample, we expect to see above 90% of SNPs in dbSNP_132. Table 1: Properties of SNPs Discovered in Sample NA12878, Exome. Settings NumSNPs NumIndbSNP %indbSNP NumHets HetsindbSNP %HetsindbSNP NumHoms HomsindbSNP %HomsindbSNP v1.5.1 21,154 20,342 96.16 11,580 10,958 94.63 9,574 9,384 98.02 v2.0.0 25,334 23,729 93.66 15,943 14,444 90.6 9,391 9,285 98.87
The figure below comes from combining of 3 Ion AmpliSeq runs (ANG189, ANG190 and ANG191) specifically designed to have more indels than a typical experiment. We separate all indel predictions into three classes: Pos: true positive predictions, Neg: false positive predictions, and Off: off-target predictions (when a BED file is not used). We plot the Bayesian score for the three classes of predictions, and it is clear from the plot that there is a separation between the true positive and false positive sets. Off-target predictions all have extremely low Bayesian scores. Please note that there is no upper limit of Bayesian score. The figure truncates around 150. We recommend a cutoff of around 25.
Version 2.2
Figure 1: Distribution of Bayesian score of Positive (Pos, set of true variants), Negative (Neg, set of false prediction), and off-target (Off, set of predicted off target) sets. Here the X-axis represents Bayesian score values, and the Y-axis represents the count of predictions with this Bayesian score.
Workflows
Torrent Variant Caller 2.0 Workflow
First, the user selects a reference, and, if not using the whole genome application, a target region BED file. Next, the user may use the Plan page to define one of six workflows consisting of three possible region sizes (Whole Genome, TargetSeq, and Ion AmpliSeq), each of which may be called at two frequencies (germ-line, somatic). The user may define the workflow at the PGM system, or after the run, but use of the Plan page is encouraged.
Version 2.2
Once sequencing is performed, reads are mapped to the reference with TMAP 2.2, followed by a conversion to a flowspace alignment, and barcode splitting. Reads are trimmed of primers for more accurate variant calling (primers are not from the sample). Summary reports of coverage, variants, chromosomes, and HotSpots are generated. The individual SNP and indel detectors are run, followed by a Bayesian Rescorer which employs flowspace realignment, and filtering by parameters specific to the application. HotSpot variants are annotated and left-aligned, and a report is generated containing the summary, variant, and HotSpots tables, along with links to files that may be of interest to the user (BAM files, VCF files for SNPs and Indels, target region BED files). The variant calling algorithms are separately optimized for three workflows: Ion AmpliSeq Workflow, which includes the Ion AmpliSeq Cancer Panel and Custom Ion AmpliSeq panels. TargetSeq Workflow, with parameters optimized for hybridization-based target enrichment ("TargetSeq"). Whole Genome Workflow, which does not assume enrichment and does not require a target regions file. These methods can be extended to TargetSeq, and validation of exome data. Previously we have extensively sequenced the HuRef exome, and reliably know where true positive variants are located. Ion AmpliSeq Workflow We report only variants within the target region (defined by the target BED file). One specific module for AmpliSeq: Primer Trimming For AmpliSeq, we assign each read to the amplicon it most likely belongs to, and trim away any primer sequence, leaving only the insert sequence. This allows us to elegantly manage overlapping amplicons. Several methods are used to validate indels. We obtained and sequenced 15 samples from the HapMap project, for which sequencing and variant data from the 1000 Genomes Project is publicly available. This provides us with, in principal, true positives for comparison, although they are somewhat limited in number per sample. To increase the absolute number of true positives, a dataset of simulated indels was generated by mapping reads from a HapMap sample to a modified HG19 reference. The strength of this method is that there are many true positives that are definitively known; however, simulation of this type is limited to homozygous calls. Early-access customer feedback is also a valuable source of information for validating indels, especially when attempting to discover the reasons behind false positives. In some cases, customers have Sanger sequenced variants, providing us with the ability to examine known false negatives. TargetSeq Workflow We report only variants within the target region (defined by the target BED file). Whole Genome Workflow We report variants across the entire sequenced region.
Version 2.2
Quickstart
Quickstart
What's Covered Here?
This guide gives you an overview of concepts and methods involved in working with Ion Torrent analysis data. A high-level view of the run data flow, On Dataflow, should give you a point of reference for understanding the different stages of an analysis run. Learn how to start and monitor a run by reading Starting a Run. Learn the basics of how to access and interpret the reports generated by a run, in Viewing Runs and Reports. Finally, in What Next, you can find suggestions for additional reading that gives you more in-depth information about the concepts you have just learned.
Prerequisites
These instructions assume that you have already set up your Torrent Server and can access Torrent Browser. Detailed information about setting up your server be found in the Torrent Server Administration Guide. To manually run an analysis, you must also have a run defined in the run list with experiment data uploaded from the PGM sequencer to Torrent Server. Contents Quickstart On Dataflow Starting a Run Viewing Runs and Reports What Next
On Dataflow
Quickstart
On Dataflow
The Ion Torrent dataflow involves the transfer of raw sequencing data from the PGM sequencer to the Torrent Server for analysis and reporting.
Version 2.2
The following table shows a high-level view of the dataflow, including approximate file sizes for the Ion 314 chip, Ion 316 chip, and Ion 318 chip:
Step Flows Loading Library Reads Classification Raw Image Acquisition Image Processing Signal Processing and Base Calling FASTQ Conversion BAM Conversion
Resulting File Type ---bfmask.bin DAT
Ion 318 Chip 520 81% 4994815 25 MB 232 GB
Ion 316 Chip 520 70% 2167130 14 MB 126 GB
Ion 314 Chip 520 71% 660938 2.9 MB 41 GB
WELLS SFF
16 GB 9.2 GB
7.4 GB 4.0 GB
1.7 GB 1.2 GB
FASTQ BAM
2.4 GB 3.9 GB
1.0 GB 1.8 GB
0.3 GB 0.5 GB
File sizes will vary depending on the number of flows, the number of wells generating signal, and the number of library reads available. Your file sizes may be different.
The PGM sequencer outputs raw sequencing data in the form of DAT files. The raw measurements are the conversion of the raw pH value in a well to a digital representation of the voltage. These raw data are transferred to the Torrent Server for analysis pipeline processing. The analysis pipeline converts the raw signal measurements into incorporation measures and, ultimately, into basecalls for each read. Output files
Version 2.2
at different pipeline stages include WELLS, SFF, FASTQ, and other file formats. See Technical Note - Analysis Pipeline for a detailed description of analysis pipeline processing. The Torrent Browser provides a web interface for viewing processing results, showing metrics and graphs.
From PGM Sequencer to Analysis Files During a PGM sequencer run one acquisition (acq_<flowNumber>.dat) file is generated for each nucleotide flow. The acquisition file contains the raw signals measured in the chip wells. One data acquisition file is created for each nucleotide flow. Acquisition files are temporarily stored on the PGM sequencer hard drive(s).
Dataflow Steps
1. Data are generated on the PGM sequencer as acq_<flowNumber>.dat files with one file per flow. 2. The acq_<flowNumber>.dat files are transferred to the Torrent Server using a standard computer network cable. On the server, data transfer is controlled by the Crawler service. 3. After the data associated with the initial flows of the PGM sequencer are available on the Torrent Server, analysis pipeline processing of the data begins. The first processing step compiles data from the many acq_ <flowNumber>.dat files into one 1.wells file, for a given run. 4. Basecalling is performed on the 1.wells file data, producing SFF and FASTQ file formats. Transferring Data to the Torrent Server Data are automatically transfer from the PGM sequencer to the Torrent Server. You may monitor the transfer progress by viewing the FTP Status column in the Runs tab of the Torrent Browser user interface. The display shows the status of each PGM transfer:
Version 2.2
Ideally, the PGM sequencer and Torrent Server are directly connected using a Category 6 cable, to handle the large data transfer. A direct connection provides the easiest setup and fastest data transfer options, and transfer performance varies greatly because of network infrastructure differences. Please consult the PGM sequencer site preparation guide for more detailed instructions on connecting the PGM sequencer to the Torrent Server.
Contents Quickstart On Dataflow Starting a Run Viewing Runs and Reports What Next
Quickstart TOC
Contents
Quickstart On Dataflow Starting a Run Viewing Runs and Reports What Next
Starting a Run
Quickstart
Starting a Run
Version 2.2
Torrent Browser provides a User Interface for managing runs and viewing reports. Accessing the Run Management Interface 1. When you start Torrent Server, Torrent Browser displays the start page shown in the following figure. Click the CLICK TO BEGIN button to begin using Torrent Browser features:
2. Login to Torrent Browser, entering your username and password, and clicking Log In:
3. The interface is arranged by tabs corresponding to the various UI functions. Click the Runs tab to start an analysis:
Manually Starting an Analysis
Version 2.2
Flagging a run on the PGM sequencer automatically starts analysis. You can also start (or restart) a run, manually, from Torrent Browser Run tab: 1. Find the row with your run name and click the Analyze button next to the run name, which begins with the PGM sequencer name:
2. Click the Advanced button:
See Working with Runs for a description of these arguments.
Automated Analysis and Run Naming Run analysis automatically initiates report generation, using the following run naming convention: auto_<runName
Version 2.2
>_<uniqueID>. The runName is the PGM sequencer name followed by two dashes and the incremental run number for that PGM. For example: B1--120. Monitoring an Analysis Run analysis can take several hours, depending on many factors. You can monitor analysis progress, using the Reports tab. In the row for your run, a sequence of horizontal boxes shows each analysis stage and the status of your run for that stage:
Use your cursor to hover over each box to display the processing stage names, which have the following meaning:
Stage Well Characterization
Description The process of determining the contents of each well on the chip. Estimating the incorporation at each nucleotide flow for each well. The result is a 1.wells file. Determining the sequence in bases, from the 1.wells file data. Creating the FASTQ file from the trimmed SFF file, containing only filtered reads. The QC stage that involves aligning reads produced by the pipeline to a reference genome.
Signal Processing
Basecalling
Creating FASTQ
Aligning Reads
A run stage has the following color code legend:
Version 2.2
Color Green Yellow Gray
Meaning Complete Active processing Pending
Re-running Re-running an analysis does not affect the signal processing part of the analysis, which combines multiple acq_*.dat files into a single 1.wells file.
You may re-run the basecalling, analysis and plugin steps, beginning with the 1.wells file: 1. Click the Torrent Browser Reports tab (next to the Runs tab). 2. Click the "re-do" icon next to the report name (placing the cursor over the icon displays Re-Analyze from a 1.wells file):
The "re-do" option is only available for Completed runs. 3. Follow the prompt to provide a new analysis report name. Notice that this only applies if the base calling algorithm has been updated, or if parameters are changed using the advanced command line arguments. Please, contact Ion Torrent technical support for more information. Running Plugins Your Torrent Suite includes the following pre-installed plugins: Plugin Alignment Coverage Analysis Description Performs a new alignment to the reference you specify. Provides statistics and graphs describing the level of sequence coverage produced for targeted genomic regions.
Version 2.2
Run RecognitION
Considers candidate runs for inclusion in Ion Community leaderboards. Within each chip type (Ion 314 chip, Ion 316 chip, and Ion 318 chip), top runs are ranked according the number of AQ20 bases mapped. Calls SNP and InDel variants across a reference or within a targeted subset of that reference. Has the ability to call variants down to a 5% level of variant frequency. Can optionally show which variants coincide with predefined HotSpot positions on the reference sequence.
Torrent Variant Caller
Plugins can be set to run automatically on every analysis, and also can be run manually on a completed analysis. See the following pages for instructions on running plugins: Plugin Summary Running the Installed Plugins Torrent Variant Caller Plugin
Features offered by the Torrent Variant Caller include the following: Increased accuracy of variant calls due to new algorithms (with optimized variant calling parameters) Using the new algorithms, calls include 10% less false positive SNPs, 50% less false positive insertions, and 60% less false positive deletions than previous versions. Ability to analyze standard and custom panels by uploading a BED file BED files for standard panels have been improved, and you can now upload custom BED files which are provided with our custom-designed products. New and expanded details reports The new details reports include variant and coverage information, and gene symbol and HotSpot ID annotations. Standard UI provides a single, efficient method for multiple analysis types Streamline your workflow by running multiple analysis types from a single plugin.
Viewing Runs and Reports

Quickstart
Version 2.2
Viewing Runs and Reports

The Torrent Server hosts an embedded Web service, called Torrent Browser, which provides a GUI interface to control and view analysis runs, reports and various configuration parameters. You can access Torrent Browser from another computer connected to your network or from any computer connected to the Internet. You do not need to connect a monitor and keyboard directly to Torrent Server. Consult with your Torrent Server Administrator for details on how to access the Torrent Server hosting your data.
When you access Torrent Browser by entering the Torrent Server URL in your browser address bar, the initial page prompts you to start a browser session:
Click the CLICK TO BEGIN button, followed by entering your username and password credentials, to access Torrent Browser features. The main menu has a tabbed arrangement corresponding to UI features and functions:
In this guide, we are introducing only the Runs and Reports tabs. A run is one chip run on the PGM sequencer. A report is the analysis report generated on the Torrent Server. General UI Features First, it is helpful to understand some common Runs and Reports UI features:
Filtering
Version 2.2
The following filtering methods are provided: Filtering by Run/Report Parameter Filtering by Date Filtering by Star Filtering by Run/Report Parameter Runs and Reports have a filter panel at the top of the page for selecting the runs and reports to display: Runs
Reports
Column labels correspond to filter parameters. Use the drop-down menu to select the filter parameter value. Filtering by Date You can also filter runs and reports by date, using the Date start and Date end dialog:
Using the drop-down menus, select the beginning and ending dates, inclusive, of the runs/reports in the interval you want to view. Filtering by Star In the Runs tab, clicking the checkbox in the star column, next to the run name, selects runs to be displayed when S tarred filtering is set:
Version 2.2
In the Runs and Reports tabs, check the Starred checkbox to display only starred runs:
Searching
Enter a run/report text string to search for a run or report by name:
Viewing All Reports for a Run Click the plus (+) symbol to the left of the run name to display all reports generated for the run:
When a run is suitable for re-analysis, multiple reports can accumulate for the run. Click the minus (-) symbol to compact the run display:
Viewing an Analysis Report In the Runs tab, you can choose to view a preview report or a full report. A couple of options are available for viewing the full report. You can view a report while an analysis is in progress, as indicated by a Started status.
Viewing the Preview Report
The preview report provides initial run information, including well loading performance and the skeleton of the full report. Click the run name to expand the panel, which shows the run(s) and run preview information.
Viewing the Full Report
Version 2.2
See Torrent Browser Analysis Report Guide for a detailed description of the full report. The full report may be viewed using the Runs and Reports tabs. To view the report in the active window, click the report name:
To view the report in new window, click the icon to the right of the report name:
Viewing Test Fragment Data In the Reports tab, expanding the plus (+) symbol next to a run name displays test fragment summaries:
If the plus symbol is black, the field can be expanded because test fragment data are available. If a finished run shows a gray plus symbol, no test fragment data of sufficient strength are available to display. More metrics are shown, summarizing the performance of individual test fragments. Comparing Data Across Multiple Runs The Torrent Browser UI does not include a feature for comparing data across multiple runs. But a CSV file can be downloaded to use with external applications, such as Excel, for data analysis. Click the Download CSV button to download analysis data (this link is at the bottom of the Reports tab summary page):
What Next
Quickstart
Version 2.2
What Next?
Use these suggested references to learn more about Torrent Server. Documentation These documents describe the Torrent Server functionality in detail. The documentation is classified according to the intended audience:
Administrator Documentation
Torrent Server Administration Guide Administrator documentation describes how to set up and maintain your Torrent Server. It also includes information on working directly with the database, which involves working with user configuration and run parameters.
User Documentation
Torrent Suite User Documentation In addition to this quickstart guide, user documentation includes the following kinds of documents: Torrent Browser User Interface Guide The user interface guide describes the Torrent Browser interface features, including how to use them and how they relate to the function you want to perform. Torrent Browser Analysis Report Guide The analysis report guide gives a detailed description of the default analysis report. This guide also contains information about the Torrent plugins, including the Torrent Variant Caller: Running the Installed Plugins Torrent Variant Caller Plugin Use Cases This set of documents describes how to perform common operations in your day-to-day work with Ion Torrent. Technical Notes and Whitepapers These documents elaborate on particular topics, providing a significant, in-depth presentation of each topic. Other Torrent Browser Functionality Explore each of the Torrent Browser interface tabs as a hands-on way to learn more about the system: Planning The Planning tab provides a way to enter the PGM run information in advance, reducing hands-on instrument time and allowing an opportunity to print and review your run information. Runs
Version 2.2
You have already had a basic introduction to the functionality available using the Runs tab. Experiment with changing parameters not already discussed in this document to see how they affect a run. Notice the effect changes in run parameters have on items in the Reports tab. Reports You have also already had a basic introduction to the functionality available using the Reports tab. Experime nt with changing parameters not already discussed in this document to see how they affect viewing a report. Services Functionality found in the Services tab involves managing Torrent Server jobs, which are the server software processes that implement the functionality, and archiving data. These are usually considered advanced functions. References The References tab provides a way to define Test Fragments, reference library sequences, and barcode sets. Config The Config tab permits you to modify database items, including user accounts and run behavior. This includes a list of installed plugins and the ability to enable and disable plugins. Functionality exposed through this interface usually involves more advanced use cases. About The About tab lists the Torrent Server software components and the current version of each component installed on your server.
Version 2.2
Torrent Browser Analysis Report Guide

Introduction
This document describes the following Torrent Browser Analysis Report topics: Library Summary Test Fragment Report Ion Sphere Particle Summary Report Information File Links Plugin Summary Running the Installed Plugins You can use your mouse to hover over many table heading and data cells to get helpful tool tips while viewing the report.
Contents Torrent Browser Analysis Report Guide Library Summary Library Summary Overview Library Summary Report Barcode Reports Test Fragment Report Ion Sphere Particle Summary Report Information File Links Plugin Summary Running the Installed Plugins Alignment Plugin combineAlignment Plugin Coverage Analysis Plugin IonReporterUploader Plugin Run RecognitION Plugin sampleIdentifier Plugin Torrent Variant Caller Plugin
Library Summary
Version 2.2

Library Summary
This section provides Library Summary background information and a detailed description of the report. Library Summary Overview Library Summary Report Contents Torrent Browser Analysis Report Guide Library Summary Library Summary Overview Library Summary Report Barcode Reports Test Fragment Report Ion Sphere Particle Summary Report Information File Links Plugin Summary Running the Installed Plugins Alignment Plugin combineAlignment Plugin Coverage Analysis Plugin IonReporterUploader Plugin Run RecognitION Plugin sampleIdentifier Plugin Torrent Variant Caller Plugin
Library Summary Overview

Library Summary Overview This Library Summary section of the Torrent Browser Analysis Report gives performance metrics for reads whose initial bases match the library key. These reads are generated from the input library, not from the positive control Test Fragments.
Performance is measured based on either predicted quality or quality as measured following alignment.
Version 2.2
Using Predicted Quality (Q20) Using Quality Following Alignment (AQ20)

Using Predicted Quality (Q20)
The number of called bases with a predicted quality of Q20 is reported. The predicted quality values are reported on the Phred scale, defined as -10log10 (error probability). Q20, therefore, corresponds to a predicted error rate of one percent. Refer to http://en.wikipedia.org/wiki/Phred_quality_score for a more complete description of Phred values.
Using Quality Following Alignment (AQ20)
Alignment of reads can be a useful process to assess the quality of the sequencing reaction and the quality of the underlying library where an accurate reference is available. Reads are aligned to a reference genome. Any discrepancy in alignment to a reference (whether biological or technical, meaning a real variant or a sequencing error) is listed as a mismatch. Alignment performance metrics are reported depending on how many misaligned bases are permitted. Torrent Suite reports alignment performance at two quality levels: AQ20 Perfect
How Is Aligned Read Length Calculated?
The aligned length of a read at a given accuracy threshold is defined as the greatest position in the read at which the accuracy in the bases up to and including the position meets the accuracy threshold. So for example the AQ20 length is the greatest length at which the error rate is 1% or less. The "perfect" length is simply the longest perfectly aligned segment. For all of these calculations the alignment is constrained to start from position 1 in the read - in other words, no 5' clipping is permitted. The underlying assumption is that the reference to which the read is aligned represents the true sequence that should have been seen. Suitable caution should be taken when interpreting AQ20 values in situations where the sample sequenced has substantial differences relative to the reference used, such as working with alignments to a rough draft genome or with samples that are expected to have high mutation rates relative to the reference used. In these situations the AQ20 lengths might be short even when sequencing quality is excellent. Specifically, the AQ20 length is computed as follows: 1. Every base in the read is classified as being correct or incorrect according to the alignment to the reference. 2. At every position in the read the total error rate is computed up to and including that position. 3. The greatest position at which the error rate is one percent or less is identified and that position defines the AQ20 length. For example, if a 100bp read consists of 80 perfect bases followed by 2 errors followed by 18 more perfect bases, the total error rate at position 80 is zero percent. At position 81 the total error rate is 1.2% (1/81), at position 82 the error rate is 2.4%, continuing up to position 100 where it is two percent (2/100). The greatest length at which the error rate is one percent or less is 80 and the greatest length at which the error rate is two percent or less is 100, so the AQ20 lengths are 80 and 100 bases, respectively.
How Is Alignment Performed?
Within Torrent Browser, the objective is to provide you with a view on alignment that helps determine run and library quality.
Version 2.2
There are many alignment algorithms available within the marketplace and you are encouraged to consult with a bioinformatician for the most appropriate alignment algorithm for your downstream analysis needs. Alignment algorithms are also embedded in many of the commercial software tools available within the Ion Torrent Web store. You are also encouraged to experiment with these tools. Alignment within Torrent Browser is performed using TMAP. TMAP is currently an unpublished alignment algorithm, created by the authors of the BFAST algorithm. Please, contact Ion Torrent Technical Support for more information about TMAP. Technical Note - Analysis Pipeline Technical Note - TMAP Alignment Although TMAP is unpublished and a reference is not currently available, the precursor to TMAP, BFAST, is based on the ideas in the following publications: Homer N, Merriman B, Nelson SF. BFAST: An alignment tool for large scale genome resequencing. PMID: 19907642 PLoS ONE. 2009 4(11): e7767. http://dx.doi.org/10.1371/journal.pone.0007767 Homer N, Merriman B, Nelson SF. Local alignment of two-base encoded DNA sequence. BMC Bioinformatics. 2009 Jun 9;10(1):175. PMID: 19508732 http://dx.doi.org/10.1186/1471-2105-10-175
Which Reads Are Used in the Alignment Process?
The alignment stage involves aligning reads produced by the pipeline to a reference genome and extracting metrics from those alignments. By default, Torrent Suite aligns all reads to the genome, however there may be situations, particularly with large genomes, where the alignment takes longer than the user is willing to wait. So for such circumstances the Torrent Suite also has the capability to define on a per-reference basis the maximum number of reads that should be aligned from a run. For more detail on how to enable and specify this reference-specific limit see the Adding a Reference Sequence section of Working with Reference Sequences. When the number of reads in a run exceeds a genome-specific maximum, a random sample of reads is taken and results are extrapolated to the full run. By sampling a quickly-aligned subset of reads and extrapolating the values to the full run, the software gives you enough information to be able to judge the quality of the sample, library and sequencing run for quality assessment purposes. The outputs of the alignment process is a BAM file. The BAM file includes an alignment of all reads, including the unmapped, with exactly one mapping per read. When a read maps to multiple locations, the mapping with the best mapping score is used. If more than one such mapping exists, a random mapping is used and given a mapping quality of zero. If sampling is in effect for a run and full read alignment is needed, the analysis can be manually re-run using the Runs tab Advanced option and checking the Override alignment sampling ch eckbox.
Version 2.2
Library Summary Report

Library Summary Report Performance, based on either predicted quality or quality as measured following alignment, is provided in the Librar y Summary section of the Detailed Report. This section of the report contains the following information:
Based on Predicted Per-Base Quality Scores - Independent of Alignment
The Based on Predicted Per-Base Quality Scores - Independent of Alignment section gives performance measurements based on predicted quality:
Version 2.2
Parameter Total Number of Bases [Mbp]
Description Number of filtered and trimmed million base pairs reported in the SFF and FASTQ files. Number of bases with predicted quality of Q20 or greater. Total number of filtered and trimmed reads independent of length reported in the SFF and FASTQ files. Average length, in base pairs, of all filtered and trimmed library reads reported in the SFF and FASTQ files. Maximum length, in base pairs, of all filtered and trimmed library reads reported in the file.
Number of Q20 Bases [Mbp]
Total Number of Reads
Mean Length [bp]
Longest Read [bp]
For more information on filtering and trimming, please see Technical Note - Filtering and Trimming.
Read Length Histogram
The Read Length Histogram is a histogram of the trimmed lengths of all a reads present in the SFF file. The following figure illustrates an example graph:
Consensus Key 1-Mer
The Consensus Key 1-Mer graph shows the strength of the signal from the first three one-mer bases of the library key. This graph represents the consensus signal measurement of release of H+ during nucleotide incorporation.
Version 2.2
The y-axis shows signal strength, measured in Counts, which is an arbitrary but consistent unit of measure. The x-axis shows time as nucleotide Flows over the chip. There is a known key at the beginning of every library read. Typically, three bases are shown of the 4-mer key. Note that the graph is displayed in "flow order" rather than "base order." For example, the four-base library key is typically TCAG for nucleotides one through four and is graphed as TCA, representing the nucleotide flow order. Negative flows are not displayed. Q: If the library key is 4 bases, why are only three bases displayed in the key one-mer graph? A: The next base after the last key base is the first library base. This base varies depending on the library fragment. If the last key base is a G and the first library base is a G, both of these are incorporated in the same flow resulting in a signal roughly 2X of a one-mer. Thus, the last base of the library read is not informative for quality purposes because that flow can contain library information in addition to key information. The one-mer key pass graph only contains n-1 flows for an n-mer library key.
Reference Genome Information
The Library Summary includes the Reference Genome Information:
Parameter Genome Name Genome Size Genome Version Index Version
Description Name of reference genome. Number of bases in the reference genome. Version information for the genome used. Version information for the genome index used.
If an alignment error occurred, a message prompts you to view the report log for information about the error.
Based on Full Library Alignment to Provided Reference
This section of the Library Summary report shows performance as measured following alignment.
Version 2.2
Parameter Total Number of Bases [Mbp]
Description Number of million of base pairs that have been aligned to the genome at the specified quality level. Average length, in base pairs, of all library reads that aligned to the genome at Q20 and Perfect (the longest perfectly aligned segment). Maximum length, in base pairs, of all library reads that aligned to the genome at a specific quality level. Average number of times that a base was independently sequenced and aligned to the reference genome. 1X means that every base was sequenced and aligned, on average, once. 2X means that every base was sequenced and aligned, on average, twice. Percentage of the reference genome that is covered at a minimum of 1X by filtered library reads at a specific quality.
Mean Length [bp]
Longest Alignment [bp]
Mean Coverage Depth
Percentage of Library Covered [bp]
Using the TMAP Alignment Algorithm
When the reference genome is a large genome, alignment is performed on a random subset of the reads in the SFF and FASTQ files. You can specify sampling and the number of reads to sample on a per-genome basis. Sampling only occurs if this number is set during reference upload. Otherwise, complete alignment is performed. The values in this table are extrapolated to the full number of reads. Refer to the Working with Runs section of the Torrent Browser User Interface Guide for a description of how to use the Override alignment sampling flag to force all reads to be used for alignment.
Read Alignment Distribution
Summarizing the alignments produces the following report format:
Version 2.2
The number of rows displayed varies depending on the read length.
Each column shows data based on alignment of the sample. Column Read Length [bp] Description The number of bases in each read considered for the row in the table. Number of reads with at least Read Length bases. Number of reads that TMAP could not map. Number of reads mapped but not having 90% accuracy in first 50 bases. Number of reads mapped and with accuracy of greater than 90% in first 50 bases, but with align length less than the Read Length threshold. Number of aligned reads with zero mismatches in the first Read Length bases. Number of aligned reads with one mismatch in the first Read Length bases. Number of aligned reads with two or more mismatches in the first Read Length bases.
Reads Unmapped Excluded
Clipped
Perfect
1 mismatch
2 mismatches
SAM/BAM files for reports with sampled data only include alignment information for the sampled subset.
Version 2.2
Test Fragment Report

Test Fragment Report
The Test Fragment Summary section of the Analysis Report provides information about the performance of each Test Fragment included in the experiment. The report has a heading for each Test Fragment, in the form Test Fragment - <TestFragmentName>. Test Fragments are used during analysis to predict the CF/IE/DR values for each Test Fragment, regionally. Analysis results for a Test Fragment are displayed when there are at least 1000 high-quality Test Fragments, where there is an 85% match against the appropriate template in the Test Fragment list. This includes CF/IE/DR estimates and performance calculations. The number of TFs reported includes lower quality TFs, down to 70% match, to better represent the run quality from all TF's.
Version 2.2
Test Fragment Summary The Test Fragment Summary part of the Test Fragment Report displays the following information:
Test Fragment List
Consensus Key 1-Mer
The Consensus Key 1-Mer graph shows the strength of the signal from the first three one-mer bases of the Test Fragment key. This graph represents the consensus signal measurement of release of H+ during nucleotide incorporation.
The y-axis shows signal strength, measured in Counts, which is an arbitrary but consistent unit of measure. The x-axis shows time as nucleotide Flows over the chip. There is a known key at the beginning of every read. Typically, three bases are shown of the 4-mer key. Note that the graph is displayed in "flow order" rather than "base order." For example, the four-base Test Fragment key is typically ATCG for nucleotides one through four and is graphed as ATC, representing the nucleotide flow order. Negative flows are not displayed. Quality Metrics
The Quality Metrics part of the Test Fragment Report displays the following information:
Version 2.2
Parameter TF Name
Description Test Fragment name, as defined in the Templates tab of Torrent Browser. Test Fragment sequence. Number of filtered & trimmed reads identified for this Test Fragment. Average read length with Q17, or better, for this Test Fragment. Number of reads for this Test Fragment with a minimum of 50 base pairs in length and an error rate of 1 in 50, Phred-like 17, or better. Quality is based on alignment, not predicted quality.
TF Seq Num
Avg Q17 read length
50AQ17
Graphs
The Graphs part of the Test Fragment Report displays the following information:
AQ17 Read Lengths
The AQ17 Read Lengths graph is a histogram of read lengths, in bp units, that have a Phred-like score of 17 or better, or one error in 50 bp. Distributions skewed to the right are ideal, showing longer read lengths (remembering that Test Fragments are a discrete length). It is likely that the sequence can extend all the way through the Test Fragment, if enough flows are
Version 2.2
run, so the histogram only displays a maximum size based on the length of the Test Fragment.
Average Corrected Ionogram
In the Average Corrected Ionogram graph, the x-axis is in flow space and the y-axis shows the signal intensity. A unit of one indicates that there is only one base. A unit of two indicates that there are two bases of that specific nucleotide incorporated during the single flow event.
Version 2.2
Ion Sphere Particle Summary

Ion Sphere Particle (ISP) Summary
The Ion Sphere Particle (ISP) Identification Summary section of the Analysis Report gives summary statistics of Ion Sphere Particle performance:
Version 2.2
This section of the report is available before the Library Summary or Test Fragment Summary sections, providing a quick determination of whether or not the analysis should be permitted to continue. The report displays:
Well Information The data generated in this section are created "early" in the analysis process, before base calling, and are intended to be a coarse, initial assessment of run performance. The well data are subject to more stringent filtering in later stages of the analysis.
Parameter Total Addressable Wells
Description Total number of addressable wells.
Percentage Not calculated
Version 2.2
Wells with ISPs
Number (and percentage of addressable wells) of wells that were determined to be "positive" for the presence of an ISP within the well. "Positive" is determined by measuring the diffusion rate of a flow with a different pH. Wells containing ISPs have a delayed pH change due to the presence of an ISP slowing the detection of the pH change from the solution. Number (and percentage of wells with ISPs) of wells that contained an ISP with a signal of sufficient strength and composition to be associated with the library or Test Fragment key. This value is the sum of the following categories: Test Fragment Library
Wells with ISPs / Total Addressable Wells
Live ISPs
Live ISPs / Wells with ISPs
Test Fragment ISPs
Number (and percentage of Live ISPs) of Live ISPs with a key signal that was identical to the Test Fragment key signal. Number (and percentage of Live ISPs) of Live ISPs that have a key signal identical to the library key signal. These reads are input into the Library filtering process.
Test Fragment ISPs / Live ISPs
Library ISPs
Library ISPs / Live ISPs
Library ISP Details The Library ISP Details table is available after basecalling and read filtering are complete. This table provides information on a collection of read filters, which are applied after basecalling to ensure only high-quality reads are written to the final results. The Polyclonal filter removes ISPs carrying clones from two or more templates. The Primer dimer filter removes reads carrying an insert of fewer than 8 bp. The Low quality filter removes reads that fail to attain a high level of accuracy.
Version 2.2
Parameter Library ISPs/Percent Enrichment
Description Predicted number of Live ISPs that have a key signal identical to the library key signal (the same value as shown in the Well Information table). The Percent Enrichment value reported is the number of loaded ISPs that are Library ISPs, after taking out Test Fragment ISPs.
Percentage Library ISPs / (No. of Loaded ISPs minus TF ISPs)
Filtered: Polyclonal
ISPs carrying clones from two or more templates. Insert length of less than 8 bp. Low or unrecognizable signal. Number (and percentage of Library ISPs) of reads passing all filters, which are recorded in the SFF and FASTQ files. This value may be different from the Total number of reads located in the Library Summary Section due to technicalities associated with read trimming beyond a minimal requirement resulting in Total number of reads being slightly less than Final Library Reads.
Polyclonal ISPs / Library ISPs
Filtered: Primer dimer Filtered: Low quality Final Library Reads
Primer dimer ISPs / Library ISPs Low quality ISPs / Library ISPs Final Library / Library ISPs
Chip Loading Image The Ion Sphere Particle Identification Summary section includes a Chip Loading Image, similar to the image shown in the following figure:
Version 2.2
This is a pseudo-color image of the Ion CHIP showing percent loading across the physical surface. The Loading Density percentage is displayed above the image, with the percentage value that considers only potentially addressable wells. Contents Torrent Browser Analysis Report Guide Library Summary Library Summary Overview Library Summary Report Barcode Reports Test Fragment Report Ion Sphere Particle Summary Report Information File Links Plugin Summary Running the Installed Plugins Alignment Plugin combineAlignment Plugin Coverage Analysis Plugin IonReporterUploader Plugin Run RecognitION Plugin sampleIdentifier Plugin Torrent Variant Caller Plugin
Report Information
Version 2.2

Report Information
Analysis Info
The Analysis Info report displays the following information:
Parameter Run Name
Description Name of the PGM sequencer run entered. This value is typically entered on the PGM sequencer. Date and time the PGM run was started.
Run Date
Version 2.2
Analysis Name
Name of the analysis provided in Torrent Browser when the analysis was initiated. If the analysis was scheduled to auto-start, this is the default analysis name. Date the analysis was performed. Number of PGM cycles analyzed for this report. Note that this number can differ from the total number of cycles run on the PGM sequencer. Number of PGM nucleotide flows analyzed for this report. Note that this number can differ from the total number of flows occurring on the PGM sequencer. Name of the project assigned to the run. This is typically assigned on the PGM sequencer. Name of the sample assigned to the run used to generate this analysis. This is assigned on the PGM sequencer. Name of the library assigned to the run used to generate this analysis. This library name is used to specify the reference genome used for alignment. Name of the PGM sequencer where the run was performed. A series of tests on reference wells (about 10% of the chip in non-addressable areas) is performed to ensure that the chip is functioning at a basic level. The value of this field is either Passed or Failed. Type of chip used on the PGM. Usually, 314, 316, or 318 (for the Ion 314 chip, Ion 316 chip, and Ion 318 chip.) A letter follows the numbers, indicating the chip version. The name of the barcode set assigned to the run. Blank for non-barcode libraries. A space for text notes entered during the PGM sequencer run.
Analysis Date Analysis Cycles
Analysis Flows
Project
Sample
Library
PGM
Chip Check
Chip Type
Barcode Set
Notes
Version 2.2
Flow Order
Flow order selected on PGM sequencer: Samba = TACGTACGTCTGAGCATCGATCGATGTACAGC [ Default] Regular = TACG The "regular" flow order adds bases most rapidly to sequenced molecules but is vulnerable to phase errors. The Samba flow order consists of a 32-base sequence, repeated. This flow order resists phase errors by providing opportunities for out-of-phase molecules to catch up and is designed to sample all dimer (nucleotide pair) sequences, efficiently. Samba is the default flow order because it improve sequencing accuracy for longer reads by resisting phase errors.
Library Key
A short known sequence of bases used to distinguish the library fragment from the Test Fragment. Example: "TCAG"
Software Version
The Software Version report display includes version information for the modules installed on your Torrent Server. The version numbers shown in the example may be different from your current version of the software depending on the age of the analysis. See the About tab in the Torrent Browser for a complete list of modules and version on your server. See the Torrent Suite Release Notes for the package versions in a specific release.
Version 2.2
Parameter Torrent Suite
Description Version of Torrent Suite software used to generate the analysis. Version of the Datacollect package. Version of the LiveView package. Version of the Script package. Version of the Torrent Suite alignment module used for this analysis. Version of the Analysis Pipeline used to generate the analysis. Version of the ion-dbreports package. Version of the NVIDIA Tesla GPU driver. Version of the pre-installed plugins. Version of the TorrentR stats package. Version of the TMAP alignment package.
Datacollect LiveView Script ion-alignment
ion-analysis
ion-dbreports ion-gpu ion-plugins ion-torrentR tmap
Version 2.2
File Links
File Links
These links permit you to directly download the data and report files. Some files are compressed, using the .zip for mat, to provide data integrity and to reduce download time. Right-click the wanted file type and follow the Save link as dialog to save the file to your local computer. Most output files can be loaded into third-party viewers (such as IGV) for visualization. The barcode links only appear for runs on barcoded data.
Version 2.2
File Type Library Sequence (SFF)
Description Compressed (.zip) Standard Flowgram Format (SFF) -formatted file that contains "flow space" data. The bases called in a run are stored in two formats: SFF and FASTQ. Both files contain the nucleotide calls and associated quality values, the SFF files, additionally, contain signal values in flow space and a mapping between sequence and flow spaces. The data are organized on a per flow basis, and contain information about nucleotide flows that both did and did not result in base incorporation. (See Technical Note - Filtering and Trimming.) Compressed (.zip) FASTQ-formatted file containing data organized in a per-base basis, including quality scores. The reads contained in the file are unaligned reads. (See Technical Note - Filtering and Trimming.) Binary Sequence Alignment/Map (BAM), is a compressed, binary form of the SAM file. BAM files can be indexed, using the BAM Index file, for quick access to sequence alignment data. See http://samtools.sourc eforge.net for more a more detailed description of the SAM/BAM file format. Many tools are available for working with SAM files. See Library Summary Overview for a description of the alignment data included in the BAM file. The reads in the file are sorted by reference location.
Library Sequence (FASTQ)
Library Alignments (BAM)
Version 2.2
Library Alignments (BAM Index)
Binary Sequence Alignment/Map Index (BAI) -formatted file. A BAM index file speeds up the access time for a coordinate-sorted BAM file, enabling software to more quickly access random parts of the genomic information in a BAM file. Each BAM index file is associated with a BAM file so be careful not to confuse them. They share the same file name but the BAM index file has a .bai extension. To access information in a BAM file, a BAM index file is not required but does improve time-to-access. A summary file of alignment metrics for barcodes. The metrics include the quality and read lengths at which each barcode aligns to reference. This link appears on barcode runs only. The same format and purpose as the Library Alignments (BAM) and Library Alignments (BAM Index) files, with content for specific barcodes. The BAM and BAM Index files for each barcode are zipped together. This link appears on barcode runs only. The same format and purpose as the Library Sequence (FASTQ) file, with content for specific barcodes. The FASTQ files for each barcode are zipped together. This link appears on barcode runs only. The same format as the Library Sequence (SSF) file, with content for specific barcodes. The SFF files for each barcode are zipped together. This link appears on barcode runs only. Test Fragment data are provided as a compressed, binary Standard Flowgram Format (SFF) file. See http:/ /www.ncbi.nlm.nih.gov/Traces/trace.cgi?cmd=show&f=f ormats&m=doc&s=format#sff for a detailed description of the file format. Complete detailed Analysis Report in PDF format. ZIP archive containing the PDF and HTML version of the run report as well as useful logs in case troubleshooting is required.
Barcode Alignments Summary
Barcode-specific Library Alignments (BAM and BAM Index)
Barcode-specific Library Sequence (FASTQ)
Barcode-specific Library Sequence (SFF)
Test Fragments (SFF)
PDF of this Report Customer Support Archive
Version 2.2
Plugin Summary
Plugin Summary
The Plugin Summary lists the plugins associated with the analysis, and provides an interface for running and monitoring your plugin(s).
Plugins can have one of the following behaviors: Run without user input.
Version 2.2
Run with a user interface for getting plugin parameters. Plugins without a user interface display a confirmation message that the plugin has been submitted to run. Plugins requiring input parameters display a user interface dialog and are launched after you click Submit. Running Plugins The following plugins are pre-installed on Torrent Server. See Running the Installed Plugins for a description of these plugins. Alignment Coverage Analysis Run RecognitION Torrent Variant Caller
To manually run a post-analysis plugin on the report data: 1. Click Select Plugins To Run. 2. The Plugin List pops up and displays a list of available plugins:
3. Click the plugin you want to run. If the plugin does not require user input, it begins execution. This displays the user interface for your plugin. In this example, the Alignment plugin is selected, displaying the Alignment plugin dialog (See Running the Installed Plugins for more information about running pre-installed plugins.):
Version 2.2
4. Select the desired plugin options and click Submit. This runs your plugin, listing the run status in the Plugin Summary panel. 5. For plugins that take a long time to run, click Refresh Plugin Status to update the plugin display status. When your plugin completes, the status is updated to Started. The Plugin Summary List After a plugin runs, it is listed in the Plugin Summary panel:
Some plugins, such as Alignment, display a preview results window in the Plugin Summary list. Plugin Reports Plugin results, results summaries, links to output files, and other information are available in the plugin report pages. See Running the Installed Plugins for a description of report pages for the installed plugins. Click the plugin html link in the Plugin Summary section to open that plugin's report page. Plugin Log Files 1. To view the plugin log file, hover your mouse pointer over the log icon to the right of your plugin to display the plugin log file:
2. Click the icon to display the log:
Version 2.2
3. Click Close to exit the display. Contents Torrent Browser Analysis Report Guide Library Summary Library Summary Overview Library Summary Report Barcode Reports Test Fragment Report Ion Sphere Particle Summary Report Information File Links Plugin Summary Running the Installed Plugins Alignment Plugin combineAlignment Plugin Coverage Analysis Plugin IonReporterUploader Plugin Run RecognitION Plugin sampleIdentifier Plugin Torrent Variant Caller Plugin
Running the Installed Plugins
Version 2.2

Running the Installed Plugins
Available Plugins Manually Run a Plugin Automatically Run Plugins Available Plugins
Plugin Alignment Plugin combineAlignment Plugin
Description Performs a new alignment to the reference you specify. Combines reads aligned to the specified reference from multiple run reports. Intended for use when multiple runs analyze the same tissue sample, for example when a tissue sample is run on more than one chip. Does not support barcoded runs. Provides statistics and graphs describing the level of sequence coverage produced for targeted genomic regions. Transfers run results files to your Ion Reporter server (available under a separate license). Considers candidate runs for inclusion in Ion Community leaderboards. Within each chip type (Ion 314 chip, Ion 316 chip, and Ion 318 chip), top runs are ranked according the number of AQ20 bases mapped. Uses sample fingerprinting to identify any cross-contamination between samples or between barcodes in a run. Calls SNP and InDel variants across a reference or within a targeted subset of that reference. It has the ability to call variants down to a 5% level of variant frequency. It can also show which variants coincide with predefined HotSpot positions on the reference sequence.
Coverage Analysis Plugin
IonReporterUploader Run RecognitION Plugin
sampleIdentifier Plugin
Torrent Variant Caller PluginTorrent Variant Caller Plugin
Manually Run a Plugin The Plugin List is the mechanism to manually run plugins. The Plugin List is accessed in the run report of completed analysis runs. Only enabled plugins are listed. Follow these steps to manually run a plugin: 1. Click the Reports tab, then click the link for your completed analysis run. 2.
Version 2.2
2. In the run report, scroll down to Plugin Summary section.
The Plugin Summary also lists any plugins that executed on your run (not shown in this example). 3. Click Select Plugins To Run to see the list of available plugins. See Available Plugins for the pre-installed plugins. Your server may have additional plugins installed.
4. Click the desired plugin name to run a plugin. If the plugin does not require user input, it starts immediately, without a confirmation screen. 5. Click Close to close the Plugin List without running a plugin. Automatically Run Plugins Your Torrent Server administrator can set plugins to be executed automatically on every run; however, the preinstalled plugins are not good candidates to be run automatically. For example, the Torrent Variant Caller plugin fails on runs that do not use both the hg19 reference and the Ion AmpliSeq Cancer Panel or custom product data.
Version 2.2
The Alignment and RunRecognitION plugins cannot be be executed automatically because they require manual user input before every run. For Ion Reporter users, enabling Auto-Run with the IonReporterUploader plugin could cause you extra expense, by transferring unnecessary results to the Ion Reporter server. (The Planning tab allows you to turn off the plugin for a planned run.)
Alignment Plugin
Alignment Plugin The Alignment plugin performs a new alignment to the user-selected reference file. Output files include SAM/BAM and FASTQ files of sampled reads. This plugin cannot be auto-run because it requires user input before execution. The following instructions describe steps to run the plugin and to review the plugin output report. 1.
Version 2.2
1. To run this plugin, in the report for your run, scroll down to the Plugin Summary section, and click Select Plugins to Run. 2. In the the Plugin List select Alignment. The plugin displays the plugin user interface:
3. From the Libraries drop-down list, select the desired library. 4. Select the desired Sampling option. 5. Click the checkbox for each of the desired plugin output formats: .sam, .bam, and .fastq. 6. Click Submit to run the plugin with the specified parameters. The interface provides feedback that the plugin has started:
On completion, the Alignment displays a minimal output preview in the plugin list of the Plugin Summary pa nel:
Click the Alignment.html link to display the full plugin output.
Version 2.2
combineAlignments Plugin
combineAlignments Plugin The combineAlignment plugin combines reports from multiple runs that analyze the same tissue sample. The plugin is used, for example, when you run a tissue sample on more than one chip. The plugin combines the reports from those runs into one report. You run the plugin from any one of the related reports. All runs must use the same reference. Analyses using barcoded libraries are not supported in this release.
Run the Plugin
You run the plugin from the Report tab run report of one of the runs to be combined:
Version 2.2
That report appears in the selected reports section. Click search to find related reports:
Related reports appear in the search section. Your original report has the Add button grayed out:
Click Add for each run to be combined. Additional search results can be seen by clicking the More button on the lower right. When you have all runs selected and appearing in the selected report section (the lower table), select a name for the combined alignments report, and click Submit.
Plugin Notes
the plugin includes these notes in the submission page: This plugin combines reads aligned to the specified reference from multiple run reports. The resulting combined
Version 2.2
alignment files may be downloaded from the plugin report page or used as input for other Torrent Browser plugins that support the use of these files, such as the Torrent Variant Caller. Use the filters in the Report Locater to find reports then click the Add button to add them to the Selected Reports table. Initially the list of selected reports will contain the current report name. This report, or any selected report, may be removed from the list by clicking the Remove button. The alignment file from selected reports is combined by this plugin after clicking the Submit button. You may also modify the default name for the combined alignment files generated. Note that the Report Locator will only include reports that are completed and does not support barcoded runs. Also, the Report Locator only lists reports for runs associated with the selected reference. Contents Torrent Browser Analysis Report Guide Library Summary Library Summary Overview Library Summary Report Barcode Reports Test Fragment Report Ion Sphere Particle Summary Report Information File Links Plugin Summary Running the Installed Plugins Alignment Plugin combineAlignment Plugin Coverage Analysis Plugin IonReporterUploader Plugin Run RecognitION Plugin sampleIdentifier Plugin Torrent Variant Caller Plugin
Coverage Analysis Plugin

Coverage Analysis Plugin The Coverage Analysis plugin provides statistics and graphs describing the level of sequence coverage produced for targeted genomic regions.
Version 2.2
You can run the Coverage Analysis plugin automatically or manually. To run the Coverage Analysis plugin automatically, you must first set up a plan and apply it. Refer to the Planning Tab section of the Torrent Browser User Interface Guide for information about how to set up and apply a plan. To run the Coverage Analysis plugin manually, perform the following steps: 1. In the Torrent Browser, select a run report by clicking a run link, then clicking a report from the dropdown area. The run report opens.
2. On the run report page, scroll about halfway down the screen to the Plugin Summary area.
3. Click Select Plugins to Run. The Plugin List appears.
Version 2.2
4. Select coverageAnalysis. The Coverage Analysis Plugin interface appears.
5. Select a library type. 6. If you have one and would like to use it, select a targeted regions file. 7. If you would like to pad the target by a number of bases, enter the desired number. If you do not enter a number, the default of 0 will be used. 8. If you would like the option to examine unique starts, select the checkbox.
Version 2.2
9. When you are satisfied with your selections, click Submit. The analysis runs and a group of output reports is created. The following sections of this document describe the output reports generated by the Coverage Analysis plugin. Coverage Analysis Plugin Output The Coverage Analysis Plugin output is a report called the Coverage Analysis Report. This report includes statistics about all reads, and the following information in graphical form: Target Coverage Binned Target Coverage Target Coverage by Chromosome Individual Target Coverage On/Off Target Read Alignment Normalized Target Coverage In addition, from the bottom of the Coverage Analysis Report, you can download a BAM file or its related BAM index file.
All Reads
The All Reads section of the Coverage Analysis Report provides statistics about all reads. The items in this report are described in the table below.
Version 2.2
Statistic Number of mapped reads Number of reads on target
Description Total number of reads mapped to the reference. Total number of reads mapped to any targeted region of the reference. A read is considered to be on target if at least one aligned base overlaps a target region. A read that overlaps a targeted region but where only flanking sequence is aligned, for example, due to poor matching of 5' bases of the read, is not counted. The percentage of reads mapped to any targeted region relative to all reads mapped to the reference. The total number of bases covered by reads aligned to the reference. The total number of target bases covered by any number of aligned reads.
Percent of reads on target
Total aligned base reads
Total base reads on target
Version 2.2
Percent base reads on target
The percent of all bases covered by reads aligned to the reference that covered bases in target regions. The total number of bases in all target regions of the reference. The total number of target bases that had at least one read aligned over the proximal sequence. Only the aligned parts of each read are considered. For example, unaligned (soft-cut) bases at the 5' ends of mapped reads are not considered. Covered target reference bases may include sample DNA read base mismatches, but does not include read base deletions in the read, nor insertions between reference bases. The average number of reads of all targeted reference bases... The percentage of all target bases covered by at least 0.2x the average base coverage depth. The maximum number of times any single target base was read. The average number of reads of all targeted reference bases that were read at least once. The standard deviation (root variance) of the read depts of all targeted reference bases that were read at least once. The percentage of target bases covered by at least one read. The percentage of target bases covered by at least ten reads. The percentage of target bases covered by at least 20 reads. The percentage of target bases covered by at least 50 reads. The percentage of target bases covered by at least 100 reads.
Bases in targeted reference
Bases covered (at least 1x)
Average base coverage depth
Uniformity of coverage
Maximum base read depth
Average base read depth
Std.Dev base read depth
Target coverage at 1x
Target Coverage Graph
In the Target Coverage graph, target base coverage is plotted against read depth, where read depth is the number of times a particular base is read and coverage is the number of counts of bases read at that read depth. This plot
Version 2.2
includes the number of target bases that were not read (to 0x coverage). The right-hand y-axis shows the cumulative coverage as a percentage of the total number of targe base reads, corresponding to the blue line. This may be used to gauge the target number of bases read to a particular depth.
Binned Target Coverage Graph
In the Binned Target Coverage graph, as in the Target Coverage graph above, target base coverage is plotted against read depth, where read depth is the number of times a particular base is read and coverage is the number of counts of bases read at that read depth. However, in the Binned Target Coverage graph, the read depths are binned to better represent coverage where the maximum read depth is high. The binning size is chosen accordingly. For example, an x-axis value of 20x indicates the sum of coverage for reads from 1 to 20, or 11 to 20 if the preceding bar was labeled 10x, etc. This plot does not include the coverage for non-covered target bases (0x coverage).
Version 2.2
Target Coverage by Chromosome Graph
In the Target Coverage by Chromosome graph, the number of reads that align to each chromosome of the reference are plotted as a bar chart. The number of reads that have aligned sequence overlapping any part of the target region are represented by the gray portion of the plot. Those aligned off-target are represented by the white region.
Version 2.2
Individual Target Coverage Graph
In the Individual Target Coverage graph, the number of aligned base reads are counted for individual target regions of the reference and normalized by dividing by the length of the target. These valuea are plotted on a bar chart for each chromosome of the reference to a common scale, set by the highest normalized count. Red areas of the bars show the fraction of the target bases uncovered by any read. For example, 20:80 red:gray indicates only 80% of the original target was read. Note that with large numbers of targets, details for individual target regions (bars) may not be visible. Download the data file using the link provided to examine the individual target coverage in full detail.
Version 2.2
On/Off Target Read Alignment Graph
In the On/Off Target Read Alignment graph aligned read starts are counted for each 100 bases of the reference and plotted as a bar if the count as at least five. Hence zero- or low-coverage regions, including isolated read mappings, are not represented in these plots. Plot color is alternated to show continuous regions of coverage. Red and blue represent the reads starting in a 100 base region overlapping a target region. Black and gray represent reads starting outside of a target region. Peaks are contiguous and aligned to 100 base counts along the reference but no distance between any two peaks is depicted. Note that with large numbers of targets, peak shape and resolution between on- and off-target peaks may not be discernible. Download the data file using the link provided to examine the coverage in full detail.
Version 2.2
Normalized Target Coverage Graph
In the Normalized Target Coverage graphl, the fraction of total target bases covered is plotted against normalized coverage, where normalized coverage is the number of times a particular base is read (read depth) divided by the read depth of all target bases. The y-axis is the cumulative read depth divided by the total number of target base reads.
Version 2.2
Version 2.2
Run RecognitION Plugin

Run RecognitION Plugin The Run RecognitION plugin allows you to submit your best runs to the Ion Community leaderboards. The leaderboards are available on the Ion Community, and are organized in leagues according to chip type:
Version 2.2
Your Torrent Suite software must be at least version 1.5.1 to use this plugin.
Running the RunRecognitION Plugin Follow these steps to run to the RunRecognitION plugin: 1. Go to the report page for your run. Scroll down to the Plugin Summary section, and click Select Plugins To Run. 2. In the Plugin List, click RunRecognitION:
Version 2.2
The leaderboard for your chip type is shown, with a message indicating where your ranks among the leaderboard runs.
How Do I Submit My Run to the RunRecognitION leaderboard?
Follow these steps to submit a candidate run to the leaderboard: 1. Go to the report page for your run. Run the RunRecognitION plugin, as described in #Running the RunRecognitION Plugin. The Run RecognitION leaderboard is displayed for your chip type.
Version 2.2
2. Scroll to the Ion Community Login section below the leaderboard. If you do not have an Ion Community account, click create an account to register. Enter your community user name and password. 3. Optionally enter the any of this information about your run: Your site name The reference genome used in this run The application type for this run 1. Below the information fields, click Terms and Conditions and carefully read that information. Click the checkbox "I agree to the Terms and Conditions". 2. Click Submit at the bottom of the page. If your run qualifies, it is added to the leaderboard:
Version 2.2
What Information About Me Does RunRecognitION Make Public?
If your run is published to the leaderboard, your Ion Community user name and avatar are visible to other members of the community. Contents Torrent Browser Analysis Report Guide Library Summary Library Summary Overview Library Summary Report Barcode Reports Test Fragment Report Ion Sphere Particle Summary Report Information File Links Plugin Summary Running the Installed Plugins Alignment Plugin combineAlignment Plugin Coverage Analysis Plugin IonReporterUploader Plugin Run RecognitION Plugin sampleIdentifier Plugin Torrent Variant Caller Plugin
Version 2.2
The IonReporterUploader Plugin

IonReporterUploader Plugin The IonReporterUploader plugin transfer results files directly from a Torrent Server analysis to your Ion Reporter server (available under a separate license). You can run the plugin on a completed analysis, in the Report tab, or create a run plan that launches the plugin automatically when the PGM analysis completes. When you run the plugin, you have these options: Transfer and analyze Select the Ion Reporter workflow to analyze the newly-transferred files. The Ion Reporter analysis begins automatically when file transfer completes. In this release, only single-sample workflows are supported. Transfer but do not analyze automatically Select No Workflow to have the files transferred to the server and imported into Ion Reporter application. An analysis is not automatically launched, but the transferred files appear in the Search Sample page, and an analysis can be launched from the Ion Reporter UI. Do not transfer Select Do not upload or launch Ion Reporter Analysis to turn off the plugin in a run plan. This feature prevents the unnecessary transfer of results when Auto-Run is set for the plugin. This option does not appear in the Select Plugin to Run submission page. There is a one-time configuration of the plugin in both the Ion Reporter UI and the Torrent Browser, before the plugin can be used.
Role Requirements
An Ion Reporter administrator role must generate an authentication token in the Ion Reporter UI. This token is used to configure the IonReporterUploader plugin in the Torrent Browser (as described below). Either a regular user or an ionadmin can configure the plugin in the Torrent Browser.
Transfer Limitations
The following limitations apply to the Uploader plugin: The Uploader plugin transfers results files for a completed run that executed on the Torrent Server where the plugin is configured. You cannot copy results file from a different Torrent Server and have the plugin transfer those files. Only files appearing in the File Links section of a run report are transferred. You cannot add supplemental files to the results files of a run, in order to have the plugin transfer those files.
Run the IonReporterUploader Plugin from a Run Report
Follow these instructions to run the plugin on a completed PGM analysis. 1. Log into your Torrent Browser, and go to the Reports tab. 2. Open the run report for the PGM sequencing data to be transferred, and scroll down to the Plugin Summary section.
Version 2.2
3. Check that the VariantCaller plugin is not in progress. 4. Click the Select Plugins to Run button. 5. Double-click the IonReporterUploader entry. The Ion Reporter Uploader submission screen opens:
Do not change the pre-configured fields. 6. If you select an Ion Reporter workflow, the uploader plugin launches an analysis on the transferred results files. In the Launch Ion Reporter analysis section, select one of the following in the Ion Reporter workflow menu: No Workflow The plugin transfers your files to the Ion Reporter system, but does not automatically launch an analysis. A specific workflow name After the plugin transfers your files to the Ion Reporter system, it launches an analysis using the workflow you select with this menu. Note: The list of workflow names is read from your Ion Reporter department. Please allow a moment for the menu to be populated. 7. Click Submit. Because of the large size of sequencing results files, file transfer typically takes some time. The Ion Reporter analysis cannot be started until file transfer is complete. 8. Check the progress of the analysis on the import role Home tab or in the Analysis tab > Analysis Tracker p age.
Add the IonReporterUploader Plugin to a Run Plan
To set a run plan to automatically transfer files (after the completion of the PGM analysis), use the Ion Reporter menu on the Edit Plan page.
Version 2.2
The plugin supports the following options: Transfer and analyze Select the Ion Reporter workflow to analyze the newly-transferred files. The Ion Reporter analysis begins automatically when file transfer completes. In this release, only single-sample workflows are supported. Transfer but do not analyze automatically Select No Workflow to have the files transferred to the server and imported into Ion Reporter application. An analysis is not automatically launched, but the transferred files appear in the Search Sample page, and an analysis can be launched from the Ion Reporter UI. Do not transfer Select Do not upload or launch Ion Reporter Analysis to turn off the plugin in a run plan. This feature prevents the unnecessary transfer of results when Auto-Run is set for the plugin. This option does not appear in the Select Plugin to Run submission page. The menu by default is set to not transfer:
The plugin supports automatic launch of an Ion Reporter analysis on the transferred files. In this release, single-sample analyses are supported (related samples are not supported). For automatic launch, select a workflow:
To transfer files without analysis launch, select No Workflow in the menu.

Plugin Not Configured or Not Enabled
If the Uploader plugin is not enabled for Auto-Run, the Ion Reporter menu does not appear in in the Edit Plan page. If the Uploader plugin is not configured with the Ion Reporter authentication token, the plan page provides a link to the Config tab:
Version 2.2
Configure Ion Reporter for the Plugin
An Ion Reporter administrator generates an authentication token. If a second authentication token is generated, the previous token becomes invalid immediately.
Configuration Limitation
You cannot configure the plugin to support more than one Ion Reporter organization on the Ion Reporter server.
Configure the IonReporterUploader Plugin
You configure the plugin in the Torrent Browser with the authentication token that an Ion Reporter administrator generated in the Ion Reporter UI. The plugin is configured once, and then is available for all users on your Torrent Server. Follow these steps to configure your IonReporterUploader plugin: 1. Log into the Torrent Browser as either an ionadmin or a regular user. 2. Go to the Config tab and scroll down to the plugin section. Find the entry for IonReporterUploader. 3. Click the Manage link for IonReporterUploader, and select Configure IonReporterUploader:
4. The Ion Reporter Uploader Configuration page opens:
Version 2.2
If the configuration page already contains an different authentication token, that configuration might now be invalid.
5. Enter the authentication token from the Ion Reporter administrator in the authentication token field, and click Save. 6. If the IonReporterUploader is not enabled, click that checkbox to enable it:
7. Torrent Browser plugins optionally support an Auto-Run feature. If Auto-Run is enabled for IonReporterUploader, the plugin attempts to transfer all results files from all completed runs on this Torrent Server. In most situations, this behavior is not appropriate and could be expensive. The run plan page supports a plugin option to not transfer the results of run being planned. 8. If you enable Auto-Run for the plugin, select the plugin option Do not upload or launch in the Torrent Browser Planning tab to avoid transferring files you do not want. 9. This configuration of the plugin makes all files transferred by the plugin available to the Ion Reporter organization that generated the authentication key.
Version 2.2
The sampleIdentifier Plugin

The sampleIdentifier Plugin The sample identifier plugin provides a way to examine a barcoded run to determine if there is any look at cross-contamination between barcodes. Example plugin output is shown below:
Version 2.2
The sampleIdentifier plugin does not take user input. The plugin executes immediately when you click it in the Select Plugins to Run list.
Version 2.2
Torrent Variant Caller Plugin

----
Torrent Variant Caller Plugin
This page contains preliminary information. Click here for the latest Torrent Variant Caller Plugin,
The Torrent Variant Caller plugin is a single plugin that combines the functionality of three previously available plugins (AmpliSeqCancerVariantCaller, Germ-lineVariantCaller, and targetSeq) into a single, unified interface with the following features and benefits:
Version 2.2
Increased accuracy of variant calls due to new algorithms (parameters optimized on a much larger training data set). Using the new algorithms, calls include 10% less false positive SNPs, 50% less false positive insertions, and 60% less false positive deletions than previous versions. Ability to analyze standard and custom panels by uploading a BED file. BED files for standard panels have been improved, and you can now upload custom BED files which are provided with our custom-designed products. New and expanded details reports. The new details reports include variant and coverage information, and gene symbol and HotSpot ID annotations. Standard UI provides a single, efficient method for multiple analysis types. Streamline your workflow by running multiple analysis types from a single plugin.
Upgrading to Torrent Suite Software 2.0 and the Torrent Variant Caller is recommended due to substantial improvements over previous versions. For example, Torrent Variant Caller in Torrent Suite 2.0 is over twice as fast as AmpliSeq Variant Caller in Torrent Suite 1.5.1. However, if you continue to use Torrent Suite Software 1.5, you will continue to have access to the individual AmpliSeqCancerVariantCaller, Germ-lineVariantCaller, and targetSeq plugins.
Using the new, unified Torrent Variant Caller plugin, you can perform six possible analysis types by selecting one of three library types (Whole Genome, Ion AmpliSeq, Ion TargetSeq) and one of two variant frequencies (Germ-line or Somatic). In addition, if you have Target Region and/or HotSpot Region files in BED format, you can upload them and select them from the dropdown menu, and they will be applied to your analysis. If you do not select a Target Regions file or a HotSpot Regions file, the Torrent Variant Caller plugin assumes there are no targeted regions when it runs the analysis.
Running the Torrent Variant Caller Plugin
There are two ways to run the Torrent Variant Caller plugin: automatically, by preconfiguring the plugin to run as soon as primary analysis has completed, or manually, allowing you to run the plugin at any time.
Running Torrent Variant Caller Automatically
To run the Torrent Variant Caller plugin automatically, perform the following steps:
1. Upload your BED files.
For information about working with BED files, see Working with BED Files.
2. Plan your run and select Run Type, Variant Frequency, and Reference.
Version 2.2
For information about planning a run, see Planning Tab. The Torrent Variant Caller will not run if you select Generic Sequencing as the run type.
3. If the run type is AmpliSeq or TargetSeq, select the proper target and HotSpot region files.
4. Run the plan on the PGM.
For information about running a plan, see Working with Runs.
The Torrent Variant Caller runs automatically after primary analysis has completed. The Torrent Variant Caller takes a significant amount of time to complete. Setting it up to run automatically saves time compared to running it manually.
Running Torrent Variant Caller Manually
To run the Torrent Variant Caller plugin manually, perform the following steps:
1. In Torrent Suite Software 2.0, select a run report by clicking a run link, then clicking a report link that drops down. The run report opens.
2. On the run report page, scroll about halfway down the screen to the Plugin Summary area.
Version 2.2
3. Click Select Plugins to Run.
A new window with a list of plugins appears.
4. Select variantCaller.
The Torrent Variant Caller Plugin interface appears.
Version 2.2
The first item, Reference Genome, displays the reference used for mapping, which will also be used for variant calling in the current analysis. This reference comes from selections made on the PGM or on the plan page during the run configuration. The reference can be changed on the PGM or on the plan page, and a new analysis can be run using the correct reference. The reference cannot be changed from within the Torrent Variant Caller Plugin interface.
5. Select a Library Type for your analysis from the dropdown menu.
Options include Whole Genome, Ion AmpliSeq, and Ion TargetSeq.
6. Select a Variant Frequency for your analysis from the dropdown menu.
Options include Germ-line or Somatic.
7. If they are needed for your analysis, select Targeted Regions or HotSpot Regions BED files by doing one of the following: If the Target Regions file and/or HotSpot regions files you want to use already appear in the dropdown menu, select them. If you want to use a Target Regions or HotSpot Regions file that is not already included as an option in the dropdown menu, use the BED file uploader on a reference page to upload the BED file. Once you have uploaded a new BED file, it appears in the dropdown menu, and you can select it for use in your analysis. BED files provided by Life Technologies can be downloaded from the Ion Community. Use the BED file uploader to include these files as options in the dropdown menu. For more information about BED files, see Working with BED Files. 8. When you are satisfied with your selections, click Submit.
Version 2.2
The analysis runs and a group of output reports is created. The following sections of this document describe the output reports generated by the Torrent Variant Caller plugin. Torrent Variant Caller Plugin Output The Torrent Variant Caller plugin output includes the following details reports: Barcode Coverage and Variants Report Variant Caller Report Variant Calls Summary Variant Calls Allele Coverage for All Bases in HotSpot Regions In addition, there is a File Links section, which is a list of output files generated by the Torrent Variant Caller Plugin. The plugin output files can be loaded into the Broad Institute's IGV or other third-party tools for further visualization or analysis. The following details reports are output reports generated by the Torrent Variant Caller plugin.
Barcode Coverage and Variants Report
The Barcode Coverage and Variants Report is a table that provides a summary list of reports per barcode. For each barcode, various types of data are displayed that can help you evaluate the quality of your run.
The following table provides descriptions of the Barcode Coverage and Variants Report columns. Column Variant Caller Reports Description Individual reports for each barcode. For a non-barcoded run, this report is not generated. Number of reads that map to the reference genome. Number of reads that overlap the target region. Number of bases that overlap the target region.
Mapped Reads Reads On-Target Bases On-Target
Version 2.2
Read Depth
Average coverage in all targeted regions. (The number of bases in reads that overlap the targeted region, divided by total length of the region.) Percentage of the targeted region that has at least 1x coverage. Percentage of the targeted region that has at least 20x coverage. Percentage of the targeted region that has at least 100x coverage. Number of variants detected in the targeted region.
1x Coverage
20x Coverage
100x Coverage
Variants Detected
Variant Caller Report
Variant Caller Reports are individual reports (per barcode) which can be selected from the Barcode Coverage and Variants Report shown above. While the Barcode Coverage and Variants Report provides a summary of all available barcode reports, the Variant Caller Report provides additional data, such as variant categories and information in HotSpot regions.
The following tables provide descriptions of the Variant Caller report columns. Report Header Column Number of mapped reads Percent reads on target Description Total number of reads mapped to the reference. The percentage of reads mapped to any targeted region relative to all reads mapped to the reference. The total number of target bases covered by any number of aligned reads.
Number of mapped bases
Version 2.2
Percent bases on target
The percent of all bases covered by reads aligned to the reference that covered bases in target regions.
Report Tables - Target Regions / HotSpot Regions Column Bases in target regions Description The total number of bases in all target regions of the reference. The average number of reads of all targeted reference bases. This is the total number of base reads on target divided by the number of targeted bases, and therefore includes any bases that had no coverage. The percentage of target bases covered by at least 0.2x the average base coverage depth. The percentage of target bases covered by at least 1 read. The percentage of target bases covered by at least 20 reads. The percentage of target bases covered by at least 100 reads. Number of called heterozygous SNPs in target regions or loci. Number of called homozygous SNPs in target regions or loci. Number of called heterozygous INDELs in target regions or loci. Number of called homozygous INDELs in target regions or loci.
Average base coverage depth
Uniformity of coverage
Coverage at 1x
Coverage at 20x
Coverage at 100x
Heterozygous SNPs
Homozygous SNPs
Heterozygous INDELs
Homozygous INDELs
Heterozygous SNPs and INDELs differ from the reference in only one copy of the chromosome, while homozygous SNPs and INDELs differ from the reference in both copies of the chromosome.
Variant Calls Summary
The Variant Calls Summary table shows the number of variants per chromosome. In this table, you can select the number of entries to display, filter the table based on search terms, or export the table for use with other third-party data analysis tools.
Version 2.2
The Variant Calls Summary table is new in the Torrent Suite Software 2.0 release. You can use this report monitor the quality of your run on a per-chromosome basis. For example, if one chromosome has an unexpected number of variants, this might indicate an issue with the run.
The following table provides descriptions of the Variant Calls Summary Report columns. Column Chromosome Description The chromosome (or contig) name in the reference genome. The total number of variants called (in the target regions of) the reference. The total number of heterozygous SNPs called (in the target regions of) the reference. The total number of homozygous SNPs called (in the target regions of) the reference. The total number of heterozygous INDELs called (in the target regions of) the reference. The total number of homozygous INDELs called (in the target regions of) the reference. The total number of variants identified with one or more HotSpots.
Variants
Het SNPs
Hom SNPs
Het INDELs
Hom INDELs
HotSpots
Variant Calls
The Variant Calls report has been updated for the Torrent Suite Software 2.0 release. New fields include: Gene Sym, Var Type, Ploidy, and HotSpot Annotation.
Version 2.2
The following table provides descriptions of the Variant Calls report columns. Column View Description Click IGV to open the variant in the Broad Institute's Integrative Genomics Viewer to see all reads covering the variant. The IGV is hosted by the Broad Institute, so internet connectivity is required from the machine running your web browser. Use of the IGV also requires a Java Runtime Environment installed on the desktop.
Chromosome
The chromosome (or contig) name in the reference genome. The one-based position in the reference genome. Gene Symbol for the gene where the variant is located. This value is not available (N/A) if no target regions were defined (full genome was used). Name of the target region where the variant is located. This value is not available (N/A) if no target regions were defined (full genome was used). Type of variation detected: SNP (single nucleotide polymorphism), MNP (multinucleotide polymorphism), IN (insertion), or DEL (deletion).
Position Gene Sym
Target ID
Var Type
Version 2.2
Ploidy
Assigned ploidy of the variation: HOM (homozygous), HET (heterozygous), or NC (no call). The reference base(s). Variant allele base(s). Frequency of the variant allele. Estimated probability that the variant could be produced by chance. (The smaller the p-value, the higher the confidence in the variant call.) The total reads covering the position. The number of reads covering the reference allele. The number of reads covering the variant allele. The HotSpot ID for one or more starting locations matching the identified variant. (Examples: cosmic ID, dbSNP ID.)
Ref Variant Var Freq P-value
Coverage Ref Cov Var Cov HotSpot ID
Allele Coverage for all Bases in HotSpot Regions
The Allele Coverage for all Bases in HotSpot Regions report provides allele- and strand-specific coverage information for each location of interest, as defined in a HotSpot BED file. The purpose of this table is to provide investigators with additional information about locations of interest, in the event that a variant wasn't called at the location of interest and wasn't shown in the Variant calls table. The Allele Coverage for all bases in HotSpot regions report has been updated for the Torrent Suite 2.0 release. New columns include: HotSpot ID, Cov (+), and Cov (-).
The following table provides descriptions of the Allele Coverage for All Bases in HotSpot Regions report columns.
Version 2.2
Column Chromosome
Description The chromosome (or contig) name in the reference genome. The one-based position in the reference genome. Name of the target region containing the HotSpot variation site. Name of the HotSpot variant site. (Examples: cosmic ID, dbSNP ID.) The reference base(s). The total reads covering the position. Number of reads calling A. Number of reads calling C. Number of reads calling G. Number of reads calling T. Number of reads calling either an insertion or deletion at this base location. Number of forward reads aligned over the reference base. Number of reverse reads aligned over the reference base.
Position Target ID
HotSpot ID
Ref Coverage A C G T INDEL
Cov (+)
Cov (-)
File Links
The File Links section contains links to the plugin results files. The variant calls files (VCF files) are in VCF 4.1 format. The plugin output files can be loaded into the Broad Institute's IGV or other third-party tools for further visualization or analysis. If your run includes barcoded data, barcode file links also appear in this section.
Version 2.2
The following table provides descriptions of the files available in the File Links section. File Type Variant calls file Description Variant calls generated by this plugin, in table format: t extfile.xls Allele counts generated by this plugin, in table format: t extfile.xls SNP calls generated by this plugin, in VCF format: textf ile.vcf Index of the SNP calls file, generated from the SNP calls file. Format: binaryfile.vcfidx INDEL calls generated by this plugin, in VCF format: te xtfile.vcf Index of the INDEL calls file, generated from the INDEL calls file. Format: binaryfile.vcfidx Target regions file generated by this plugin, in BED format: textfile.bed Target HotSpots file generated by this plugin, in BED format: textfile.bed Mapped reads file generated by this plugin, in BAM format: binaryfile.bam Index of the mapped reads file, generated from the mapped reads file. Format: binaryfile.bai Primer-trimmed mapped reads file generated by this plugin, in BAM format: binaryfile.bam
Allele counts file
SNP calls file
SNP calls VCF index file
INDELs file
INDEL calls VCF index file
Target regions file
Target HotSpots file
Mapped reads file
Mapped reads index file
Primer-trimmed mapped reads file
Version 2.2
Primer-trimmed mapped reads index file
Index of the primer-trimmed mapped reads file, generated from the primer-trimmed mapped reads file. Format: binaryfile.bam
TBAR Table Of Contents
Version 2.2
Contents
Torrent Browser Analysis Report Guide Library Summary Library Summary Overview Library Summary Report Barcode Reports Test Fragment Report Ion Sphere Particle Summary Report Information File Links Plugin Summary Running the Installed Plugins Alignment Plugin combineAlignment Plugin Coverage Analysis Plugin IonReporterUploader Plugin Run RecognitION Plugin sampleIdentifier Plugin Torrent Variant Caller Plugin
Barcode Reports
Barcode Reports
The Barcode Reports section displays histograms for the following metrics per barcode:
Chart Total number of reads
Description Total number of filtered and trimmed reads independent of length. This number is reported in the barcode SFF and FASTQ files. Number of base pairs of sequence from reads with aligned quality score of AQ20 or better.
AQ20 Bases
Version 2.2
Mean AQ20 read length
An AQ20 read length is the length, in bp units, that after alignment has a Phred-like score of 20 or better, or one error in 100 bp. This histogram charts the mean of these AQ20 read lengths per barcode. The number of reads with an aligned quality score of AQ20 or better.
AQ20 Reads
Version 2.2
Version 2.2
The number of barcodes shown in the reports varies according to the barcode set used in your run and on the barcodes actually present in the sample. Only data for barcodes present in the run are displayed in the chart. Each bar is labeled with the barcode ID. Data labeled as barcode ID X reports the number of unclassified barcodes: reads which could not be classified as matching one of the expected barcodes in the barcode set. Contents Torrent Browser Analysis Report Guide Library Summary Library Summary Overview Library Summary Report Barcode Reports Test Fragment Report Ion Sphere Particle Summary Report Information File Links Plugin Summary Running the Installed Plugins Alignment Plugin combineAlignment Plugin Coverage Analysis Plugin IonReporterUploader Plugin Run RecognitION Plugin sampleIdentifier Plugin Torrent Variant Caller Plugin
Version 2.2
Torrent Browser User Interface Guide

Introduction
Torrent Browser, the main Torrent Suite software user interface, provides access to the following tasks: Plan future runs to be executed on PGMs. View a Detailed Report for a specific run. View summary statistics from several analyses. Find a specific run or report, using filter or search criteria. Restart an analysis from a completed run. Configure various Torrent Suite software parameters to control archiving, reporting and other administrative functions. You can also access all documentation, support, and licensing information at the bottom of the main page:
The Torrent Browser interface is organized into a collection of tabs:
The easiest way to learn more about Torrent Browser functionality is to review the topics presented in each user interface tab: Planning Tab Runs Tab Reports Tab Services Tab References Tab Config Tab About Tab For paired end runs, see also Working with Paired-End Data.
Version 2.2
Contents Torrent Browser User Interface Guide Planning Tab Runs Tab Basic Runs Tab Interface Working with Runs Reports Tab Services Tab References Tab Working with Test Fragments Working with Reference Sequences Working with BED files Working with Obsolete Reference Sequences Managing Barcodes and Barcode Sets Working with Paired-End Data Config Tab About Tab
Planning Tab
Torrent Browser User Interface Guide Planning Tab
Use the Planning tab to create and edit run information for runs to be executed on PGM sequencers associated with this server instance. In the Planning tab you enter the same information as is required on the PGM Run Info screen. The Planning tab provide alternate methods (and timing) of entering the same data that is otherwise entered on the PGM Run Info screen. With the Planning tab, you can enter the information in advance, and have an opportunity to print and review your entries. Use of the Planning tab reduces your hands-on time on the instrument. If you do not create pending runs here in the Planning tab, you enter the run information on the PGM Run Info screen. Overview Run Planning Workflow Creating a Planned Run Executing a Pending Run on a PGM Sequencer For information on run plans for paired-end data, see Paired-End Planning Tab. Overview
Version 2.2
The Planning tab is divided into two areas. The top panel provides for adding plans either manually or uploading from a CSV file:
The second panel lists the pending runs not yet executed on a PGM sequencer. Each row represents an individual run and contains the following information about the run: Label Plan Name Short Code Value Name of the planned run. A short code identifying the planned run. Name of the project. Name of the sample. Name of the reference library used. Date the planned run was created.
Project Sample Library Date
The following is an example of a Planning tab with several pending runs. You select your pending run on the PGM Run Info screen.
Run Planning Workflow
Run planning allows you to enter run information via the torrent browser rather than at the PGM sequencer's Run Info screen. On the PGM sequencer, the pending run information is applied to the current Run Info screen by
Version 2.2
entering the pending run's short code or selecting the pending run from a list of pending runs. Any run information applied to the Run Info screen by selecting a pending run can be overwritten (changed) at the PGM system. Creating a Planned Run
A planned run can be created by selecting the Add Plan button in the planning tab or by uploading a CSV file by selecting the Upload Plans button. With the Add Plan button, you enter the run information in the Torrent Browser Add new plan screen:
Version 2.2
The plans now use the number of flows instead of the number of cycles. The Target Regions and HotSpot Regions BED files must be valid for the reference genome used for alignment. The BED files must contain the correct coordinates for the reference genome (and for the particular version of the reference genome). BED files are added existing references, in the References tab. (The BED files must be added in the References tab before they are available here.) See Working with Reference Sequences.
Version 2.2
The following Run Types are supported: Generic Sequencing (GENS): Generic sequencing, the PGM default. Does not support the Torrent Variant Caller plugin. AmpliSeq (AMPS): The Ion AmpliSeq workflow, which includes the Ion AmpliSeq Cancer Panel and Custom Ion AmpliSeq panels. TargetSeq (TARS): The TargetSeq workflow, with parameters optimized for hybridization-based target enrichment. Whole Genome (WGNM): Whole genome workflow, which does not use enrichment or a target regions file. Variant Frequency applies to TargetSeq and Ion AmpliSeq workflows. The following values are supported: Germ Line: Call only homozygous or heterozygous SNPs and indels. Somatic: Detects variants occurring at frequencies greater than 5%. If your Torrent Server uses the IonReporterUploader plugin, and the plugin is configured and enabled, an Ion Reporter workflow menu also appears in the plan page. See The IonReporterUploader Plugin for instructions for a run plan with the plugin. (Ion Reporter software is available under a separate license.) Paired-end data must be analyzed as paired-end. Check the "Is this plan intended for a Paired-End run?" checkbox. You must analyze paired-end data as paired-end even if you only run the forward direction.
Click Save Plan when the plan is ready. The new plan appears as a pending run in the Planning page, and can be selected on the PGM sequencer. (See Executing a Pending Run on a PGM Sequencer.)
Adding Multiple Plans from a CSV File
With the Upload Plans button, you can add one or more plans from an existing CSV file. 1. Click the Upload Plans button to open the Add new plans screen.
2. To see an example of the expected CSV file, click Download the example. 3. When your CSV file of plans is ready, click the Browse button, select the CSV file, and click Open. (If the Torrent Browser detects a problem with the CSV file, the "Plans CSV file" area is highlighted in yellow or red.) 4. Click Upload & Save. Executing a Pending Run on a PGM sequencer
A pending run created on the Torrent Server is executed on the PGM sequencer by selecting it from the PGM sequencer's Run Info screen. With the browse button you can select a pending run from a list of pending runs previously created on the Torrent Server. The change button allows you to select a pending run via its short code.
Version 2.2
The pending run information is populated into the PGM Run Info screen. You can optionally change run information on the Run Info screen. When ready, click Next --> to start your PGM run. Your pending run is removed from the Planning tab when you approve the run confirmation.
The pending run short code can be entered by entering it manually from the touch screen or by scanning in its barcode available by selecting the barcode button in the pending run list header.
You can also type the pending run short code (for example, P7GNF) into the Pending Run: text field on the PGM Run Info screen:
Version 2.2
Runs Tab
Runs Tab
Use the Runs tab to search, filter and sort the PGM system runs to display, and to start and terminate a run.
Version 2.2
Basic Runs Tab Interface Working with Runs Working with Paired-End Runs Contents Torrent Browser User Interface Guide Planning Tab Runs Tab Basic Runs Tab Interface Working with Runs Reports Tab Services Tab References Tab Working with Test Fragments Working with Reference Sequences Working with BED files Working with Obsolete Reference Sequences Managing Barcodes and Barcode Sets Working with Paired-End Data Config Tab About Tab
Basic Runs Tab Interface

Torrent Browser User Interface Guide Basic Runs Tab Interface
Version 2.2
Use the Runs tab to search, filter, and sort PGM runs to be displayed.
The dialog is divided into three areas: The top panel provides the filtering criteria for selecting runs to be displayed. The middle panel displays a list of runs that matched the run filtering specification. The bottom panel displays the current page number. Use the left and right arrows to navigate between the previous and next pages. Metadata tags entered on the PGM (Project, Sample) are listed in the run entry to aid in finding the run of interest.
Filtering
To filter the runs to be listed: 1. Select from the metadata tags available using the drop-down menus (PGM, Project, Sample, Library, and S torage). 2. For additional filtering, select the Starred checkbox, to the left of the run, to filter the runs to those that are checked. 3. Click Go, on the right side of the filter bar, to apply the filters you specified. Only those runs matching all metadata criteria are displayed.
Searching
To find and display a run: 1. Enter a full or partial run name in the Search by Run text edit window, on the right. 2. Click Go. The search is a simple text search and is not case-sensitive. Boolean operators and wildcards are not supported.
Version 2.2
Sorting
Sort the list of displayed runs using the up arrow (ascending order) and down arrow (descending order) next to each column heading, as shown in the following figure:
Expanding the Run View
Click the plus icon to the left of the run name to expand the run entry and see all of the reports generated for a run. The expanded view of reports shows run status and summary statistics for the run. The report Status shows Starte d if report generation is in progress. A list of the possible values displayed in the expanded view of a run is described in Reports Tab. Click the report name, in the Report field, to display the full report.
Monitoring Storage Capacity
A color-coded storage indicator light is located to the left of the Storage header. This indicator allows you to monitor space available on your Torrent Server hard drive.
Color Green
Meaning Your Torrent Server hard drive is less than 70% full. The height of the green bar corresponds to the percentage of disk space in use. Your Torrent Server hard drive is between 70% and 90% full. Your Torrent Server hard drive is more than 90% full.
Yellow
Red
For additional information about disk space, please refer to the Monitoring Free Disk Space secti on of the Torrent Server Administration Guide.
Controlling Raw Data Storage for a Run
Use the drop down menu under the Storage heading for each run to select the data archiving option: Archiving Option Delete Raw Operation Delete the data without archiving.
Version 2.2
Archive Raw
Archive a copy of the data. The DAT files are copied onto the archive drive. The Torrent Server is updated with a symbolic link to the archived version of the files. Retain the data on the Torrent Server.
Keep
The Archive Raw option requires configuration. The Delete Raw option only takes affect if the Archive Raw option is configured. For additional information about data archiving, please refer to the Archiving and Deleting Data s ection of the Torrent Server Administration Guide.
Working with Runs

Working with Runs Starting an Analysis Job Terminating an Analysis Run Changing the Analysis Reference Changing Run Metadata For information on paired-end runs, see Paired-End Runs Tab.
Version 2.2
Starting an Analysis Job
1. Click the Runs tab and find your run name.
2. Click Analyze, to the left of the run name, to display the run analysis dialog shown in the following figure.
Click the Advanced button to display additional run parameter options.
Version 2.2
The fields are described below: Parameter Report name Paired-end forward Description Text string of the report name. For use with paired-end runs, to create a composite report from separate forward and reverse runs. See Paired-End Runs Tab. For use with paired-end runs. See Paired-End Runs Tab.
Paired-end Reverse
Version 2.2
Start reanalysis from
The Analysis Pipeline proceeds through three stages: Signal Processing, Base Calling, and Alignment. Normally report generation proceeds through all three steps, but if you have already generated a report, it is possible to reanalyze the experiment and skip the earlier stages of the pipeline. For example, you may wish to change the genome that is used for Alignment. After changing the genome for the experiment on the Runs page using the Edit field, you need to reanalyze data to produce a new report using the new genome. But since there is no need to repeat the time consuming Signal Processing and Basecalling steps, you can use the output from an existing report as a starting point for Alignment, and the report will be completed much more quickly. You can restart the analysis from these points: Signal Processing (Default) Does not use the Use data from previous report field. You can optionally use both the Analysis args and Basec aller args fields. Base Calling Uses the Use data from previous report field and optionally the Basecall er args field. Does not use the Analysis args fiel d. Alignment Uses the Use data from previous report field. Does not use either of the Analysis args and Basecaller args fields. The selected option is in blue.
Use data from previous report
This option applies only when starting reanalysis from Base Calling or Alignment. In these cases, the results from a previous report are used as input for reanalysis. Analysis command line arguments. Should not be modified unless instructed by Ion Torrent Technical Support. Basecaller command line arguments. Should not be modified unless instructed by Ion Torrent Technical Support. Sequence used to identify library reads. Example: "TCAG". Sequence used to identify test fragment reads. Example: "ATCG".
Analysis args
Basecaller args
Library Key
TF key
Version 2.2
TF config
Name of an external file that specifies Test Fragment information. If a file is not specified, the default Test Fragment file is used. Custom arguments for the TMAP aligner. Leave blank to use the default TMAP parameters.
TMAP Args
3. Enter the parameter values, or accept the default values. Click Start Analysis to start the run, which checks the analysis status before displaying the run confirmation message:
4. You may view run progress by clicking Report or by selecting the run name in the Reports Tab:
5. Click Log to view run progress details in text form, clicking your browser refresh button to update the log.
Terminating an Analysis Run
To terminate a run that has not yet completed, go to the Services tab and click Terminate to the right of the run name. (See also: Terminating an Analysis Run, under Use Cases)
Changing the Analysis Reference
Use the following procedure to change the reference for an analysis: 1. In the Reference column, click the reference short name for the desired run:
Version 2.2
2. In the pop-up window, click the new reference name:
A Making change notification displays while the change is in progress. The Reference name is updated for the selected run:
Changing Run Metadata
The edit link in the Runs tab allows you to change the metadata for a run, for example, to add a note or to correct the following run metadata: Sample name Project name Run type Reference name Barcode (index) Library kit barcode Sequencing kit barcode
Version 2.2
You must restart or re-analyze your run for these changes to take effect. When you change the metadata, you change the information in the run database. Because the analysis pipeline is initialized with the run database information at the time that an analysis starts, changing metadata does not affect an analysis that is in progress. For a running analysis, you must terminate the run and start analysis manually. For a completed analysis, you must re-analyze the run.
Points to know about changing metadata: The run report (in the Reports tab) always shows the metadata in effect for the run. If your changes are not shown in the run report, the changes were not in place at the time the report was generated. If you add or change an entry in the Notes field, that note does not appear in the run report unless you restart or re-analyze the run (even though the note does not affect the analysis results). The location of the edit link for a run:
The edit link opens the Edit Run page:
Version 2.2
Version 2.2
Follow these steps to change metadata for a run: 1. Locate the run in the Runs tab, and click the edit link for that run. 2. In the Edit Run page, make your corrections to the metadata. 3. When you are done, click Save Experiment. 4. Confirm your changes in the Runs tab. 5. Restart the run: a. If the run is in progress, terminate the run and restart it. 6. If the run is completed, re-analyze the run.
Your changes do not affect a run that is in progress. Notes: For information about the paired-end checkbox, see: Working with Paired-End Data Paired-End Planning Tab Paired-End Runs Tab The ChipBarCode field contains a chip identifier. The chip barcode should not be confused with chemical barcodes and barcode sets (which are described here: Managing Barcodes and Barcode Sets).
Version 2.2
Reports Tab
Reports Tab
Use the Reports tab to search, filter and sort analysis reports for runs on this server instance. For information on paired-end reports, see Paired-End Reports.
Version 2.2
The dialog is divided into the following areas: The top panel provides for selecting run display filtering criteria. The second panel enables the export of analysis report metrics to a CSV file for download. The third panel lists the reports that matched the filtering criteria. A row represents an individual run and contains the following information about the run:
Label Report Status Run Chip Flows Lib Key Signal
Value Name of the report. Visual analysis report progress bars and status text. Original run name. Chip type. Number of nucleotide flows that were analyzed. Strength of the signal associated with the library key. Number of base pairs of sequence in predicted Q20 stretches. Number of reads 100 bp or longer at Q20 or better based on alignment. Number of base pairs of sequence from reads with Q20 or better based on alignment.
Q20 Bases
100 bp AQ20 Reads
AQ20 Bases
Version 2.2
Date
Date the analysis report was generated.
The bottom navigation panel displays the current page number. Use the left and right arrows to navigate between the previous and next pages. Filtering, Searching, and Sorting
Filtering and Searching
In the search/filter panel, you can select a metadata tag from the drop-down menus to select reports to display only reports with a matching tag. You can also enter the name, or partial name, of a report in the Search by Report text edit window to display that report.
Sorting
The resulting report list can be sorted using the arrows next to the report metadata parameter, in the list heading:
Sorting can be specified in ascending (up arrow) or descending (down arrow) order. Viewing Report Progress
A progress bar, to the right of the report name, shows the analysis phase for runs with a Status of Started:
Analysis progresses through the following stages: Well characterization Signal processing Base calling FASTQ creation Read alignment A yellow box indicates processing is in progress for the associated stage. A green box indicates processing has successfully completed. The Status changes to Completed when the run completes and the Re-Analyze from a 1.wells file icon replaces the progress bar. The Status column can have the following text, indicating run status: Status Started Completed Meaning Analysis is currently processing. Run analysis has completed.
Version 2.2
ERROR TERMINATED Checksum Error
Run analysis failed. Check run log for specific error. User terminated analysis job. One of the raw signal (DAT) files is corrupt. Try again to transfer the data from the PGM sequencer. Unexpected raw data values. Check the PGM sequencer for clogs or problems with the Chip.
Separator Abort
Viewing an Expanded Individual Report
A report entry in the list can be expanded by clicking the plus icon to the left of the entry. Separate rows display Test Fragments detected in the run above a minimum performance threshold. Additional details about Test Fragment performance are provided in the Viewing a Default Analysis Report section. Name TF Name TF Reads Value Test Fragment label. Number of reads that are associated with this Test Fragment. Measure of strength of the voltage signal associated with this Test Fragment. Average length of this Test Fragment, in base pairs.
TF Key Signal
TF AQ17 Mean
Viewing a Default Analysis Report
The Default Analysis Report can be viewed using one of the following methods: Click report name, under Report, to open the detailed report in the current window. Click the icon to the right of the report name to open the report in a new window.
Refer to the Torrent Browser Analysis Report Guide for a description of the report.
Re-analysis from a Wells File
Signal processing and base calling are separate analytical Torrent Suite software processes, and algorithms can change with different versions of the software. Following a software upgrade, an updated version of the base calling software may be available, indicated by a new
Version 2.2
version number for the ion-analysis component in the About Tab. If you would like to re-run the base calling part of the algorithm, starting from a wells file, without changing the (more lengthy) signal processing steps, click the Re-Analyze icon to the right of the experiment name, as shown in the following figure:
Services Tab
Services Tab
Version 2.2
Use the Services tab to view the current status of the essential system services: Jobs Server Service Crawler Service Archive Service
The Job Server and Crawler services should always be active. The Archive service is only active if archiving has been configured. If a service is no longer active, refer to the Troubleshooting sec tion of the Torrent Server Administration Guide.
Jobs Server Service
The Jobs Server panel lists active and queued analysis jobs on the Torrent Server, including server information and active (running) jobs:
Version 2.2
When no job is currently active, the Jobs Server panel displays a No active jobs message.
Starting a Job
Job requests are initiated by clicking the Analyze button in the Runs Tab for a given run or by specified auto-analys is on the PGM for the run. After data transfer following PGM auto-analysis completes, an analysis job starts automatically for that run. Up to two jobs can be run concurrently, by default, using the Sun Grid Engine (SGE). Running jobs in parallel results in a longer run time for each job but gives a shorter overall processing time compared to running the jobs concurrently.
Stopping a Job
In the Active Jobs panel, you can stop a job by clicking Terminate on the line for the job. Crawler Service
The Crawler panel displays information about processes transferring data from PGM sequencers to the Torrent Server.
Field Crawler Uptime
Description How long the Crawler has been scanning directories for data from the PGM sequencers since its last restart. Total number of PGM runs that have been added to this Torrent Server since the original initiation of this Torrent Server. Displays a list of the last several PGM runs added to this Torrent Server. Displays a changing list of folders showing where the Crawler is looking for data when it is in "working" mode. Current state of the Crawler: Working or Sleeping.
Number of Runs Added
Recently Added Runs
Currently Inspecting Folder
State
Version 2.2
Running on Host
Name of the computer hosting Torrent Suite.
Archive Service
The Archive panel is used to configure archive locations for storing raw data. The Archive panel also displays a dashboard of file space and archive space usage, and lists completed runs that are set to have their raw data archived or deleted.
Refer to the Archiving and Deleting Data section of the Torrent Server Administration Guide for detailed instructions on using the Archive panel, including mounting storage devices and specifying the Archive directory. Contents Torrent Browser User Interface Guide Planning Tab Runs Tab Basic Runs Tab Interface Working with Runs Reports Tab Services Tab References Tab Working with Test Fragments Working with Reference Sequences Working with BED files Working with Obsolete Reference Sequences Managing Barcodes and Barcode Sets Working with Paired-End Data Config Tab About Tab
Version 2.2
References Tab
References Tab
Use the References tab to enter nucleotide sequence Test Fragments, Reference Genomes for aligning reads, and Barcodes for barcode set management.
Working with Test Fragments Working with Reference Sequences Working with Targeted Regions BED files and Hotspots BED files Working with Obsolete Reference Sequences Managing Barcodes and Barcode Sets
Version 2.2
Working with Reference Sequences

Working with Reference Sequences As part of the standard analysis process, reads are aligned to a genomic reference, using the TMAP aligner that comes pre-installed on the Torrent Server. The alignments and some summary statistics based on the alignments are included in the Library Summary Report of the Detailed Analysis Report. For a new genome sequence, use the References tab to add the new reference genome. (These reference sequences are also displayed on the PGM when you load a sample.) Internet Explorer 6.0 and earlier are not compatible with this tool. Please download the latest version of Internet Explorer, Firefox, or Chrome.
Adding a Reference Sequence

Prerequisites
Create a FASTA format reference sequence file (on your client machine).
Version 2.2
FASTA files can be found at: http://www.ncbi.nlm.nih.gov/sites/genome Download the FASTA file to your local client machine. It is important that the format of your FASTA file conform to Ion Torrent requirements. Define a long form of the genome name. Enter a description for the genome. Define the number of reads to randomly sample for alignment. Create a regions of interest file or HotSpots file (on your client machine).
Procedure
Adding a reference sequence involves the following operations: Enter genome information. Upload a FASTA file. Wait for the genome index to be created. Upload a regions of interest file or HotSpots file. Click Cancel at any time to end your session without adding a new reference. 1. Click Add in the References Sequences bar:
2. In the Enter Genome Information dialog, enter the required genome names and version, and any optional information:
Enter the following information in the applicable fields:
Version 2.2
Field Description of the genome
Description [required] This entry may be any text string. The description usually includes the genus-species, version, and other identifying information. The description entered here is displayed in various report output, and is listed in the Reference Sequences section of the References tab. [required] A shortened form of the genome name, the short form of the genome name may be any alphanumeric character and the underscore (_) character. The name should not match any existing references installed in the /results/referenceL ibrary /<index_type>/<genome_shortname>/ directo ry, including previous unsuccessful attempts at creating reference sequences. Undesired sequences can be removed (see #Deleting a Reference Sequence). Deletion allows the short name to be used for a new genome. [required] Enter any string for the genome version number. The accession number, if there is one, is a good choice. The version entered here is displayed in various report outputs. [optional] Use this field to specify the number of reads to randomly sample for alignment. If left empty, the default is to align all reads. For larger genomes use smaller values to keep the alignment QC runtime short.
Short form of genome name
Genome version
Number of reads ...
Notes
[optional] Use this field to record any notes about the reference genome.
3. Click Select file to browse the genome FASTA file to upload.
When working with larger genomes, performance improves if you first zip the FASTA file. The create index tool supports a zip archive, provided the file contains only a single FASTA file. It is important that the format of your FASTA file conform to Ion Torrent requirements. 4. The "Select file to upload" file browser opens. Your Torrent Server host is named in the title:
Version 4. 2.2
5. Use the file browser dialog to select a FASTA file to upload, and click Open.
FASTA files can be found at: http://www.ncbi.nlm.nih.gov/sites/genome
6. The Torrent Browser returns to the "Information about the Genome" dialog. This screen now displays the filename and size of the file you selected.
7. Click Upload file and create reference to start uploading the file to your server.
Version 2.2
To provide a better uploading experience, Adobe Flash or Microsoft Silverlight plugins are required to be installed for your browser. You may need to contact your local system administrator for assistance. Silverlight can be downloaded from http://www.silverlight.net/getstarted/. Adobe Flash can be downloaded from http://get.adobe.com/flashplayer/.
8. The "Information about the Genome" dialog displays a progress bar as the file is uploaded.
The time required to upload the file depends on the genome size. Small genomes can be uploaded in less than a minute, whereas the larger genomes can take a couple of hours. 9. When upload completes (100%), the Torrent Browser creates an index for the genome:
10. The Torrent Browser opens the References Tab. A started status indecates the genome index is being prepared:
A found status indicates that genome index creation is complete:
The reference sequence is enabled by default. 11. Follow the steps in Modifying Reference Sequence Information to add a regions of interest file or a HotSpots file to this reference.
Version 2.2
Modifying Reference Sequence Information
1. In the Reference Sequences bar, click the Name of the reference sequence you want to modify:
2. The current reference sequence information is displayed. Modify any field, as wanted.
3. If you are not adding a BED file (a regions of interest file or a Hotspot file), click Save to save your changes. Clicking Save returns you to the References tab dialog. Skip the remaining steps in this section. 4. To upload a regions-of-interest file or a Hotspot file, click Upload BED Files. The following dialog opens. Click the Select a New BED File button:
Version 2.2
5. In the file browser, browse to the BED file (on your local client machine), and click Open. Click Upload. The new BED file appears above the Select a New BED File button:
6. Click Upload. The Torrent Browser validates the BED file:
7. If an error is reported, click the Details link to see the error message. If no error is found, click Save.
See also Working with BED files for more detailed information about using BED files with your references.
Deleting a Reference Sequence
1. In the Reference Sequences bar, click the Name of the reference sequence you want to delete.
2. This displays the reference sequence management options. Click Delete to delete the reference sequence:
Version 2.2
2.
3. Click Delete to confirm your delete request and remove the reference sequence. Click Cancel to exit the dialog without deleting the reference sequence:
4. The dialog confirms deletion. Click X to return to the References tab:
Version 2.2
The deleted reference sequence is removed from the Reference Sequences list.
Error Handling
If you uploaded an invalidly formatted FASTA file, the following error displays when you attempt to view the reference sequence associated with the file:
To recover from the error: 1. Delete the existing reference sequence entry. 2. Identify and correct formatting errors in the FASTA file. 3. Retry Adding a Reference Sequence
Version 2.2
Working with Test Fragments

Working with Test Fragments Use the References tab to enter the Test Fragment nucleotide sequence to search for within the sequenced nucleic acids. You can give a Name label and Key to your Test Fragment sequence. If any sequence is detected in the system that matches a template sequence, information about the sequence is displayed in the Test Fragment Report section of the Detailed Analysis Report. Ion Torrent provides four Test Fragments by default.
Be sure to enter the Test Fragment sequence using only the uppercase letters: A, T, C and G. If you enter an invalid character or duplicate test fragment, you will not be able to save your changes. Contact Ion Torrent if you have questions about the Test Fragment templates installed in your Torrent Browser.
Editing a Test Fragment
Version 2.2
If Ion Torrent provides new Test Fragments as part of an updated protocol, it will be necessary to carefully cut and paste this information into the fields. Do not modify the Test Fragment sequences for the Test Fragments that are supplied by Ion Torrent: TF_A, TF_B, TF_C, and TF_D. 1. Click the Name column label to display the ideal ionogram associated with the Test Fragment. This example shows Test Fragment TF-C selected for editing:
2. Click Edit to change the Test Fragment nucleotide sequence. 3. Modify the current values as needed:
4. Click Save to save your changes. (Click Cancel to end your edit session without modifying the Test Fragment.)
Adding a Test Fragment
1. Click the Add button at the upper right corner to add a new Test Fragment.
This displays the New Template dialog with blank fields:
Version 2.2
2. Enter your values for the new Test Fragment:
3. Click Save to save your changes. Your new Test Fragment is displayed in the Test Fragment list:
(Click Cancel to end your session without adding a new Test Fragment.) Contents Torrent Browser User Interface Guide Planning Tab Runs Tab Basic Runs Tab Interface Working with Runs Reports Tab Services Tab References Tab Working with Test Fragments Working with Reference Sequences Working with BED files Working with Obsolete Reference Sequences Managing Barcodes and Barcode Sets Working with Paired-End Data Config Tab About Tab
Version 2.2
Working with Obsolete Reference Sequences

Working with Obsolete Reference Sequences After the Torrent Server is updated to software version 2.2, all previously generated reference sequence indexes are obsolete. Reference libraries are not automatically upgraded when installing version 2.2.0 software.
The list of currently installed libraries provides a checklist of the libraries that need to be upgraded after the update.
Torrent Browser aids you in identifying the obsolete sequences by automatically recording the libraries that were installed before the upgrade. You will need to upgrade these reference sequences using the procedures described in Working with Reference Sequences. The only reference library available after upgrade is E. coli DH10B, which is displayed in the Reference Sequences panel of the References tab and on the PGM sequencer genome choice list menu. The previous default Ion Torrent reference library, E. coli K12 is permanently removed.
Version 2.2
Managing Barcodes and Barcode Sets

Managing Barcodes and Barcode Sets This section describes how to manage barcode sets in the UI under the References Tab. (For a description of how to use barcodes with a PGM run, see Using Barcodes with the PGM.) Up to 96 barcodes are supported per set. With the pre-installed Ion Torrent barcodes, you can view the barcode sets and the barcodes, incuding the barcode sequences: View a Barcode or Barcode Set With your own barcodes sets, you can do the following: View a Barcode or Barcode Set Add a Barcode Set Delete a Barcode Set Edit or Delete an Individual Barcode Add a Barcode to an Existing Barcode Set
For Pre-installed Barcode Sets
For pre-installed barcode sets, the following functionality is supported:

Version 2.2
View a Barcode or Barcode Set The two pre-installed barcode sets are seen under the References Tab:
View a Barcode or Barcode Set
1. Click the References tab, and scroll down to the Barcodes panel. Click the name of the barcode set to view:
2. This displays the barcodes in the set:
Dialog buttons are displayed to add a new barcode to this set and to delete the entire barcode set. The barcode edit and delete feature is only for custom barcode sets that you install.
Version 2.2
Do not edit, delete, or modify the pre-installed barcode sets IonSet1 or IonXPress.
IonXpress Barcodes
Here are the barcodes in the IonXPress set:
Version 2.2
Version 2.2
IonXpressRNA Barcodes
Here are the barcodes in the IonXPressRNA set:
Version 2.2
RNA_Barcodes_None Barcodes
Here is the barcode in the RNA_Barcodes_None set:
For Custom Barcode Sets
For your own barcode sets, you can do the following: View a Barcode or Barcode Set Add a Barcode Set Delete a Barcode Set Edit or Delete an Individual Barcode Add a Barcode to an Existing Barcode Set
Add a Barcode Set
To add a barcode set, packaged as a list of barcodes in a Comma-separated Variable (CSV) text file, create the CSV file then select the file to add it to the barcode set list. 1. If needed, create the CSV file containing a maximum of 96 barcodes, using Microsoft Excel, OpenOffice Calc, or equivalent program. Save the file with a .csv extension. 2. Click the References tab, and scroll down to the Barcodes panel. Click Add on the right side of the Barcode s panel:
Version 2. 2.2
3. In the Add new DNA barcodes dialog, enter the required Barcode set name in the edit window and browse to find the Barcode csv file:
4. To view an example CSV file, click Download the example file:
The following table describes the column headers:
Name id_str
Type String
Description The unique name for this barcode entry. Not used. The barcode sequence. G, C, A, and T (always upper-case) are allowed. Not used. Used to sequentially label barcodes. Used when id_str not set to dynamically create the id_str from the barcode set name and this value. Not used.
type sequence
-String
flow order index
-Integer
annotation
--
Version 2.2
adapter
String
The portion of the barcode adapter not used to identify this barcode. Often referred to as the "stuffer sequence". G, C, A, and T (always upper-case) are allowed. 0 = percentage match (Not recommended) 1 = number of errors to accept (R ecommended) A barcode is scored by comparing each read in flow space to each barcode in flow space. Errors in each flow are summed over the length of the barcode flows. Once complete, if score mode 0 is requested, the sum of errors is divided by the total number of flows in the barcode, and an error value is calculated. 1.0 minus this error value is the percentage correct for the barcode. If this value is at least as large as the score_cutoff value, then this barcode can be considered for a best match. If this barcode has the highest percentage match to the input sequence as compared to all other barcodes, and this value is also higher than the cutoff value, then this barcode is considered a match to the input sequence. In score mode 1, a similar process occurs, except that the number of mismatches is used directly. Any barcode with the number of mismatches equal to or less than the score cutoff value can be considered, and the barcode with the fewest mismatches to the input sequence is the matching barcode. With score mode 0: a percentage threshold. With score mode 1: an error count threshold. (Recommended error count of 2)
score_mode
Integer
score_cutoff
Floating point
Version 2.2
Click Upload & Save to add the new barcode set.

Delete a Barcode Set
This feature is only for your own custom barcode sets. Do not delete the pre-installed barcode sets IonSet1 or IonXPress. 1. Display the barcode set contents as described in View a Barcode or Barcode Set. 2. At the bottom of the page, click Delete Barcode Set. This displays a delete confirmation prompt:
3. Click Yes to delete the entire barcode set. Click No to keep the displayed barcodes.
Add a Barcode to an Existing Barcode Set
This feature is only for custom barcode sets that you install. Do not edit, delete, or modify the pre-installed barcode sets IonSet1 or IonXPress.
1. Display the barcode set contents as described in View a Barcode or Barcode Set. 2. Enter the barcode sequence in the edit window and click Add Barcode:
3.
Version 2.2
3. Add the barcode information and click Save Barcode. The new barcode is added to the set displayed in the current barcode set list.
Edit or Delete an Individual Barcode
The barcode edit and delete feature is only for custom barcode sets that you install. Do not edit, delete, or modify the pre-installed barcode sets IonSet1 or IonXPress.
1. Display the barcode set contents as described in View a Barcode or Barcode Set. 2. Click the References tab, and scroll down to the Barcodes panel. Click on the ID of a barcode, such as Cust omBCSet_15:
3. To edit the barcode details, make your changes and click Save Barcode. 4. To Delete this barcode from the barcode set, click Delete Barcode. This displays a delete confirmation message:
5. Click Yes to delete the barcode and remove it from the list. Click No to keep the barcode in the set.
Version 2.2
Working with BED files

Working with BED Files Browser Extensible Data (BED) files supply chromosome regions and restrict analysis to only the specified regions. These two types of BED files are frequently used: Targeted regions of interest: Specifies the amplified regions that are used with targeted sequencing. HotSpot: Specifies regions of known mutations, for example from COSMIC or dbSNP databases or from customer-defined regions. With the Torrent Browser, you add BED files to an existing reference. The reference must be listed in the Torrent Browser References tab before you can upload our BED files. BED files are then available as an option when you create a new run registration in the Planning tab. In the Add New Plan page, the following menus offer the available BED files:
Version 2.2
You can optionally upload multiple BED files to a reference. On the Planning tab run registration page you specify the BED files used for each run. BED files provide an option to the analysis of the entire reference genome. Whole genome analysis is supported by the run type Whole Genome Analysis. Do not specify a BED file on the Planning tab run registration page if the variants are to be called over the whole genome. All regions specified in your BED files are analyzed. Follow the instructions in Modifying a BED File (before uploading your BED files) to delete lines representing regions that span variants that you do not wish to call. The BED file coordinates (example: chr2 29443689 29443741) use zero-based indexing and a half-open interval. The start position is included, and the range extends up to, but not including, the end position. BED files used with Ion AmpliSeq workflows define the internal segment only, and do not include the primer sequence. A BED file is tied to specific reference. The coordinates within a BED file must match coordinates in the reference genome.
Summary of Steps
Before your analysis run or run registration (on the Planning page), you can add BED files to your genome reference on the Torrent Server: 1. You use the Torrent Browser to upload the BED file from your local client machine to the Torrent Server. 2. During file upload, the Torrent Browser validates the BED file, and ensures that the BED file's coordinate regions are valid for the genome reference. 3. BED files are then available as an option when you create a new run registration in the Planning tab.
Modifying a BED File
You can optionally modify a BED file before adding it to your reference genome on the Torrent server. You can use this technique to avoid regions for which you do not want variants called (even if the variants appear in your sample). You can modify a BED file only before uploading the file with the Torrent Browser. Follow these instructions to modify a BED file: 1. Make a copy of your BED file. Rename the two files in a way that reflects changes you make to the regions being analyzed. 2. Open the BED file with a text editor. 3. Delete the lines for regions you do not want. 4. Save the file. If the region (or regions) appear in both your targeted regions BED file and in your HotSpots BED file, you must
Version 2.2
delete the line for those regions from both types of BED file.
Uploading a BED File
These instructions upload a BED file from your local client machine to the Torrent Server. These instructions support both targeted regions of interest files and HotSpot regions files. You must upload only BED files that both match the reference and are for the correct reference version. The uploader attempts to validate the BED files, but cannot always detect the errors listed below.
You have the responsibility to avoid the following mismatch errors. The uploader does not always detect these errors: 1. Upload a BED file to a reference genome of a different version (for example, an hg18 BED file with an hg19 reference). 2. Upload a BED file for a different species. 3. Upload a HotSpots BED file as a targeted regions BED file, or upload a targeted regions BED file as a HotSpots BED file. Follow these steps to upload a target regions BED file or HotSpots BED file to a reference: 1. In the Reference tab Reference Sequences section, click the Name of the reference sequence:
2. The current reference sequence information is displayed:
Version 2.2
3. Click Upload BED Files. The following selection section opens in the Available BED Files section:
Make sure this is the correct reference for this BED file. If this is a HotSpots BED file, click the HotSpot checkbox to enable it. 4. Click the Select a New BED File button. 5. In the file browser, browse to the BED file (on your local client machine), and click Open. The filepath appears in the field next to the Browse button. 6. Click Upload.
Version 2.2
6. The new BED file appears in the Processing BED Files section:
The Torrent Browser validates the BED file:
7. If an error is reported, click the Details link to see the error message. If no error is found, click Save.
BED files the appear in the Available BED Files section are available during run registration (under the Planning tab, in the Add New Plan page). This page does not automatically refresh. To confirm that the files are successfully uploaded, refresh the browser page. The files should be seen in the Available BED Files area (and not as Editing in the Processing BED Files area).
BED File Formats and Examples
BED file formats use tab-separated fields.

Target Regions File Formats
Target regions BED files use 3-column, 4-column, and 6-column formats. 3-column Target Regions BED File Format The 3-column BED file format is used when amplicon IDs and gene names are not known. The track line is optional. If present, it includes these tab-separated fields:
Field Name Description
Type string string
Description A unique design identifier. Optional. Description of the design. Optional.
Version 2.2
The following is an example track line:
track name="ASD270245" description="AmpliSeq
Pool ASD270245"
In a 3-column target regions BED file, the coordinates lines require the following tab-separated fields:
Field chrom
Type string (chars >= 0x20, other than \tab)
Description Name of the chromosome. This name must be an exact match with a chromosome in the reference. Starting position of the feature (zero-based). Ending position of the feature (not inclusive). Must be greater than chromStart.
chromStart
unsigned int64
chromEnd
unsigned int64
Partial example of a 3-column target regions BED file:
chr9 133738312 133738379 chr9 133747484 133747542 chr9 133748242 133748296 chr9 133748388 133748452 chr9 133750331 133750405 chr9 133738312 133738379 chr9 133747484 133747542 chr9 133748242 133748296 chr9 133748388 133748452 chr9 133750331 133750405 chr14 105246407 105246502 chr14 105246407 105246502 chr14 105246407 105246502 chr2 29432658 29432711
4-column Target Regions BED File Format The 4-column BED file format is used when gene names are not known and some or all amplicon IDs are known. The track line is optional. If present, it includes these tab-separated fields:
Field
Type
Description
Version 2.2
Name Description
string string
A unique design identifier. Optional. Description of the design. Optional.
The following is an example track line:
Pool ASD270245"
In a 4-column target regions BED file, the coordinates lines require the following tab-separated fields:
Field chrom
Description Name of the chromosome. This name must be an exact match with a chromosome in the reference. Starting position of the feature (zero-based). Ending position of the feature (not inclusive). Must be greater than chromStart. Amplicon ID. If missing, the following string is used "chrom" + ":" + "chromStart" + "-" + "chromEnd"
chromStart
unsigned int64
chromEnd
unsigned int64
AmpliconID
string
chr9 133738312 133738379 amplID73150 chr9 133747484 133747542 amplID73075 chr9 133748242 133748296 amplID73104 chr9 133748388 133748452 491413 chr9 133750331 133750405 74743 chr9 133738312 133738379 73150 chr9 133747484 133747542 73075 chr9 133748242 133748296 73104 chr9 133748388 133748452 491413 chr9 133750331 133750405 74743 chr14 105246407 105246502 329410 chr2 29432658 29432711 34014
Version 2.2
6-column Target Regions BED File Format The 6-column BED file format is used when some or all of the gene names are known. The track line is required in a 6-column target regions BED file. The following is an example track line:
Pool ASD270245"
type=bedDetail
The track line includes these tab-separated fields:
Field Name Description Type
Type string string string
Description A unique design identifier. Optional. Description of the design. Optional. Must be "bedDetail" (without quotes). Required.
In a 6-column target regions BED file, the coordinates lines require the following tab-separated fields: Field chrom Type string (chars >= 0x20, other than \tab) Description Name of the chromosome. This name must be an exact match with a chromosome in the reference. Starting position of the feature (zero-based). Ending position of the feature (not inclusive). Must be greater than chromStart. Amplicon ID. If missing, the following string is used "chrom" + ":" + "chromStart" + "-" + "chromEnd" GeneID or MiscID. If missing, set to '.'. This field is not used currently. Gene name. If missing, set to 'N/A'.
chromStart
unsigned int64
chromEnd
unsigned int64
AmpliconID
string
ID
string
GeneSymbol
string
Version 2.2
track name="ASD270249_v1" description="AmpliSeq Pool ASD270249" type=bedDetail chr9 133738312 133738379 AM73150 NM_005157 ABL1 chr9 133747484 133747542 AM73075 NM_005157 ABL1 chr9 133748242 133748296 AM73104 NM_005157 ABL1 chr9 133748388 133748452 AM491413 NM_005157 ABL1 chr9 133750331 133750405 74743 NM_005157 ABL1 chr9 133738312 133738379 73150 NM_007313 ABL1 chr9 133747484 133747542 73075 NM_007313 ABL1 chr9 133748242 133748296 73104 NM_007313 ABL1 chr9 133748388 133748452 491413 NM_007313 ABL1 chr9 133750331 133750405 74743 NM_007313 ABL1 chr14 105246407 105246502 329410 NM_001014431 AKT1 chr14 105246407 105246502 329410 NM_001014432 AKT1 chr14 105246407 105246502 329410 NM_005163 AKT1 chr2 29432658 29432711 34014 NM_004304 ALK
HotSpots File Format The track line is required in a HotSpots BED file. The following is an example track line:
track name="ASD270245" description="HotSpots locations for AmpliSeq ASD270245" type=bedDetail
The track line includes these tab-separated fields:
Field Name Description Type
Type string string string
Description A unique design identifier. Optional. Description of the design. Optional. Must be "bedDetail" (without quotes). Required.
In HotSpots BED files, the coordinates lines require the following tab-separated fields:
Field chrom
Description Name of the chromosome. This name must be an exact match with a chromosome in the reference. Starting position of the feature (zero-based).
chromStart
unsigned int64
Version 2.2
chromEnd
unsigned int64
Ending position of the feature (not inclusive). Must be greater than chromStart. This ID is either the COSMIC ID, dbSNP ID, or user-defined. If missing, the following string is used "chrom" + ":" + "chromStart" + "-" + "chromEnd" This field describes the variant, using this format (see examples below): REF=reference_allele;OBS=observ ed_allele;ANCHOR=base_before_a llele Amplicon ID. If missing, the following string is used "chrom" + ":" + "chromStart" + "-" + "chromEnd"
HotSpotName
string
HotSpotAlleles
string
AmpliconID
string
This section describes the HotSpotAlleles field. This field specifies the alleles involved in variant calls, using this format: REF=reference_allele;OBS=observed_allele;ANCHOR=base_before_allele. Example of TT insertion with 1-base prior at reference C: REF=;OBS=TT, ANCHOR=C. Example TT deletion with 1-base prior at reference G: REF=TT; OBS=; ANCHOR=G. For insertions and deletions, the reference base prior to an indel must be present as specified in ANCHOR. This field is required for left-aligning the region. If any of the REF, OBS, or ANCHOR fields are missing, then left alignment is not performed. If present, REF, OBS, and ANCHOR should be on the forward genomic strand. There should be one alternative allele per line. Insertion alleles should have the same start and end position, and that position corresponds to a region between two bases. SNV, MNV, deletion, and complex variants should correspond to the reference bases that are spanned by the event. HotSpotAlleles are always reported based on the allele information from the positive strand of the reference sequence. Even if the allele strand is negative, the REF, OBS, and ANCHOR bases still report the alleles on the positive strand. For example, if there is a hotspot either on the positive strand or on the negative strand on a genomic coordinate, the strand information makes no difference to what is reported on the HotSpotAlleles column. HotSpotAlleles column always reports the alleles on the positive strand. In the following example, the strands are different, but the reported alleles are always from the positive strand: chr 143815007 43815009 ID1 0 - REF=TG;OBS=AA;ANCHOR=G AMPL1 chr 143815007 43815009 ID2 0 + REF=TG;OBS=AA;ANCHOR=G AMPL2 Partial example of a HotSpots BED file:
Version 2.2
track name="ASD270249_v1" description="AmpliSeq Pool ASD270249" type=bedDetail chr9 133738312 133738379 73150 NM_005157 93843 (c.726C>G, p.I242M, 133738326) chr9 133747484 133747542 73075 NM_005157 12602 (c.827A>G, p.D276G, 133747520) chr9 133748242 133748296 73104 NM_005157 43361 (c.931T>A, p.F311I, 133748270) chr9 133748388 133748452 491413 NM_005157 43361 (c.931T>A, p.F311I, 133748270) chr9 133750331 133750405 74743 NM_005157 43360 (c.1109T>G, p.L370R, 133750278) chr9 133738312 133738379 73150 NM_007313 93843 (c.726C>G, p.I242M, 133738326) chr9 133747484 133747542 73075 NM_007313 12602 (c.827A>G, p.D276G, 133747520) chr9 133748242 133748296 73104 NM_007313 43361 (c.931T>A, p.F311I, 133748270) chr9 133748388 133748452 491413 NM_007313 43361 (c.931T>A, p.F311I, 133748270) chr9 133750331 133750405 74743 NM_007313 43360 (c.1109T>G, p.L370R, 133750278) chr14 105246407 105246502 329410 NM_001014431 93893 (c.155T>G, p.L52R, 105246445) chr14 105246407 105246502 329410 NM_001014432 93893 (c.155T>G, p.L52R, 105246445) chr14 105246407 105246502 329410 NM_005163 93893 (c.155T>G, p.L52R, 105246445) chr2 29432658 29432711 34014 NM_004304 28058 (c.3833A>C, p.Y1278S, 29432655)
Config Tab
Config Tab
Version 2.2
The Config tab provides access to administrative functions.
Please refer to the Administration section of the Torrent Server Administration Guide, which describes Config tab functionality. The BED publisher is used internally to support BED file uploading to a reference (described in Working with BED files). User functions that also involve direct database access are described in Use Cases.
Version 2.2
About Tab
Help/About Tab
The Help/About tab lists the version numbers of the currently installed Torrent Suite software components, and provides links to other resources:
Version 2.2
The component version number should agree with the required version specified in the Release Notes. Refer to the latest Release Notes for a description of how to install components.
Version 2.2
UI Table Of Contents
Contents
Torrent Browser User Interface Guide Planning Tab Runs Tab Basic Runs Tab Interface Working with Runs Reports Tab Services Tab References Tab Working with Test Fragments Working with Reference Sequences Working with BED files Working with Obsolete Reference Sequences Managing Barcodes and Barcode Sets Working with Paired-End Data Config Tab About Tab
Working with Paired-End Data

Working with Paired End Data
Paired-end analyses involve a few different steps from other analyses. Paired-end data is analyzed as two separate runs, one for forward and one for reverse. Then the reports for the two runs are combined into a composite report. The composite report joins the data from two physical runs into one data set. The basic paired-end scenario is the following: You define the run in the Planning tab: You can specify you want to run both forward and reverse directions. You can run only the forward direction.
Paired-end analysis must account for adapter differences. Even a single-direction forward paired-end run must be specified as a paired-end run in the Planning tab.
These pages support paired end runs: Planning tab When you create a run plan for paired-end data, you specify the forward and reverse runs in the plan. (See Paired End Planning Tab.) You specify these in the run plan: The Forward Library Key and Forward 3-prime Adapter The Reverse Library Key and Reverse 3-prime Adapter
Version 2.2
Run tab With the Analyze button on the Runs tab, you start a new run to create a composite report on a related forward run and reverse run (see Paired-End Runs Tab):
Also, in the Runs tab, you can re-analyze a paired-end run. See Paired-End Runs Tab. Reports tab The Reports tab has separate reports for the forward run and the reverse run, and also a composite report. Plugins Plugin results for Paired End data are generated from the composite report.
Paired-End Adapters
Paired-end data has paired-end adapters. This data can only be analyzed as a paired-end run. If you do not analyze in both directions, the data must be analyzed as a paired-end forward run .
See also: Paired-End Runs Tab Paired-End Reports Tab Paired-End Planning Tab User Interface Guide Table Of Contents Report Guide Table Of Contents
Paired-End Planning Tab

Version 2.2
Creating a Plan for Paired-End Data Paired-end data is analyzed as two separate runs, one for forward and one for reverse. When you create a run plan for paired-end data, you specify the forward and reverse runs in the plan. Your selections include a paired-end checkbox and these keys and adapters: The Forward Library Key and Forward 3' Adapter The Reverse Library Key and Reverse 3' Adapter You create a new plan by selecting the Add Plan button, and fill out the main run information. Then you must indicate this is a planned Paired-End analysis by clicking the check box for Paired-End run as shown below.
When you click the check box, four drop-down menus are presented with options for Forward Library Key, Forward 3' Adapter, Reverse Library Key, and Reverse 3' Adapter as shown below.
Version 2.2
Tip: Use the Notes field to keep track of which two runs belong to the same paired-end data. Your notes appear in the Run tab, in the details for the two runs (the forward run and the separate reverse run).
You can use the defined default settings for both the Forward and Reverse Library Keys as well as the Forward and Reverse 3' Adapters. Alternatively, you can access the drop-down menu to choose the Forward and Reverse Library Keys and 3' Adapters you wish to run with your analysis as shown below.
For the reverse run, the correct Library Key is automatically presented as the default key. For most scenarios you select the default Library Key for the reverse run.
Click Save Plan when the plan is ready. The new plan appears as a pending run in the Planning page, and can be selected on the PGM sequencer. In the Planning Tab, your paired-end runs are listed under Pending Runs.
The forward and reverse runs are listed separately and are in separate plans. If before launch you need to edit common run information, you must edit both plans.
A Single Direction Run on Paired-End Data
Version 2.2
If your data in paired-end data, the data has paired-end adapters. This data can only be analyzed as a paired-end run (in order to have the correct handling of adapters). Even if you do only a forward run, and do not want to analyze in both directions, the data must be analyzed as a paired-end forward run. Paired-end data can only be analyzed as a paired-end run, whether you do both directions or only the forward direction.
Changing Default Run Plan Paired-End Settings
To add, delete, or change the settings for the Library Keys or 3' Adapters, you need administrative access to reset the attributes through the Config tab Django Admin Interface. Executing a Pending Paired-End Run on a PGM sequencer This section describes how a paired-end run that you created in the Planning tab is executed on the PGM instrument. A pending run created on the Torrent Server is executed on the PGM sequencer by selecting it from the PGM sequencer's Run Info screen. With the browse button you can select a pending run from a list of pending runs previously created on the Torrent Server. You can then choose from the list of plans to use for the run as shown below.
If the Paired-End planned run is chosen, the Reverse run button is disabled.
Version 2.2
See also: Working with Paired-End Data Paired-End Runs Tab Paired-End Reports Tab UI Guide Table Of Contents Report Guide Table Of Contents
Paired-End Reports
Reviewing the Paired-End Report
Once your Paired-End analysis is complete, you can access the Report either by clicking on the hyperlink from the Reports listed in the Run tab under the Forward runs file associated with the Paired-End analysis, or by using the Re ports tab in the Torrent Browser.
Version 2.2
Library Summary
The Library Summary section of the Report offers information about the Predicted Per-Base Phred Quality Scores that is independent of the Alignment to the reference genome. This Qualtiy Score matrix gives details based on the following distributions: Union - As in set theory, the concept of the Union of two data sets, Foward Reads U Reverse Reads, defines the Forwards reads AND the Reverse reads, as depicted in the Venn diagram below. Corrected - As in set theory, the Quality scores that are listed under the Corrected values can be associated with the Intersection both reads sets, also depicted in the Venn diagram. Unpaired Forward and Unpaired Reverse - The Per-Base Quality Scores for unpaired forward and reverse reads are calculated as reads that do not have a corresponding match.
Version 2.2
In the Library Summary matrix, calculations for Phred-based Qualtiy standard scoring are shown for Total Number of Bases (Mbp), for Q20 (Mbp). Also available are calculations for Total Number of Reads, as well as the Mean Length distribution in basepairs and the length of the Longest Read in basepairs for each category of the Paired-End analysis as shown in the example below.
Below the Library Summary matrix, you are presented with the Read Length Histogram distribution for all combined reads.
Version 2.2
Ion Sphere Particle (ISP) Identification Summary
Running a Paired-End Analysis will generate two ISP Summary Heat Maps for each runs file. The first Heat Map and Summay Matrix is associated with the Forward Reads file.
The second Heat Map and Summary Matrix is associated with the Reverse Reads file.
Paired-End Report Information
Analysis Info
Version 2.2
For both the Foward and Reverse runs, you can review the following information: Run Name Run Date Analysis Name Analysis Date Analysis Cycles Analysis Flows Project Sample Library PGM Chip Check, Chip Type and Chip Data Notes Barcode Set Run Ids Flow Order Library Key Information that is different between the two runs appears in red:
Software Version
The Software Version displays the release version for the current Torrent Software Suite for the Torrent Browser.
Version 2.2
File Links
For each Library reports generated from your analysis - Union, Corrected, Singleton Forward and Singleton Reverse - you can download and save the Library Sequence files: Library Sequence (SFF) Library Sequence (FASTQ) Full Library Alignments (BAM) Full Library Alignments (BAI) A single BAM file is generated for the Paired-End analysis, allowing you to run further analysis in the Plug-Ins section at the bottom of the Reports workspace. There is a link to generate a PDF of the full Paired-End Report and a link for the Customer Support Archive.
Plug-In Summary
To run the Variant Caller Plug-In from the Reports Tab, click on the Select Plug-Ins to Run button in the Plug-In Summary section of the Reports workspace.
You will be presented with a display box to chose which plug-in you wish to run, as in the example below:
See also:
Version 2.2
Running the Installed Plugins Paired-End Runs Tab Paired-End Planning Tab Working with Paired-End Data UI Guide Table Of Contents Report Guide Table Of Contents
Paired-End Runs Tab

Creating a Paired-end Composite Report This page describes how to combine the results from your completed forward and reverse paired-end runs into one composite report. By default, a composite report is generated for your paired-end runs. If your run has multiple forward and multiple reverse reports, you can use these steps to create a composite report which combines your choice of specific forward and reverse reports. From the Runs Tab in the Torrent Browser, the forward and reverse runs are listed separately. To create a composite report, you must click the Analysis button that is associated with the forward run.
In the Run Analysis workspace, enter the Report Name for the new composite report and select the Paired-End Forward run and Paired-End Reverse run from the drop-down menus. (You must use your own convention to keep track of the matching forward and reverse runs. Either use related run names, or the Notes field, or both.)
Once you have defined the forward and reverse runs, click the Start Analysis button in the lower right corner of the workspace. Notice that in the Runs tab, your composite report run appears under the forward run as Started.
Version 2.2
When your composite report is complete, click the hyperlink for the completed run as listed under Reports, to open the full report page under the Reports Tab. See also: Paired-End Reports Tab Paired-End Planning Tab Working with Paired-End Data User Interface Guide Table Of Contents Report Guide Table Of Contents Reanalyze a Run as Paired-End Because of differences in adapters, paired-end data must always be analyzed as paired-end, even if you only analyze the data in one direction. This section describes how to take a run that was incorrectly set up and analyzes as a regular forward run. Follow these steps to reanalyze as paired-end: 1. Find the run in the Runs tab, and click the edit link:
2. The Edit Run page opens. You fill out this page the same as the Planning tab Edit Plan page. Scroll down to the "Is this plan intended for a paired-end run?" checkbox, and enable the checkbox:
3. Select the Forward Library Key and the Forward 3' Adapter. The library key is typically the same in a regular forward run and a paired-end forward run. The adapter is different in a paired-end run. 4. Click Save Run. 5.
Version 2.2
4.
5. Back in the Runs tab, click Analyze for run you edited.
Reanalyze a Paired-End Run This section describes how to redo a paired-end analysis, for instance to rerun with different advanced arguments. 1. Find the run in the Runs tab, and click the Analyze button. The following page opens:
2. Click Advanced. (The Report Name and Paired-End Forward and Reverse menus are for composite reports, not for re-analyzing.) The Advanced arguments page opens:
Version 2.2
3. See Working with Runs for a description of the advanced arguments. Change the run arguments as required, and click Start Analysis.
Version 2.2
Use Cases
Use Cases
Introduction
This document describes several common use cases. These include topics whose operation or application: May not be intuitive from the Torrent Browser UI. May cross UI functional boundaries. May have alternative command line interfaces. Realign Run to Different Reference Genome Reanalyzing Run with a Different Barcode Set Using Barcodes with the PGM Scanning Sequencing Kits Handling a Failed Analysis Run Terminating an Analysis Run Working with Files Working with the Database Contents Introduction Realign Run to Different Reference Genome Reanalyzing Run with a Different Barcode Set Using Barcodes with the PGM Scanning Sequencing Kits Handling a Failed Analysis Run Determining the Fault Cause Restarting a Run Terminating an Analysis Run Working with Files Working with the Database Changing the Report Name Changing the Sample Name Changing the Run Date Adding or Changing a PGM Changing Your Torrent Browser Password

Use Cases
Version 2.2
Use the following procedure to terminate an analysis job for a run that has started but not completed: 1. Click the Services tab. 2. In the Active Jobs panel, find the run Name you want to terminate and click the Terminate button associated with the job (the Status Message column indicates job is running).
3. In the confirmation dialog, click Terminate to end the run or click Cancel to let the analysis job continue:
4. The dialog displays a termination confirmation message. Click the X in the upper-right corner to complete your termination session.
The run is removed from the Active Jobs list, which displays No active jobs if no other runs are active:
On the Reports tab, the deleted report shows a Terminated status.
Version 2.2
You can always start a new analysis run. Contents Introduction Realign Run to Different Reference Genome Reanalyzing Run with a Different Barcode Set Using Barcodes with the PGM Scanning Sequencing Kits Handling a Failed Analysis Run Determining the Fault Cause Restarting a Run Terminating an Analysis Run Working with Files Working with the Database Changing the Report Name Changing the Sample Name Changing the Run Date Adding or Changing a PGM Changing Your Torrent Browser Password
Handling a Failed Analysis Run

Use Cases
Handling a Failed Analysis Run
If an analysis run fails, you need to attempt to determine the cause of the failure an, possibly, restart the run. Determining the Fault Cause Restarting a Run
Version 2.2
Contents Introduction Realign Run to Different Reference Genome Reanalyzing Run with a Different Barcode Set Using Barcodes with the PGM Scanning Sequencing Kits Handling a Failed Analysis Run Determining the Fault Cause Restarting a Run Terminating an Analysis Run Working with Files Working with the Database Changing the Report Name Changing the Sample Name Changing the Run Date Adding or Changing a PGM Changing Your Torrent Browser Password
Determining the Fault Cause

Use Cases
Determining the Fault Cause If an analysis run fails, make the following checks: 1. Has the PGM sequencer completely transferred the data for the run? Go to the PGM sequencer Data Management screen to verify complete data transfer. You can re-transfer the data if you are not sure it was completely transmitted. 2. Click the Torrent Browser Runs tab and see that the file transfer was complete. Also, check if there are any error messages, such as User Aborted. 3. Click the Reports tab and look for any error messages. 4. If the report was generated, check if there are any messages on the report, itself. 5. There is a link at the bottom of the report for the Analysis Log File. You can download the file and review it, or send it to your Ion Torrent contact for review. 6. If you cannot determine the cause of the fault, try restarting the run.
Version 2.2
Restarting a Run
Use Cases
Restarting a Run This section describe how to restart a run that does not analyze paired-end data. To restart a paired-end run, see the following pages: Paired-End Runs Tab Working with Paired-End Data To restart an analysis run: 1. Click the Runs tab to access run management functions. 2. Next to your experiment name, click the Analyze button:
3. The Start Analysis window opens:
Version 2.2
3.
Do not modify the Report Name or the Paired-End fields these are used to generate a composite run report from a related forward run and reverse run. 4. Modify the Advanced options, if needed. See Working with Runs for information on Advanced fields.
Version 2.2
5. Click the Start Analysis button. 6. Verify that the run has started by clicking the Services tab to view your job
Version 2.2
Use Cases Table of Contents
Contents
Introduction Realign Run to Different Reference Genome Reanalyzing Run with a Different Barcode Set Using Barcodes with the PGM Scanning Sequencing Kits Handling a Failed Analysis Run Determining the Fault Cause Restarting a Run Terminating an Analysis Run Working with Files Working with the Database Changing the Report Name Changing the Sample Name Changing the Run Date Adding or Changing a PGM Changing Your Torrent Browser Password
Version 2.2
Realign Run to Different Reference Genome

Use Cases
Realign Run to Different Reference Genome
You may have one of the following reasons for wanting to choose a different reference genome: 1. Log in to Torrent Browser and click the Run tab. 2. In the Reference column for the experiment you want to rerun, click the reference short name:
3. In the pop-up window, click the new reference name:
Version 2.2
A Making change notification displays while the change is in progress. The Reference name is updated for the selected run:
Working with Files

Use Cases
Working with Files
Analysis Results File Location For a standard Torrent Server configuration, analysis results files are located in the following directories: Type of Data Directory Name
Version 2.2
Raw Processed
/results/<PGM_name>/<Run_name>/ /results/analysis/output/Home/<Report_na me>/
Standard Reference File Location Standard reference files are stored in the following location:
/results/referenceLibrary/<index_type>/<genome_shortname>/
Understanding Log Files in the /results Folder Many log files, shown in the following table, are generated for different parts of the Analysis pipeline. Some files only appear when a problem occurs. You do not need to login to see these files. Opening a report and removing the report name gives you a directory listing of all of the files, which you can open directly as text files. Be careful that you do not open a large file, such as a SFF file, using the web browser. Filename version.txt Description Lists the versions of the Ion software packages that were installed at the time the report was generated and the host name of the server. This information is also displayed on the default report. Lists all of the Test Fragment Templates that were used for generating this report. If the file size is zero and there are no data in the file, either no templates are installed or none are flagged isofficial. Analysis only checks against the templates that are marked isofficial, which is set using the Template s tab in the browser. Lists problems uploading data to the database. If there were no errors, this file has a file size of zero. If the file contains data, a problem occurred uploading data to the database. If analysis results are not being displayed in the browser, check this file. Analysis run status. If the analysis completed successfully, this file has a value of zero. A value of -1 indicates a failure occurred, requiring that you check other log files to determine the cause. No specific error information is provided in this file.
DefaultTFs.conf
uploadError
status.txt
Version 2.2
processParameters.txt
Run events and duration. The command-line passed to the Analysis program is also included, which is useful if you want to re-run the same analysis. The same data as the log file shown in the browser, without HTML formatting. Always check for errors in this file, especially the first and the last pages. Post-Analysis events. Error messages related to processes called after the primary analysis. This has a value of zero if the analysis completed successfully. Useful troubleshooting information generated during the alignment process. This file is only created when there is a problem. A memory dump listing, usually caused by a critical fault. You should see a related "exception" or "core dump" message in the ReportLog.html file. Errors related to TMAP. If the file is not present, it is likely that TMAP was not called.
ReportLog.html
drmaa_stdout.txt drmaa_stderr.txt
analyzeReads_err.txt
core
alignmentQC_out.txt
Working with the Database

Version 2.2
Use Cases
Working with the Database
User configuration information and experiment descriptions can be accessed and modified directly in the database, provided you have administrator privileges. Warning! Care must be taken when modifying database items. Setting fields to incorrect values may corrupt the database or produce unpredictable results.
Accessing the Database For all of the common database operations, you must begin by logging in to the database interface: Making changes to the database requires Torrent Server administrator-level user access. 1. In the Torrent Browser Config tab, click Admin Interface to access database administration functions:
Version 2.2
2. Enter the administrator Username and Password, and click Log In:
Common Database Operations Changing the Report Name Changing the Sample Name Realign Run to Different Reference Genome Changing the Run Date Adding or Changing a PGM
Version 2.2
Changing the Report Name

Use Cases
Changing the Report Name If you manually started an analysis and realize that you typed the report name incorrectly, you can change the report name using the following procedure. These steps require admin login. It is not safe to change the report name while the report is being processed. 1. Select the Results dialog.
Version 2.2
2. On the Select results to change page, select the name of the run you want to change, in the ResultsName column:
Version 2.2
3. Enter the new report name in the ResultsName field:
Version 2.2
4. Click Save to save your change. Contents Introduction Realign Run to Different Reference Genome Reanalyzing Run with a Different Barcode Set Using Barcodes with the PGM Scanning Sequencing Kits Handling a Failed Analysis Run Determining the Fault Cause Restarting a Run Terminating an Analysis Run Working with Files Working with the Database Changing the Report Name Changing the Sample Name Changing the Run Date Adding or Changing a PGM Changing Your Torrent Browser Password
Changing the Sample Name

Version 2.2
Use Cases
Changing the Sample Name When you start to process a chip on the PGM sequencer, one of the things you must enter is a sample name. The sample name can be any text that helps you to remember your data. When the PGM sequencer run completes, the data are transferred to the server for analysis and the sample name is displayed in the Torrent Browser, as shown in this example:
It may sometimes be necessary to change the sample name because of one of the following conditions: The PGM user entered the wrong information or a typo. The format for tracking sample names changed. You decided another name is more useful. Use the following procedure to change the sample name: 1. Select the Experiments option.
Version 2.2
2. Scroll down the list of experiments and click to select the experiment that has the sample name that you want to change. 3. Scroll to the Sample field for the selected experiment and change the contents of that field, as wanted. Warning Again, do not change any other fields or you risk corrupting the database. If you think you may have accidently modified another field, click Back in your browser and do not press S ave.
Version 2.2
4. When you are done changing the Sample name, click the Save button in the lower-right corner. 5. Go back to the Runs view of the Torrent Browser, and you should immediately see that the sample name is changed.
Version 2.2
Changing the Run Date

Use Cases
Changing the Run Date Occasionally, the PGM sequencer cannot get a date/time from the internet time server. When this occurs, the internal PGM sequencer date is set to January 1, 1969. The date of the run is encoded in the folder name, which is parsed and used as the Run Date in the database. This causes the new run to be displayed with the incorrect date. With a date of January 1, 1969, the run is the last item on the last page of the Runs tab list. Use the following procedure to change the date for this run: 1. In the Torrent Browser Config tab, click Admin Interface and login, if prompted. 2. Click to open the Experiments database item for modification:
Version 2.2
3. Find your run in the experiment name list. The list is sorted by date, starting with the newest runs in the database. Because the run from 1969 is at or near the end of the list, it is convenient to re-sort by date, in ascending order (oldest at top). Re-sort by clicking the Date column heading:
Version 2.2
4. Click the ExpName for your run to select it and display the following run information:
5.
Version 2.2
5. Use one of the following two options to change the date: a) Click the Today and Now buttons to set the Date and Time values to the current date and time in one click.
The automatic method is recommended because it places this run at the top of the Runs li st, of the Runs tab.
b) Manually edit the date/time strings. 6. Click Save, on the bottom right to save the new date:
7. Return to the Runs tab when done.
Version 2.2
Adding or Changing a PGM

Use Cases
Adding or Changing a PGM Do not add or change a PGM through the Site administration Rigs configuration item. During initial setup, PGM sequencers automatically register themselves with the Torrent Browser. Changes to a PGM are also communicated directly to the Torrent Browser. Never delete a Rig using the Site administration Rigs configuration item. That action could corrupt the database.
Here is the Site administration Rigs configuration item that is not to be used:
Version 2.2
Refer to Managing PGM Settings from Torrent Browser for configuring the PGM from Torrent Browser.
Version 2.2
Changing Your Torrent Browser Password

Use Cases
Changing Your Torrent Browser Password It is possible for a user to change the UI password and lock themselves out of the Torrent Browser Admin screen, or locked out of the Torrent Browser UI altogether. The UI password is stored in a database field, so if you have access to the database, you can change the password. Method One The first way change changing a username with minimal terminal interaction is to create a new super user account. This can be done be running the following commands
cd /opt/ion/iondb ./manage.py createsuperuser
Once the new superuser account has been created login to the admin page with the newly created username and password. Select the user section under Auth:
Version 2.2
Select the account you want to change the password for
Click "change password form"
Enter the new desired password and click "Change Password"
method Two
Step 1: Login to the database
Version 2.2
Log into the torrent server and get an interactive postgres database command prompt
ionadmin@my-torrent-server:~$ psql -U ion -d iondb psql (8.4.5) Type "help" for help. iondb=>
Step 2: Display the user list
In our example, the user "ionadmin" forgot the password, but we know the "ionuser" password. This command provides a list of users and passwords
iondb=> SELECT username, password from auth_user; username | password ----------+----------------------------------------------------ionuser | sha1$7e254$476582a5fa365cdd6081a80ac161c1904cc9c374 ionadmin | sha1$93099$b7da0df453d8db1c7715cabef9651c73003de849 ion | sha1$7798b$c025c463682f84b66cf3b5168356a04e3ce3b899 (3 rows)
Step 3: Copy the password from another user
The passwords are hashed in the database, so we don't know what the actual password is. But we know the "ionuser" password is "ionuser" so we can copy that hashed password to "ionadmin", and that will change the "ionadmin" password to "ionuser" WARNING The UPDATE command modifies the database. Do not proceed with this step if you are not comfortable with SQL commands
iondb=> UPDATE auth_user set password='sha1$7e254$476582a5fa365cdd6081a80ac161c1904cc9c374' where username='ionadmin';
Step 4: Check that the password has been changed
Query the database one again to verify that the password has been changed. See that ionadmin and ionuser now have the same password
Version 2.2
iondb=> SELECT username, password from auth_user; username | password ----------+----------------------------------------------------ionuser | sha1$7e254$476582a5fa365cdd6081a80ac161c1904cc9c374 ionadmin | sha1$7e254$476582a5fa365cdd6081a80ac161c1904cc9c374 ion | sha1$7798b$c025c463682f84b66cf3b5168356a04e3ce3b899 (3 rows)
Step 5: Reset the password
Now you can log in via the UI as ionadmin, and reset the password. Remember to change the password via the "change password" form. Contents Introduction Realign Run to Different Reference Genome Reanalyzing Run with a Different Barcode Set Using Barcodes with the PGM Scanning Sequencing Kits Handling a Failed Analysis Run Determining the Fault Cause Restarting a Run Terminating an Analysis Run Working with Files Working with the Database Changing the Report Name Changing the Sample Name Changing the Run Date Adding or Changing a PGM Changing Your Torrent Browser Password
Using Barcodes with the PGM

Use Cases
Using Barcodes with the Ion PGM Sequencer

Overview
The Torrent Suite supports barcoded runs, which allow you to process multiple barcoded samples in a single run on the Ion PGM sequencer.
Version 2.2
Torrent Server comes pre-installed with 2 barcode sets, ionSet1 and ionXpress. These barcode sets are available for use on the Ion PGM sequencer. A barcode run on the Ion PGM requires a sample-prep kit such as the IonSet1 or Ion Xpress barcode adapter kits. You select a barcode adapter kit when you set up your Ion PGM run. The barcode sequences for the IonSet1 and Ion Xpress barcode adapter kits are included with the Torrent Server software, and are called the IonSet1 barcode set and the IonXpress barcode set. This barcode set information is used during analysis to separate out reads by barcode, to remove the barcode and adapters from the read, and to output reads by barcode into FASTQ files. Reads are aligned against the reference genome, and the results stored in BAM and BAM index (BAI) files for each barcode. Reads that could not be classified as being one of the barcodes in the designated set are grouped into a "no-match" group, and alignment against the reference also performed on this group. The new barcode results files are availabe in the run report File Links section. Alignment metrics for each barcode are available as a CSV link from within the Report page for the given run. The Report default summary page shows Q20 performance metrics for all barcodes in the set, providing a quick glance at the high level quality of each barcode, and includes the total number of reads found per barcode, the mean AQ20 alignment length, the total number of AQ20 bases, and the total AQ20 reads per barcode. You can add additional barcode sets using the Torrent Browser References tab. See Managing Barcodes and Barcode Sets, which describes how to do the following: View barcode sequences. Add and delete your own barcode set. Add and delete individual barcodes of your barcode sets. Workflow The standard workflow for a barcoded sample is similar to a normal Ion PGM run and analysis. This section provides an overview of the workflow, with the new steps involved on a barcode run.
Summary of the Recommended Workflow
Here is an overview of the recommended workflow for a barcode run. Screenshots and more details are provided below. 1. Set up your run in the Planning tab, with the Add Plan button. Select one of the available barcode sets from the drop-down list for the Barcode selection, and fill out the other run information (the same information as entered on the Ion PGM Run Info screen). 2. The Torrent Server assigns a name to your pending run, and generates a retail-style scannable barcode image representing your pending run name. Your run information is stored on the Torrent Server as a pending run until you are ready to start the run on the Ion PGM sequencer. 3. When you are ready to start the run, on the Ion PGM Run Info screen you select your run from a list of pending runs. The Torrent Server populates the Ion PGM Run Info with the information you entered in the Planning tab. (You may optionally change information on the Run Info screen.) 4. You start the Ion PGM run as usual. 5. When the run and report are complete, you can review the performance of the barcoded reads in the default
Version 2.2
5. report page. The following additional barcode-specific files are available for download from the File Links download section: A zip of SFF files for each barcode A zip of FASTQ files for each barcode A zip of BAM and BAM index (BAI) files for each barcode A csv-style spreadsheet summarizing the barcode performance for each barcode
Run Planning - Set Up Your Barcode Run in the Planning Tab
Follow these steps to set up your run information as a pending run: 1. Click the Planning tab, and under Pending Runs, click the Add Plan button on the left, to open the Add new plan screen. 2. Fill out the run information just as you would on the PGM Run Info screen. An example completed Add new plan screen is shown below. Only the Plan name is required. However, you are recommended to enter all fields here on the Torrent Browser. You may still enter any remaining fields on the PGM sequencer when you start the run.
3. When your run information is complete, click the Save Plan button (on the bottom right). Your pending run
Version 2.2
3. appears in the Planning tab:
The Torrent Server assigns a short code name to your pending run. The example short code here is 67HYE. 4. (Optional) Click the Barcode button next to the Short Code column to display a scannable barcode image. The Torrent Server generates this retail-style barcode image from your short code run name.
You can print this barcode image and scan it at the Ion PGM to import your pending run information into the Ion PGM Run Info screen. Click the Barcode button again to return to the short code display. See the User Interface Guide Planning Tab page for more information about creating pending runs.
Start Your Pending Run on the Ion PGM sequencer
This section describes how to go from a pending run to an actual run on the Ion PGM sequencer. You must first create a pending run, as described in Run Planning - Set Up Your Barcode Run in the Planning Tab , before using the instructions in this section. 1. Open the Run Info screen on the Ion PGM sequencer. 2. Click on the Browse button (near the middle of the screen, to the right of the Pending Run field).
3. The Select Pending Runs pop-up opens with a list of available Pending Runs. Your pending run is identified by short code and plan name (as listed under the Planning tab). Select your run and click OK. (This example shows only one pending run. You may have multiple runs in the Select Pending Run window.)
Version 2.2
Your selection appears in the Pending Run field:
The Ion PGM Run Info fields, including your barcode set, are populated with information from your pending run.
If required, you can update any Run Info fields now. 4. Click Next --> to start your Ion PGM run, as usual. Approve your run on the confirmation screen.
Version 2.2
When you accept the confirmation screen, your pending run information is deleted from the Planning tab. If you terminate your Ion PGM run and at a later time want to start the run, you must either enter the run information on the Ion PGM Run Info screen or re-create the pending run again under the Torrent Browser Planning tab. The new pending run has a different short code and scannable barcode image.
Other Methods to Import Your Pending Run
This section describes the ways to import your pending run information into the Ion PGM Run Info screen. These are all done on the Ion PGM Run Info screen, and are all different ways to populate the PGM Run Info screen with the run information previously entered in the Planning tab. Choose the method which best fits your work environment.
Pending Run Browse Button
The pending run Browse button is described in #Start Your Pending Run on the Ion PGM Sequencer: click Brow se and select your pending run from the pop-up list.
Pending Run Short Code
You can type the short code for your pending run into the Pending Run: text field. An example short code is 67HYE.
A short code is assigned to your pending run when you enter the run information in the Planning tab (as is listed in the Planning tab).
Pending Run Scannable Barcode Image
In the Planning tab, click the Barcode button next to the Short Code column to display a scannable barcode image. The column display switches from short code to barcode image.
The Torrent Server generates the retail-style barcode image as a representation of your short code run name.
Version 2.2
Print this barcode image and scan it at the Ion PGM to import your pending run information into the PGM Run Info screen.
Barcode Reports and Output Files
This section describes new output and reports for barcode runs. The Barcode Reports section of the Torrent Browser Default Report shows key performance metrics for each barcode in the run. The number of entries in the x-axis change depending on the number of barcodes in your run. Barcodes with small numbers of reads (<1% reads of the next largest barcode) do not appear in the report when barcode filtering is enabled. The category named "X" contains barcodes that could not be matched to known members of the barcode set being used. Hover our cursor over a barcode to see a tool tip of the barcode ID (Index ID), the y-axis value for the barcode, and the barcode sequence.
Version 2.2
The File Links section of the Torrent Browser Default Report includes barcode-related results files available for download:
Version 2.2
File Type Barcode Alignments Summary
Description A summary file of alignment metrics for barcodes. The metrics include the quality and read lengths at which each barcode aligns to reference. Binary Sequence Alignment/Map (BAM), is a compressed, binary form of the SAM file. The BAM index (BAI) file speeds up the access time for a coordinate-sorted BAM file. The BAM and BAI files for each barcode are zipped together. Compressed (.zip) FASTQ-formatted file containing data organized in a per-base basis, including quality scores. The reads contained in the file are unaligned reads. The FASTQ files for each barcode are zipped together. Compressed (.zip) Standard Flowgram Format (SFF) -formatted file containing "flow space" data. The SFF files for each barcode are zipped together.
Barcode-specific Library Alignments (BAM and BAM Index)
Barcode-specific Library Sequence (FASTQ)
Barcode-specific Library Sequence (SFF)
Plugin Support for Barcodes
The following plugins supports barcode libraries: Coverage Analysis Torrent Variant Caller See Running the Installed Plugins for information on these plugins. The following plugin does not support barcode libraries:
Version 2.2
Alignment Related Information The following pages also discuss barcode usage, reports, and manipulation: Managing Barcodes and Barcode Sets describes how to manage individual barcodes and barcode sets using the Torrent Browser References tab. The User Interface Guide Planning Tab describes how to create a pending run. Running the Installed Plugins includes example barcode output for plugins. Contents Introduction Realign Run to Different Reference Genome Reanalyzing Run with a Different Barcode Set Using Barcodes with the PGM Scanning Sequencing Kits Handling a Failed Analysis Run Determining the Fault Cause Restarting a Run Terminating an Analysis Run Working with Files Working with the Database Changing the Report Name Changing the Sample Name Changing the Run Date Adding or Changing a PGM Changing Your Torrent Browser Password
Reanalyzing Run with a Different Barcode Set

Use Cases
Reanalyzing Run with a Different Barcode Set
If you choose the wrong barcode set or forget to select a barcode set during run planning, these instructions help you select the correct barcode set and reanalyze the run. Follow these steps to change the barcode ID: 1. Find the run in the Runs tab, and click the Edit link for that run:
Version 2.2
Choose the correct barcode set. If you have custom barcode sets, your list is different:
2. Click the Save Run button (in the bottom right corner). You can now reanalyze the run from the Runs Tab.
Version 2.2
Scanning Sequencing Kits
Use Cases
Scanning Sequencing Kits
Scanning sequencing kits is a new part of the Personal Genome Machine (PGM) setup workflow. The Torrent Browser run registration (under the Planning tab) also supports specifying the sequencing kit when you create a run plan. Scanning is done during the PGM system setup:
Version 2.2
Scanning the sequencing kit is preferred to selecting a checkbox, because scanning provides more detailed kit information that can be used for troubleshooting or other purposes. The choice of sequencing kit affects the nucleotide flows on the PGM system. The sequence kit barcode can entered in the Torrent Browser Planning tab during run registration. Both the Add a New Plan and Edit Plan screens support specifying the sequencing kit:
Version 2.2
Version 2.2
Torrent Suite Administrator Documentation
Version 2.2
Torrent Suite Administrator Documentation

Welcome to the Torrent Suite Administrator Documentation Home Page. Contents Torrent Server Administration Guide
Version 2.2
Torrent Server Administration Guide

Introduction
This document supports the deployment and maintenance of the Ion Torrent Server. The Ion Torrent Server is a dedicated server that performs DNA sequencing analysis of data received from the Ion Torrent Personal Genome Machine (PGM) sequencer.
Server Configurations
The Ion Torrent Server has the following hardware and software configuration: Feature CPU GPU RAM OS Network controllers Ion PGM Torrent Server Two Xeon 5650 6-core processors @ 2.66 GHz NVIDIA Graphical Processer Unit 48 GB Ubuntu Server 10.04 LTS Gigabit NIC 4 ports: - 1 internet - 3 for PGM Sequencers Eight 2 TB SATA drives; partitioned with RAID 5, for about 12 TB of usable storage Not included Dell/T7500 Precision Tower Not included (File access via SSH or web service) Torrent Suite 4462918
Hard Drives
Optical Drive Computer Chassis Keyboard, Mouse Application Software Life Technologies Part Number
For more information on Ubuntu technical specs, including version numbers of other associated software packages like Apache and PostgreSQL, please visit https://help.ubuntu.com/community/Server/TechSpecs/1004LTS.
Audience
Version 2.2
This document is intended for system and network administrators responsible for setting up and maintaining server hardware, software and network connectivity. By presenting typical use cases, we identify best-practice operational scenarios to ensure that your system installs correctly and continues to run smoothly.
Next Steps
Deploying Your System Maintaining Your System Troubleshooting You must use your ionadmin account for steps described in the Torrent Server Administration Guide. (Do not use your ionuser account.)
Version 2.2
Contents Torrent Server Administration Guide Deploying Your System Preparing Your Site Installing the Server Network Connectivity Updating Torrent Server Software Updating the IonReporterUploader Plugin Configuring the Torrent Server Verifying Functionality Managing PGM Settings from Torrent Browser Maintaining Your System Monitoring Free Disk Space Updating Reference Library Indexes Adding a New Genome Index Updating Test Fragment Templates Installing Analysis Plugins Archiving and Deleting Data Backing Up and Restoring Data Mounting a USB Drive Booting Into Single-User Mode Installing and Using a UPS Axeda Remote System Monitoring (RSM) Administration Configure Chips Configure Experiments Configure Global Configs Configure Users Management Actions View Network Settings Shutdown Server Update Server Update OneTouch Device Troubleshooting
Admin Table Of Contents
Version 2.2
Contents
Torrent Server Administration Guide Deploying Your System Preparing Your Site Installing the Server Network Connectivity Updating Torrent Server Software Updating the IonReporterUploader Plugin Configuring the Torrent Server Verifying Functionality Managing PGM Settings from Torrent Browser Maintaining Your System Monitoring Free Disk Space Updating Reference Library Indexes Adding a New Genome Index Updating Test Fragment Templates Installing Analysis Plugins Archiving and Deleting Data Backing Up and Restoring Data Mounting a USB Drive Booting Into Single-User Mode Installing and Using a UPS Axeda Remote System Monitoring (RSM) Administration Configure Chips Configure Experiments Configure Global Configs Configure Users Management Actions View Network Settings Shutdown Server Update Server Update OneTouch Device Troubleshooting
Deploying Your System

Version 2.2

Deploying Your System
This section covers initial Torrent Server installation procedures, from the time when your system is delivered to when you are able to run your first analysis. These activities include: Preparing Your Site Installing the Server Managing PGM Settings from Torrent Browser
You must use your ionadmin account for steps described in the Torrent Server Administration Guide. (Do not use your ionuser account.)
Version 2.2
Preparing Your Site

Version 2.2
Preparing Your Site

The Torrent Server requires internet connectivity to install software updates and patches from Ion Torrent. To facilitate deployment, it is important that you prepare your site network infrastructure in advance. On the scheduled deployment date, IT support personnel should be scheduled to be available to ensure that the Torrent Server is properly connected to the network and, if needed, assist the FAS in troubleshooting connectivity issues.
Review the following topics to prepare your site for server installation: Network Connectivity Preparation
Network Provisioning
Network provisioning should be completed before the scheduled installation: A network access jack should be identified and enabled in the area where the Torrent Server will be deployed. An Ethernet cable is provided with the server. The cable connects to the port labeled LAN on your Ion Torrent Server; the remaining ports are used to connect to the PGM sequencers. DNS should be configured in advance of the scheduled installation.
Port Assignment
To fully support the Torrent Server and PGM sequencers, remote monitoring must be provided using Axeda Remote System Monitoring software, enabled and able to reverse ssh into the boxes. This requires that the PGM sequencers and Torrent Servers be connected to the internet, on a private network behind a firewall, with outbound connections permitted on the following ports: Port 22 Required Yes Use Start reverse SSH tunnel for remote troubleshooting Download updates from http://updat es.iontorrent.com and http://us.archi ve.ubuntu.com (UDP) NTP access to the Internet; incoming and outgoing. Enable sending of basic status information to the remote monitoring server. Remote access to PostgreSQL database.
80
Yes
123
Yes
443
Yes
5432
No
Name Resolution
Version 2.2
A DHCP-assigned address is recommended. (The Torrent Server can be reconfigured to use a static IP address, if necessary.) Server run, report and configuration data are accessed using your workstation Web browser. You must be able to access the server UI by entering the server hostname in your browser URL address field. Network Security
Using a Firewall
The network to which the Torrent Server and PGM sequencers are connected must be secured. This typically involves placing the Torrent Server and PGM sequencers behind a stand alone firewall to prevent unauthorized access to the server or data.
You will need to add the Ion Torrent networks to the firewall rules to permit access to the Torrent Servers. Use your web browser to discover your external (public) IP address. There are several sites that can be used to determine your IP address, such as http://whatismyip.com. You may also use a text-based browser, such as Links, to find your external IP address. For Links, use the following command, which returns your external IP address:
links -dump http://whatismyip.com | grep "Your IP Address Is"
You may need to first install Links, using the apt-get install links command.
Securing Torrent Server and PGM Using iptables
The Torrent Server can further be secured by using iptables. (The administrator is assumed to be familiar with iptables and their use.) A basic iptables configuration is provided by default with the Torrent Server. Ensure the following Ion Torrent IP addresses are in the iptables.start file. For simplicity, add the entire address range as shown:
$IPTABLES -A INPUT -p tcp --dport 22 -s 38.110.159.160/27 -j ACCEPT $IPTABLES -A INPUT -p tcp --dport 22 -s 38.126.104.128/25 -j ACCEPT
Use the following procedure to configure the iptables for your location: 1. In the /etc directory, create iptables.start and iptables.stop files similar to the example files shown below. 2. Type the command sudo chmod +x iptables*. 3. Edit the rc.local file similar to the examples shown below (see ## ADD THE FOLLOWING LINE in the
Version 2.2
3. example). 4. Start iptables using the command: sudo /etc/iptables.start. PGM iptables.start
#!/bin/sh # iptables IonTorrent - PGM 1.0.0 # IPTABLES="/sbin/iptables_32" #$IPTABLES -A INPUT -m state --state ESTABLISHED,RELATED -j ACCEPT $IPTABLES -A INPUT -i lo -j ACCEPT $IPTABLES -A OUTPUT -o lo -j ACCEPT $IPTABLES -A INPUT -p icmp -j ACCEPT $IPTABLES -A INPUT -s 127.0.0.1 -j ACCEPT # PGM instrument's own IP address here $IPTABLES -A INPUT -s 128.114.63.84 -j ACCEPT $IPTABLES -F $IPTABLES -P INPUT ACCEPT $IPTABLES -P FORWARD ACCEPT # Allow ssh from these addresses: NEW IW, IE, ... elided ... $IPTABLES -A INPUT -p tcp --dport 22 -s 71.6.74.66 -j ACCEPT #$IPTABLES -A INPUT -p tcp --dport 22 -s 173.13.110.174 -j ACCEPT $IPTABLES -A INPUT -p tcp --dport 22 -s 173.13.110.163 -j ACCEPT #$IPTABLES -A INPUT -p tcp --dport 22 -s 173.13.146.13 -j ACCEPT $IPTABLES -A INPUT -p tcp --dport 22 -s 128.114.63.83 -j ACCEPT # Allow http connection to update $IPTABLES -A INPUT -p tcp --dport $IPTABLES -A INPUT -p tcp --dport # Test connection from ICE $IPTABLES -A INPUT -i eth0 -p udp $IPTABLES -A INPUT -p tcp --dport $IPTABLES -A INPUT -p tcp --dport server 80 -s 173.13.110.163 -j ACCEPT 80 -s 66.166.233.178 -j ACCEPT --dport 67:68 ... elided ... 80 -j DROP 22 -j DROP
iptables.stop
#!/bin/sh # iptables IonTorrent - PGM 1.0.0 # IPBLES="/sbin/iptables_32" $IPTABLES -F $IPTABLES -P INPUT ACCEPT $IPTABLES -P FORWARD ACCEPT
Version 2.2
rc.local
#!/bin/sh # # This script will be executed *after* all the other init scripts. # You can put your own initialization stuff in here if you don't # want to do the full Sys V style init stuff. # Create file's magic database file. if test -f /usr/share/misc/file/magic; then if test ! -f /usr/share/misc/file/magic.mgc; then file -C /usr/share/misc/file/magic.mgc fi fi touch /var/lock/subsys/local ## ADD THE FOLLOWING LINE /etc/iptables.start /software/patchcheck.sh /bin/settime /software/pciinit modprobe radeonfb fbset 800x600-75-32 /software/nanox/nano-X & /software/nanox/nanowm & sleep 0.5 cd /software/gui ./launcher & cd /software /software/datacollect & cd /software/gui/cntrl sleep 0.25 /software/gui/cntrl/Controller &
Torrent Server iptables.start
Version 2.2
#!/bin/sh IPTABLES="/sbin/iptables" $IPTABLES -F $IPTABLES -P INPUT DROP $IPTABLES -P FORWARD DROP $IPTABLES -A INPUT -m state --state ESTABLISHED,RELATED -j ACCEPT $IPTABLES -A INPUT -i lo -j ACCEPT $IPTABLES -A OUTPUT -o lo -j ACCEPT $IPTABLES -A INPUT -p icmp -j ACCEPT $IPTABLES -A INPUT -s 127.0.0.1 -j ACCEPT # Enter Server's IP address here $IPTABLES -A INPUT -s 128.114.63.83 -j ACCEPT # IP addresses to allow ssh access: NEW IE, IW, PGM, NEW IW $IPTABLES -A INPUT -p tcp --dport 22 -s 173.13.110.163 -j ACCEPT #$IPTABLES -A INPUT -p tcp --dport 22 -s 173.13.146.13 -j ACCEPT $IPTABLES -A INPUT -p tcp --dport 22 -s 128.114.63.83 -j ACCEPT $IPTABLES -A INPUT -p tcp --dport 22 -s 71.6.74.66 -j ACCEPT # allow all traffic from PGM # (FTP wants more than port 21 with extended response features) $IPTABLES -A INPUT -p tcp -s 128.114.63.84 -j ACCEPT $IPTABLES -A INPUT -p tcp --dport 80 -j ACCEPT $IPTABLES -A INPUT -p tcp --dport 443 -j ACCEPT $IPTABLES -A INPUT -i eth0 -p udp --dport 67:68 ... elided ... $IPTABLES -A INPUT -p tcp --dport 80 -j DROP $IPTABLES -A INPUT -p tcp --dport 22 -j DROP
iptables.stop
#!/bin/sh IPTABLES="/sbin/iptables" $IPTABLES -F $IPTABLES -P INPUT ACCEPT $IPTABLES -P FORWARD ACCEPT
rc.local
Version 2.2
#!/bin/sh -e # # rc.local # # This script is executed at the end of each multiuser runlevel. # Make sure that the script will "exit 0" on success or any other # value on error. # # In order to enable or disable this script ... elided ... # bits. # # By default this script does nothing. ## ADD THE FOLLOWING LINE /etc/iptables.start exit 0
Version 2.2
Installing the Server

Version 2.2
Installing the Server

Server installation should only be done by a Life Tech Field Service Engineer (FSE).
Server installation includes the following activities: Network Connectivity Configuring the Torrent Server Verifying Functionality Updating Torrent Server Software
Version 2.2
Network Connectivity
Version 2.2
Network Connectivity Use the following procedures to verify correct network configuration. If you are experiencing difficulty with network connectivity, refer to the Troubleshooting section for more detailed instructions.
Network Connectivity includes the following topics:

Testing Network Connectivity
1. The server is configured with DHCP. If DHCP is configured in the site network, the server should have already acquired an IP address. Verify that the server has acquired an IP address using the following command.
ifconfig eth0
2. To test connectivity for updates, type the following command:
links -dump http://updates.iontorrent.com/
This should return:
Welcome to the Ion Torrent Updates Server. -----------------------------------------------[ICO] Name Last modified Size Description ------------------------------------------------
followed by a few [DIR] lines. If not, check that the light is on, for the port connected to the network. Try pinging other on-site servers by name and IP address.
Using a Static IP Address
If DHCP is not available, configure the system for a static IP address. You will need root access on the Torrent Server, which user ionadmin has by default. The following procedure uses the nano text editor to open and modify the /etc/network/interfaces file. This file contains settings for the Torrent Server network interfaces. 1. Contact the site IT representative to get an available IP address, netmask and gateway. 2.
Version 2.2
2. Enter the command to edit the interfacesfile, using nano:
sudo nano /etc/network/interfaces
3. At the prompt, enter your password. 4. Look for the following lines:
# The primary network interface auto eth0 iface eth0 inet dhcp
5. Change this section to include your static IP, netmask and gateway (from step #1), as shown in the following example:
# The primary network interface auto eth0 iface eth0 inet static # set your static IP below address 192.168.1.23 # set your default gateway IP here gateway 192.168.1.1 # set your netmask here netmask 255.255.255.0 # set your DNS server here dns-nameservers 192.168.1.2
6. Exit the editor and apply your changes by restarting the network on the Torrent Server, using the following command:
sudo /etc/init.d/networking restart
7. Watch for error message after restarting the network. An ill-formed /etc/network/interfaceswill cause the server to be not bootable.
* Reconfiguring network interfaces... Don't seem to be have all the variables for eth0/inet. Failed to bring up eth0.
Version 2.2
It is not recommended to edit /etc/network/interfaces file remotely. Either log on to the Torrent Server with monitor and keyboard or ssh via Port 3 on Torrent Server.
Accessing the Web Browser
Use the following procedure to test Web browser access. To access the web browser, the site IT administrator may need to configure the hostname to point to the IP address of this server. Some networks may handle this automatically. 1. Go to another desktop or laptop on the network. 2. Open a browser and enter either http://ion-torrent-server or the new host name, as applicable. 3. If the server does not respond, try using http://<<ip-address>>. If you were able to open the browser, it indicates there is a name resolution issue on the network, which must be fixed. 4. If there is still no response, ping the hostname and the IP address of the server.
Version 2.2
Updating Torrent Server Software

Version 2.2
Updating Torrent Server Software After upgrading your Torrent Server software, go to the Config tab Plugin section and check the list of enabled plugins. You must manually enable these plugins, if required, under the Config tab Plugin section.
Visit the Management Actions page for instructions on updating your Torrent Server. Warning To get the full instructions for updating your Torrent Server, please refer to the latest Release Notes on the Ion Community. There may be additional steps and procedures required, depending on the type of software upgrade. For Ion Reporter users, if the Ion Reporter server is not also being updated, refer to Updatin g the IonReporterUploader Plugin.
Version 2.2
Configuring the Torrent Server

Version 2.2
Configuring the Torrent Server Configure your server only after updating the server software.
The server should work out-of-the-box. Server installation includes the following topics:
Changing the Hostname
Use the following commands to change the hostname:
sudo TSconfig --change-hostname
The server will need reboot after the hostname is changed, so the above commands will automatically reboot the server.
Adding an HTTP Proxy
Use the following commands to add an HTTP proxy:
sudo TSconfig --configure-proxy
Set the proxy address and authentication according to the following prompts: 1. Enter http proxy address: Enter the proxy address. (If no address is entered, you are prompted to exit the program.) 2. Enter http proxy port number [3128]: Enter a port number or carriage return to accept the default, 3128, port number. 3. Enter the username for proxy authentication: Enter a username. If you do not enter a username, no authentication is set. 4. Enter the password for proxy authentication: Enter a password. If you do not enter a password, no authentication is set. A proxy address confirmation message is displayed:
http_proxy is set to http://username:password@proxyAddress
Changing the Timezone

Version 2.2
Use the following commands to change the timezone:
sudo TSconfig --configure-timezone
Configuring Postfix for Nightly Email
#Preparing Site IT #Adding Email Recipients #Adding Web Root and Site Name to Admin and Global Config Scheduling the Mail Task (cron job) #Configuring Postfix
Preparing Site IT
It is useful to confirm the following with the site IT administrator: Outbound email from the server is allowed. Many sites block the ability for machines to send an e-mail message directly to a mail server because this is a common avenue for virus propagation. If the site has such a restriction in place, the IT administrator may need to make an exception for this server. Ideally, the domain is configured with a default "MX record." The "MX record" is the mail exchange information that automatically knows how to handle any e-mail generated by the Torrent Server. If this is not in place, the nightly e-mail might not work.
Adding Email Recipients
Use the Torrent Browser Config tab to enter destination e-mail addresses. You are cautioned to only add a couple of recipients while testing.
Adding Web Root and Site Name to Admin and Global Config
Make certain these fields are configured in the database so they will show up correctly on the e-mail message: 1. Click the Torrent Browser Config tab. 2. Click Admin Interface and login. 3. Select Global configs. 4. Make certain a value is entered in the Web root field and the Site name field, as shown in the example:
The value for Web root should be the main URL that people use to access this Torrent Browser from the desktop.
Version 2.2
Scheduling the Mail Task (cron job)
The cron program is the Linux program that automates tasks to run on a schedule. By default, a cron job is installed in the Torrent Server to run at 6AM, daily. You can verify this by logging in as ionadmin and typing the command: crontab -l.
ionadmin@kauai:~$ crontab -l 0 6 * * * /opt/ion/iondb/bin/runnightly.sh
Notice, all cron calls the runnightly.sh script at 6AM. As a test, without having to wait until 6AM, you can invoke the same script manually from the command-line and see if it generates an e-mail:
/opt/ion/iondb/bin/runnightly.sh
If you forgot to enter the Web root setting, you will an error similar to the following:
ionadmin@kauai:~$ /opt/ion/iondb/bin/runnightly.sh Traceback (most recent call last): File "nightly.py", line 70, in <module> main(sys.argv) File "nightly.py", line 67, in main send_nightly() File "nightly.py", line 49, in send_nightly if web_root[-1] == '/': IndexError: string index out of range
Configuring Postfix
Postfix is a Linux mail server ("post office") program. It is open-source software and was not designed by Ion Torrent. It is more like Exchange than Outlook in that it is an e-mail server, not an e-mail client. Postfix is designed to work on various infrastructures, so there are many configuration options for IT administrators to adjust mail routing parameters. You can find the official Postfix documentation here: http://www.postfix.org/documentation.html. There is no way to know it advance what Postfix configuration settings will work for a particular site. The following example procedure shows basic configuration settings that have worked for the servers configured to date. There is no guarantee this will work for every situation. If email is still not appearing, you will almost certainly require assistance from the site IT administrator, who may or may not have experience with Postfix. This configuration can be re-executed at anytime: 1. Go to the Linux command line and start the Postfix configuration utility:
Version 1. 2.2
sudo TSconfig --configure-postfix
2. Select the type of mail configuration. Typically, Internet Site.
3. For System mail name, enter the site domain name. This domain needs to be configured with a default "MX" (mail exchange) record.
Version 2.2
4. Leave mail recipient blank.
5. Leave other destinations blank.
Version 2.2
6. Set Force synchronous updates on mail queue to <No>.
7. For Local networks, accept the default, and select <OK>.
Version 2.2
8. For Mailbox size limit, it is recommended that you not have no limit, so enter "51200000" to not consume all disk space.
9. Leave Local address extension character blank.
9.
Version 2.2
10. For Internet protocols to use, accept the default and select <OK>.
11. After the last screen, the console returns to text mode and displays the status of the Postfix configuration files update.
Version 2.2
Running newaliases * Stopping Postfix Mail Transport Agent postfix [ OK ] * Starting Postfix Mail Transport Agent postfix [ OK ] root@kauai:~# sudo dpkg-reconfigure postfix * Stopping Postfix Mail Transport Agent postfix [ OK ] setting synchronous mail queue updates: false setting myorigin setting destinations: setting relayhost: setting mynetworks: 127.0.0.0/8 []/104 []/128 setting mailbox_size_limit: 51200000 setting recipient_delimiter: setting inet_interfaces: all setting inet_protocols: ipv4 WARNING: /etc/aliases exists, but does not have a root alias. Postfix is now set up with the changes above. If you need to make changes, edit /etc/postfix/main.cf (and others) as needed. To view Postfix configuration values, see postconf(1). After modifying main.cf, be sure to run '/etc/init.d/postfix reload'. Running newaliases * Stopping Postfix Mail Transport Agent postfix * Starting Postfix Mail Transport Agent postfix
[ OK ] [ OK ]
This completes Postfix configuration.
Version 2.2
Verifying Functionality
Version 2.2
Verifying Functionality This procedure verifies the server installation and configuration are complete, and the server is ready to run analysis programs: 1. Connect to your Torrent Server, using ssh, and verify that the Crawler and Job Server services are running:
ps -aux | grep py
This should show active crawler.py and serve.py processes. 2. Run a test analysis of the provided cropped data set and review the resulting report.
Version 2.2
Managing PGM Settings from Torrent Browser

Version 2.2
Managing PGM Settings from Torrent Browser

From Torrent Browser, users can remotely: Set FTP information Set Update server address Monitor the state of PGM, such as Idle or In experiment View the most recent alarms Get the software versions Connecting PGM to the Torrent Server On the Advanced screen, you can set Torrent Server login information, e.g. server address ( Torrent Server), username (TS UserName), and password (TS Passwd), to connect to the Torrent Server. The Torrent Server field turns green to indicate that the login information is correct. PGM uses the Torrent Browser API to communicate with the Torrent Server. The username and password are the ones used to log on to Torrent Browser. Torrent Server SSH login can be different from Torrent Browser login.
Connecting for the First Time
If this is first time that this PGM is connected to the Torrent Server, the PGM will get the FTP information and update server information from server. Any subsequent update to these fields, such as FTP server, is sent back to the Torrent Browser. If the Torrent Server has never seen this PGM before, then the PGM name is not listed under Rigs, and the Torrent Browser copies the settings from a template or the model PGM name default. You can modify this model PGM to suite your site deployment, but you should not delete the model PGM named default.
Version 2.2
Once the PGM is Connected
As mentioned before, any subsequent updates to the fields will also be updated in Torrent Browser by the PGM. Changing the name of the PGM (PGM Name) has a different implication. It creates a new rig with the information for the new PGM, rather than for the model PGM. Updating Information on PGM via Torrent Browser PGM information is stored inside Rigs under Site administration. First click on Config tab; under Database Administration section, click on Admin Interface. It prompts for user name and password. The user needs to have staff status and the permission to update Rigs (rundb | rigs). By default, ionadmin has such permission.
FTP Information
The FTP information on Torrent Browser can be updated anytime whether or not there is the experiment going on the PGM. The FTP information is sent to the PGM and is effective immediately. To change the FTP information, go to Rigs under Site administration and click on the name of the PGM. Field Name Ftpserver Ftpusername Ftppassword Ftprootdir Meaning The address of FTP server, such as 192.168.201.1 User name for login Password for login The directory path (on the FTP server) to the folder where raw data are stored
Update Home
Update home is the URL to the PGM update server. By default it is set to the Torrent Server, which acts as the PGM update server. To change the update home URL, go to Rigs under Site administration and click on the name of the PGM.
Version 2.2
Field Name Updatehome
Meaning The URL of the PGM update server
Reference Library and Barcode
On the PGM, during a run, user can enter information about the experiment, or run, on Run Info screen. PGM gets the lists of reference library and barcode set from the Torrent Browser. The information are queried in real time. For example, while at this Run Info screen on the PGM, you realize the reference library has not been added on Torrent Browser. You can go to Reference tab on Torrent Browser and add a new reference library. Back at the PGM, you see the new reference library when pressing the drop-down menu (circled in red below)
Viewing PGM Status and Information You can view the status of PGM in the State field. It can be Idle, Initing, or In Experiment. You also can view the recent alarm such "No connectivity to ftp server". You can find out the software versions in the Version field.
Version 2.2
Version 2.2
Maintaining Your System

Version 2.2
Maintaining Your System

Perform the following tasks as needed to maintain your system in good working order: Monitoring Free Disk Space Updating Reference Library Indexes Adding a New Genome Index Updating Test Fragment Templates Installing Analysis Plugins Archiving and Deleting Data Backing Up and Restoring Data Mounting a USB Drive Booting Into Single-User Mode Installing and Using a UPS Axeda Remote System Monitoring (RSM) Administration You must use your ionadmin account for steps described in the Torrent Server Administration Guide. (Do not use your ionuser account.)
Version 2.2
Updating Test Fragment Templates

Version 2.2
Updating Test Fragment Templates As part of a protocol update, Ion Torrent provides new Test Fragments (TF). Currently, the only way to enter or modify TF Templates is by using the UI Templates tab dialog. Carefully cut-and-paste the data string into the input field, as shown in the following figure:
Make sure that the Test Fragment sequence contains only uppercase letters, and includes only the letters A, T, C and G. If there is an invalid character or duplicate Test Fragment, you will not be permitted to save the changes.
Contact Ion Torrent if you have questions about the Test Fragment Templates that are installed in your Torrent Browser.
Version 2.2
Updating Reference Library Indexes

Version 2.2
Updating Reference Library Indexes
When you upgrade your Torrent Suite software to version 2.2, you must rebuild your reference indices. This process can take a few hours for larger reference genomes. Your users should not submit data analysis jobs while the reference indices are being rebuilt.
Follow these steps to rebuild your reference genome indices: 1. Log in with an ionadmin account. 2. Ensure users do not submit analyses while the rebuild is in progress. 3. Go to the References tab. 4. Click the Rebuild All Now button. 5. When the rebuild is done, the Index Version column shows the new TMAP index version, tmap-f3.
Version 2.2
Adding a New Genome Index

Version 2.2
The Torrent Browser interface is the recommended method for creating new genome indexes. Torrent Browser reference library interface described in Working with Reference Sequences. The command line instructions in this document should only be used by experts.
Adding a New Genome Index As part of the standard analysis process, reads are aligned to a genomic reference. The alignments and some summary statistics based on the alignments are included in the analysis report page. This describes the process to add a new reference genome, which is necessary when you start working with a new genome sequence. The TMAP aligner, which comes pre-installed on the Torrent Server, is used. For a standard Torrent Server configuration, files are installed in the standard reference location having the following pathname convention: /results/referenceLibrary/<index_type>/<genome_shortname>/ As of release 2.2, only the tmap-f3 index_type is supported.
Prerequisites Creating the Genome Index Follow-up Procedure

Prerequisites
Before you begin, you need your reference sequence in a FASTA format file and you need command-line access to the Torrent Server. It is important that the format of your FASTA file conform to Ion Torrent requirements. 1. Transfer the FASTA file to the torrent server, using a program like scp. 2. Move or copy the FASTA file to the standard reference location (see above). Example: /results/referenceLibrary/tmap-f3/
Creating the Genome Index
Select a Short Form of Genome Name
The short form of genome name is the name that you would like the reference option to appear when initiating a run on the PGM sequencer. The following rules apply for defining the short form of the genome name: The name must not match any of the existing references installed in the standard reference location: /results/referenceLibrary/... The name must only include alphanumeric characters, underscore ("_"), and period (".").
Create the Genome Index
The alignment package (ion-alignment) is distributed as the build_genome_index.pl script, which
Version 2.2
automates the TMAP index creation process. The script requires four inputs: A FASTA file. The short form of the genome name. The long form of the genome name. The genome version. Run the build_genome_index.pl script, with the required arguments:
build_genome_index.pl -f e_coli_dh10b.fasta -s e_coli -l "Escherichia coli dh10b" -v 1.0
The directory contains the following files, including the original FASTA file. The file sizes vary according to genome. For the human genome (3,000,000,000 bases in length) the combined size of all index files, including the original FASTA file itself, is under 8GB. For E. coli (4,600,000 bases in length) the size is about 0.4GB.
$ ls -1 /results/referenceLibrary/tmap-f3/e_coli_dh10b/ e_coli_dh10b.dict e_coli_dh10b.fasta e_coli_dh10b.fasta.fai e_coli_dh10b.fasta.tmap.anno e_coli_dh10b.fasta.tmap.bwt e_coli_dh10b.fasta.tmap.pac e_coli_dh10b.fasta.tmap.rbwt e_coli_dh10b.fasta.tmap.rpac e_coli_dh10b.fasta.tmap.rsa e_coli_dh10b.fasta.tmap.sa e_coli_dh10b.info.txt e_coli_dh10b.md5sum.txt tmap.log
Lastly, the file permission of the created indexes should be set to readable and writable by all users to be more compatible with Torrent Browser Reference Sequences database.
chmod -R a+w e_coli
Adding More Meta Information
The build_genome_index.pl script creates an info.txt file containing meta-information about the genome. The file consists of key-value pairs with one key-value pair per line. The key and value are separated by a tab, as shown in the following example:
Version 2.2
genome_name Escherichia coli dh10b genome_version 1.0 genome_length 4639675 index_version tmap-f3
Field genome_name
Required Yes
Description The long-form name of the reference set, which is displayed in reports. This name can be anything, but a common choice is "Genus species". The genome version, which is displayed in reports. This string can be anything, but good choices include the accession number, if there is one. The index version. As of software release 2.2, there is only one version supported: tmap-f3. The total number of bases in the genome, summed across all sequences, if there is more than one sequence in the genome. This value is used in coverage calculations. If this field is present, it is used to limit the number of reads that are aligned to the genome. This can be helpful when aligning to very large genomes, which can be time consuming. If more than these number of reads are called in the run, a random sample of this size is selected and used for alignment. The final report contains the results based on the sample taken and an extrapolation of the results for what you should get if all sequences were aligned.
genome_version
Yes
index_version
Yes
genome_length
Yes
read_sample_size
No
Version 2.2
read_exclude_length
No
If this field is present, it specifies the length up to which alignments are not trusted. This limit is allowed to vary by genome, because with more complex genomes longer spurious matches are possible. If not specified, a default value of 20 is used in downstream alignment summarization. Key-value pairs can optionally be defined to track other information. Such additional information is ignored.
(additional)
No
Adding the Genome to the PGM Sequencer Drop-down Menu After the indexes have been created and moved to the standard reference location, refresh the References tab. The newly created indexes are imported into the reference sequences database and added to the PGM drop-down menu.
Follow-up Procedure
The input FASTA file may be moved or deleted because there is a copy stored in the standard reference location directory.
Version 2.2
Booting Into Single-User Mode

Version 2.2
Booting Into Single-User Mode You may sometimes need to boot the Torrent Server into single user mode. One reason would be if you forgot and need to reset your password. 1. Reboot the server. A hard reboot may be needed if there is no way to login to the system. 2. Immediately after the BIOS completes, press the Esc key to display the grub menu:
3. Press the E key to edit the boot command line. 4. Scroll down to the linux ... command and scroll to the end of the line. 5. Add the word single to the end of the line:
Version 2.2
6. Press Ctrl-X to boot. 7. At the next window, scroll down to the netroot option and press Enter:
You are now logged into the system with root privileges.
Version 2.2
Archiving and Deleting Data

Version 2.2
Archiving and Deleting Data Archiving and deleting data covers the following topics: New Behavior
Run archival is enabled by default. The Torrent Server admin can change this setting is the Config tab -> Admin Interface -> Global Configs section.
A run's raw data is not set for deletion until an Acknowledge Delete checkbox is enabled. The Acknowledge Delete checkbox is in the Archive panel of the Services tab.
Defining Raw Data (DAT files)
Data generated on the Personal Genome Machine (PGM) are initially stored in acquisition files with a DAT extension . One DAT file is created per PGM flow. The DAT files are transferred from the PGM to the Torrent Server, where various algorithms are used to condense the information into a WELLS file, called 1.wells. Base calling and quality score algorithms are applied to the WELLS file to create a SFF file, which is converted into a FASTQ file.
The data in the DAT files are considered raw data and are subject to the archive operations described in this document. The WELLS, SFF and FASTQ files are considered processed data and are not affected by the archive setting described in this document.
To delete these files, you must contact your Torrent Server Administrator. These files can be deleted on the Torrent Server, using SSH. If you delete or move data in the file system, manually, the status of the data is not reflected in the Torrent Browser UI. For this reason, it is recommended that you use the Archive tool to archive and delete data.
Archiving Raw Data for a Run
Within the Run tab of Torrent Browser, you can archive the raw data for each individual sequencing run, as shown in the following figure.
Version 2.2
The following archiving options are available: Option Keep Description Retain all of the raw data (DAT files) on the Torrent Server for this run. Archive the DAT files for this run. The DAT files will be copied onto the archive source but the original files will be retained on the Torrent Server during the archive process. The file system is then updated with a symbolic link to the archived version of the files. Delete the DAT files for this run. The DAT files are removed permantently (recovery is not supported).
Archive Raw
Delete Raw
The default setting for archiving can be changed using the Config tab -> Admin Interface -> Global Config option. For example, you can configure Torrent Browser to mark all new runs as Keep, by default. Also, notice that the default configuration permits you to specify that a run should be flagged as Archive Raw, despite the lack of a properly configured archive source. The Archive Raw option requires configuration. The Delete Raw option only takes affect if the Archive Raw option is configured.
Locating Archived Data
The Archive configuration tool is located on the Services tab of the Torrent Browser.
The default state is archiving enable but not configured, as shown in the following figure:
After you configure an archive, the archive information is listed in the panel, under Archice Directory:
Version 2.2
Only one archive location at a time can be specified for a particular Torrent Server.
To populate the list if USB drives used to archive data, plug in a USB drive. The Torrent Server automatically mounts the drive. The drive should be ext3 formatted and uniquely labeled. (You may need to consult with a Linux system administrator.) If a USB drive is mounted, it automatically appears in the /media directory. If you are archiving to a Network Attached Storage (NAS) mount, the archive location is also in the /media directory.
Your Torrent Server Administrator might need to use the sudo mount command to mount the drive. After the drive is mounted it appears in the Backup directory drop down.
To add a new archive location from the drop down list of mounted USB drives, click the Edit button on the left side of the Archive tab.
Adding an Archive Location
Before the Archive schedule can be configured, a USB drive or an NFS file system must be mounted below /media. The archive service will only look for mount points under /media. Once the location has been mounted, then it will be visible as an archive directory. # Click the Add button on the Archive panel of the Services tab:
Only one archive location at a time can be specified for a particular Torrent Server.
1. Use the following dialog to configure the archive location:
Version 2.2
If there is no Archive directory present, verify that the archive file system is mounted under /media
Field Archive directory
Description Select the archive directory. The archive directory appears in the drop down menu after the drive is mounted. Define how many runs are to be moved to the archive after the archive process is initiated. Specify the time interval, in seconds, that the data archive tool delays before starting the archiving process. The default value 60 gives a one minute timeout. Other suggested values are 3600, once an hour, and 86400, once a day. Specify the Torrent Server local disk capacity before data are archived. The default value is 75%. After the local Torrent Server disks utilization reaches a percentage higher than the specified value, the archive process starts. Time interval, in hours, that experiment raw data is protected from deletion; measured from the experiment start time. The default value is 72 hours.
Number to archive (Runs)
Timeout (s)
Percent full before archive (%)
Grace period (hr)
Version 2.2
Bandwidth limit
Specify the rate at which data flows to the archive source. The following options are available in the drop down menu: - unlimited (DEFAULT) - 10 MB/sec Select the 10 MB/sec rate if the archive location is an NAS-formatted device and you are concerned about the archive process overloading the network. For USB attached devices, the unlimited the default value is recommended.
Email
Enter an email address where notifications are sent, if the archive location becomes full. When the archive location is full, one email a day is sent to this email address until the drive has been replaced. Email is sent through unauthenticated postfix, a Linux e-mail program. Check this checkbox to enable the Archive tool to run. Uncheck to disable archiving.
Enabled
Tracking Archives
When data are archived, the raw data location changes in the database for this run. Instead of a local path, a new path is created with a symbolic link to the archive location. The symbolic link is:
/media/<disk_name>/<run folder name>.
When using many USB drives to archive data, it is important that you name your archive disks using logical, unique disk names. Over time a large collection of archive sources can accumulate. Keeping track of various archive locations is facilitated if disks are uniquely named. After the raw data are archived, view the the Runs tab to confirm that the data are archived.
Monitoring the Archiving Process
You can check the status of the archive process using the Services tab. The Archive panel lists the archive directory, displays a chart of archive space usage, and lists completed runs that are set to have their raw data archived or deleted.
Version 2.2
Accessing Archived Data
There is no formal process defined to remove archived data. If you want to re-analyze archived data, use the following procedure: 1. Have the Torrent Server Administrator mount the archive drive, which does not need to be configured for active backup. 2. Click the Analyze button. The entry for the archived run has the location of the archived data listed in the Torrent Database. The software looks for the files at their archived location. If the drive is mounted, analysis proceeds typically.
Deleting Raw Data for a Run
Within the Run tab of Torrent Browser, you set the raw data for individual sequencing runs to be deleted, as shown in the following figure:
The following apply to the Delete Raw option: Deletion does not happen unless archiving is correctly configured. See #Adding an Archive Location. Deletion does not happen unless the deletion request is confirmed on each run. You confirm deletion for a run with the Acknowledge checkbox in the Archive panel Run listing. See the figure below. The run's raw data is deleted when the file server reaches a threshold set in the archive edit screen. See the Percent full before archive (%) field under #Adding an Archive Location. The Run listing in the Archive panel is shown in the following figure. The Acknowledge checkbox only appears for runs set with the Delete Raw option.
Version 2.2
Monitoring Hard Drive Space
A number of charts are available to monitor hard drive utilization. The Archive panel of the Services tab displays the following charts: #Storage Option Breakdown #File Server Space #Free Space/Fileserver #Archive Space #Average Days on Fileserver #Run Deletion Table
Storage Option Breakdown
The Storage Option Breakdown chart illustrates the archive flag status distribution, based on number of runs (not file size): Archive, Keep and Delete.
File Server Space
The File Server Space chart shows the distribution of hard drive space across all possible file servers mounted on the Torrent Server. The following figure shows a chart representing a standard Torrent Server installation, representing one file server:
Version 2.2
Free Space/Fileserver
The following figure shows the size load on file servers. Multiple file servers are shown if present.
Archive Space
The archive Space chart, shown in the following figure, indicates the used/free distribution on the currently mounted archive drive, if a drive is mounted.
Version 2.2
Average Days on Fileserver
The Average Days on Fileserver chart, shown in the following figure, estimates the average life span of data labeled as either Archive, or Delete.
The blue bar in /results/ shows that the average file labeled as archive will be retained in the /results/ directory on the torrent server for 38 days. The green bar shows the average life span of a file labeled for delete raw data. These life span estimates are a function of the number of files with each label and the total number of runs and hard drive space available. This does not mean that if you do a PGM run, today, it will be deleted in 38 days. The benefit of maintaining data online on the Torrent Server, instead of archiving, is that it gives you the ability to re-analyze the primary data to generate base calls, if necessary.
Run Deletion Table
The following table shows, in the top part, the next ten runs to be deleted and, in the bottom part, the next ten runs to be archived.
Version 2.2
Version 2.2
Mounting a USB Drive

Version 2.2
Mounting a USB Drive This describes the procedure for manually mounting and unmounting an external USB drive. To follow these steps, you should feel comfortable using the Linux command line and have a basic understanding of disk drives and partitions. By default, Ubuntu Desktop automatically mounts an external USB drive when the drive is attached to the machine, similar to Mac or Windows. Ubuntu Server, however, does not mount external hard drives automatically, so the ion-usbmount utility is included with the Torrent Server software, which automatically mounts attached USB drives in the /media directory. If a particular USB drive is not being automatically mounted by ion-usbmount, you may need to mount the drive manually. These instructions only provide an overview of the required steps, and may be a helpful reminder if you are new to the Linux operating system. For more detailed instructions and background information, refer to the Ubuntu documentation: h ttps://help.ubuntu.com/community/Mount
Mount a USB Drive
1. To see a list of the drives in the system, type the following command before connecting the USB drive:
sudo fdisk \-l" t
Make a note of the drives that are present, so you can be sure which drives are in the server. The local hard drive usually has a name such as /dev/sda, as in the following example:
ionadmin@itw-test01:~$ sudo fdisk -l Disk /dev/sda: 500.1 GB, 500107862016 bytes 255 heads, 63 sectors/track, 60801 cylinders Units = cylinders of 16065 * 512 = 8225280 bytes Sector size (logical/physical): 512 bytes / 512 bytes I/O size (minimum/optimal): 512 bytes / 512 bytes Disk identifier: 0x0004366b Device Boot Start End Blocks Id System /dev/sda1 * 1 37 291840 83 Linux Partition 1 does not end on cylinder boundary. /dev/sda2 37 60802 488092673 5 Extended /dev/sda5 37 60802 488092672 8e Linux LVM
2. To see a list of drives, including the new drive: a.

Version 2.2
2. a. Connect the USB drive. b. Wait about ten seconds and re-type: sudo fdisk -l. You should see a new drive. You need to know the device name of your USB drive, which is usually called /dev/sdb or /dev/sdc , depending on the number of drives installed. The partition is a number appended to the name of the physical drive. For example, the first partition on drive /dev/sdc would be called /dev/sdc1. In the following example, there is a 2GB partition (1953512001 blocks) attached to the system and named /dev/sdb1. It is configured with a Linux partition. (If the drive was formatted on Windows, it is either a FAT or NTFS partition.)
ionadmin@itw-test01:/$ sudo fdisk -l Disk /dev/sda: 500.1 GB, 500107862016 bytes 255 heads, 63 sectors/track, 60801 cylinders Units = cylinders of 16065 * 512 = 8225280 bytes Sector size (logical/physical): 512 bytes / 512 bytes I/O size (minimum/optimal): 512 bytes / 512 bytes Disk identifier: 0x0004366b Device Boot Start End Blocks Id System /dev/sda1 * 1 37 291840 83 Linux Partition 1 does not end on cylinder boundary. /dev/sda2 37 60802 488092673 5 Extended /dev/sda5 37 60802 488092672 8e Linux LVM Disk /dev/sdb: 2000.4 GB, 2000398934016 bytes 255 heads, 63 sectors/track, 243201 cylinders Units = cylinders of 16065 * 512 = 8225280 bytes Sector size (logical/physical): 512 bytes / 512 bytes I/O size (minimum/optimal): 512 bytes / 512 bytes Disk identifier: 0x5786fcfb Device Boot /dev/sdb1 Start 1 End 243201 Blocks 1953512001 Id 83 System Linux
3. If the drive is a Windows FAT or NTFS partition, reformat the drive as an ext3 partition to preserve the Linux file information. Be careful that you are formatting the correct hard-drive!
To reformat the drive as ext3 partition, type sudo mkfs.ext3 <your_device>. For example:
sudo mkfs.ext3 /dev/sde5
4. Create a mount point, an empty directory, where you want to attach the drive. To make a mount point under the /mediadirectory, type the following command:
Version 2.2
sudo mkdir /media/external
5. You can mount the physical drive to the mount point using the following command:
sudo mount /dev/sdb1 /media/external
Unmount a USB Drive
Before disconnecting a drive, it is recommended that you unmount it first, to ensure that all data has been completely written to disk. If you pull out the USB cable, there is a real risk of data loss. The command to unmount the drive is almost identical to the command to mount the drive:
sudo umount /dev/sdb1 /media/external
Version 2.2
Backing Up and Restoring Data

Version 2.2
Backing Up and Restoring Data The Torrent Server maintains the following types of data in separate locations: Raw PGM sequencer data are stored in the /results/<PGM_Name> directory, by default. Report data are stored in the /results/analysis/output/Home directory, by default. Database records are stored in the postgreSQL database.
Backing up Data
The easiest way to back up raw PGM sequencer data and report data is to backup the whole /results directory. If you have a backup storage location on the network with sufficient capacity, you are encouraged to use the rsync utility to periodically backup the repository over the network. The archive tool built into the Torrent Browser can also be configured to move raw data to offline, USB drive storage, but this tool is not a true backup mechanism because it does not make a copy of the data.
Backing Up the PostgreSQL Database
Database records hold the run, report and configuration data viewed using the Torrent Browser interface. Compared to the raw data and report data, database records are relatively easy to backup because the database size is only a few hundred megabytes. To backup the database, run the following command:
pg_dump -U ion -c iondb > backup_file
Use the following simple example script to automate the database backup, calling the script from a cron job:
#!/bin/bash cd /path/to/backup/directory TIMESTAMP=$(date +%Y%m%d_%k%M%S) pg_dump -U ion -c iondb > iondb.$TIMESTAMP.backup gzip iondb.$TIMESTAMP.backup rm -f iondb.$TIMESTAMP.backup
This produces a compressed backup file with a name like iondb.20100711_142442.backup.gz.

Restoring the PostgreSQL Database
The following instructions delete the current database.
To restore the database, you need a complete, working Torrent Server installation. The two scenarios for restoring a database are: Installing a new Torrent Server from the Torrent Server installation disk due to migrating the database to a new server or needing to reinstall the server.
Version 2.2
Replacing the database on an existing Torrent Server, possibly because the database is corrupted and you want to restore a previous version. To restore the database from the backup file, execute these commands on the Torrent Server:
# copy the backup file to the server and decompress it gzip -d iondb.20100711_142442.backup.gz
# stop the Torrent Server background processes sudo /etc/init.d/ionCrawler stop sudo /etc/init.d/ionJobServer stop sudo /etc/init.d/ionArchive stop sudo /etc/init.d/ionPlugin stop # login as user postgres sudo su postgres # restart the service to clear database connections /etc/init.d/postgresql-8.4 restart # drop the existing iondb database dropdb iondb # create a new empty database psql <<-EOFdb CREATE DATABASE iondb; GRANT ALL PRIVILEGES ON DATABASE iondb to ion; \q EOFdb # import data psql -e iondb < iondb.20100711_142442.backup # logout of user postgres exit # start the Torrent Server background processes sudo /etc/init.d/ionCrawler start sudo /etc/init.d/ionJobServer start sudo /etc/init.d/ionArchive start sudo /etc/init.d/ionPlugin start
Occasionally, there is a django error after completing the import data step.
If an error is displayed on the browser UI, repeat the following steps: 1. Drop database. 2. Create database. 3. Import data.
Version 2.2
Monitoring Free Disk Space

Version 2.2
Monitoring Free Disk Space It is critical that enough disk space is available on the server to avoid data loss so it is important to have a strategy that, periodically, monitors disk space and archives or deletes data as needed.
The Torrent Browser Archiving Tool
A disk space monitoring policy can be implemented using the UI Services tab Archive dialog. Refer to the Archivin g and Deleting Data section for more information about the Ion Torrent archiving tool. Additionally, your IT department may choose to deploy a monitoring tool, such as Ganglia. Ion Torrent recommends using the built-in Torrent Browser archiving tool to automatically manage disk space. Navigate to the Torrent Browser Runs tab Storage column and select the archive policy from the drop-down menu.
The Runs Tab Storage Indicator
The Torrent Browser Runs tab displays a color-coded storage indicator light to the left of the Storage header. This indicator allows you to monitor space available on your Torrent Server hard drive.
Color Green
Meaning Your Torrent Server hard drive is less than 70% full. The height of the green bar corresponds to the percentage of disk space in use. Your Torrent Server hard drive is between 70% and 90% full. Your Torrent Server hard drive is more than 90% full.
Yellow
Red
When the Torrent Browser UI is Not Available
Use the following procedure when the Torrent Browser UI is not available: 1. Log into the server using an ssh client:
$ ssh ionadmin@ion-torrent-server $ password: ionadmin
2. Enter the df command to display partitions and disk utilization:
Version 2. 2.2
$ df -h ionadmin@itw-test01:~$ df -h Filesystem Size Used Avail Use% Mounted on /dev/sda3 5.3T 372G 4.6T 8% / none 24G 200K 24G 1% /dev none 24G 0 24G 0% /dev/shm none 24G 88K 24G 1% /var/run none 24G 0 24G 0% /var/lock none 24G 0 24G 0% /lib/init/rw /dev/sda5 61G 524M 57G 1% /tmp /dev/sda1 276M 29M 233M 12% /boot /dev/sda4 3.8G 2.4G 1.3G 65% /var nas3:/c/results2 19T 17T 1.7T 91% /results2 nas2:/c/archive/tahiti 19T 13T 5.3T 71% /media/archive nas1:/c/results 19T 17T 2.1T 89% /results4 nas1:/c/results1 19T 16T 2.1T 89% /results3
Most growth is seen in the /results directories, which is where Ion Torrent data are stored.
The Use% column indicates how much space is being used. If there is insufficient space on the Torrent Server, the raw data are retained on the PGM until space becomes available.
Version 2.2
Installing and Using a UPS

Version 2.2
Installing and Using a UPS An uninterruptable power supply (UPS) can be used with a Torrent Server. While not required, it is recommended to have a UPS installed to provide some protection for a Torrent Server from the negative consequences of fluctuations in the line power, by providing consistent line power to the server. These fluctuations includes spikes as well as brown-outs. Additionally, we have developed a solution that allows for a controlled shutdown, in cases when the line power goes out completely for an extended period of time.
Hardware
The recommended device is a Dell UPS Tower 1000-Watt 100/ 120 VAC (Dell Part# : 330-7516). It is shown below.
Only this unit has been tested and verified compatible with our software. Theoretically, any of the units listed here could work as well: http://www.networkupstools.org/stable-hcl.html . However, none of these models have been tested to ensure compatibility and will likely require additional configuration, beyond what is described here.
Software
There is an optional package that should be installed if a UPS is to be used: ion-tsups. It is not installed by default and must be installed according to the procedure listed in Setup. A UPS unit can be used without this package, but the Torrent Server will not be able to initiate a controlled shutdown, if this package is not present.
Setup
The setup process is as follows. 1. Plug in and power on the UPS unit. 2. Connect the USB cable from the UPS unit to the Torrent Server. 3. Confirm that the Torrent Server and UPS unit are communicating, by issuing the following command:
upsc ionups ups.status
Version 2.2
The response should be one of the following:
OL
This means the UPS is on AC line power.
OL CHRG
This means the UPS is on AC line power and the battery is charging.
Usage
When the battery capacity drops to 30% of its full capacity, the software initiates a sequence of events which shuts down the server in a controlled manner, and then shuts off the UPS output as well. The UPS unit will not turn on its output until line power is restored. Once power has been restored to the UPS unit and, consequently, the Torrent Server, the Torrent Server must be powered back on. When the Torrent Server boots up again, it will resume normal operation, with the caveat that whatever runs were actively being analyzed at the time of shutdown will not automatically resume. These runs will need to be manually restarted. However, any runs that had been queued and not yet analyzed (due to other runs being analyzed ahead of them) will start analysis automatically without any intervention required.
Version 2.2
Axeda Remote System Monitoring (RSM)

Version 2.2
Axeda Remote System Monitoring (RSM)

Overview
The Axeda RSM (Remote System Monitoring) agent is a software component installed automatically on the Torrent Server and PGM via the software update process. This agent sends a heartbeat message to Life Technologies, approximately every sixty seconds. This information is used to track the deployment and software configuration of machines in the field. Data is collected in the Axeda monitoring database, where Life Technologies technical support personnel can review the information collected by the agents. Since the heartbeat message is sent many times an hour, Tech Support can quickly see if a machine is online, the software versions, and some technical details about the instrument such as temperature and hard drive status. The agent also allows Ion Torrent to remotely log into the PGM System and the Torrent Server, which is required for system support. Without remote access, Ion Torrent Field Application Scientists cannot access, view, and troubleshoot issues regarding machine performance.
Data automatically collected by the RSM Agents
Field names, data types, and examples of the data being collected are described in the following tables. This information is being sent automatically from the Torrent Servers and PGMs back to Life Technologies.
Torrent Server
Event Name TS.Config.configuration TS.Config.hostname TS.Config.ipaddress TS.Config.mode TS.Config.serialnumber TS.host TS.HW.HD.ResultsFull TS.TYPE TS.Version.alignment TS.Version.analysis TS.Version.dbreports
Type String String String String String String Analog String String String String
Sample Value standalone Ion-torrent-server 10.45.3.246 Master 1SMJFP1 (Dell service tag) Ion-torrent-server 58.99 (chart) TS1 1.42-0 1.40-0 1.95-3
Version 2.2
TS.Version.docs TS.Version.referenceLibrary TS.Version.tmap TS.Version.tsconfig TS.Location.country TS.Location.region TS.Location.code TS.Location.city TS.Location.ipaddress TS.Server.Crawler TS.Server.Archive TS.Server.Job TS.Server.Plugin TS.Experiment
String String String String String String String String String String String String String String
1.15-1 1.6-1 0.0.19-1 1.3-9 USA Maryland MD Rockville 37.38.39.40 ok/offline/error ok/offline/error ok/offline/error ok/offline/error serial_number:sn09c121501 chipbarcode:BA0000531 chiptype:316C flows:260 cycles:8 gain:0.63300000000000001 cal_chip_high_low_inrange:[19052, 9272, 6309130] sysSNR:12.18 aveKeyCounts:40.0 serial_number:sn09c121501 chipbarcode:BA0000531 flows:260 cycles:8 gain:0.63300000000000001 cal_chip_high_low_inrange:[19052, 9272, 6309130] sysSNR:12.18 aveKeyCounts:40.1
PGM
Event Name
Type
Sample Value
Version 2.2
Instrument.Alarm.XXX
string
All alarms without customer sensitive information are sent, e.g. PressureTooLow, BadBootDriveDetected, etc. All events without customer sensitive information are sent, e.g. CompletedRunInfo Pressure RunAborted 0 (chart) Valve Board not accessible Valve Board Down Stream Errors Valve Board Up Stream Errors Run aborted Lost chip connection, run aborted U-boots dont match Kernels dont match Results drive not accessible Bad boot drive detected Bad data drive detected 34.001 (chart) Stork 10.2 (chart) 27.06 (chart) PGM1 4 A.1 180 31
Instrument.Event.XXX
string
Instrument.Event.Pressure Instrument.Event.ValveBoard
analog string
Instrument.Event.RunAborted Instrument.Event.LostChipCon Instrument.Event.UBoot Instrument.Event.Kernel Instrument.Event.ResultsDrive Instrument.Event.BootDrive Instrument.Event.DataDrive Instrument.HW.HD1 Instrument.InstrumentName Instrument.Pressure Instrument.Temperature Instrument.TYPE Instrument.Version.Board Instrument.Version.Datacollect Instrument.Version.driver
String String String String String String String analog String Analog Analog String String String String
Version 2.2
Instrument.Version.fpga Instrument.Version.Graphics Instrument.Version.LiveView Instrument.Version.OS Instrument.Version.Scripts
String String String String String
70 15 268 12 16.3.58
Remote Access for Troubleshooting
In the event that there is a problem with the PGM or Torrent Server, this agent will allow Life Technologies support personnel to remotely: Collect log files from the systems for review Restart the device Upgrade software Provide a remote login connection to the device for further diagnostic work These types of activities would only be initiated at the request of the administrative contact, to ensure compliance with site security policies. When a issue with the PGM or Torrent Server is reported. We make all efforts to solve the issue by telephone or email. If remote access is required for additional troubleshooting a member of Life Technologies support staff will need authorization from the technical contact to initial remote connection. Only after getting authorization will anyone from Life Technologies proceed with remote troubleshooting. Once the issue is resolved you will be notified. Additional authorization will be required before initiating any further remote assistance.
FAQ
What is Axeda? Axeda is a software provider that Life Technologies has partnered with to develop this Remote Monitoring solution. Life Technologies has used Axeda successfully with other products http://www.axeda.com/community/customers/ap plied-biosystems What is the benefit of these agents? The Remote System Monitoring agents facilitate Life Technologies ability to help our customers maintain their systems in running order. The agent provides system metrics many times a hour and if a problem with the PGM or Torrent Server is detected, the RSM agent provides real time warnings to help proactively diagnose issues before they cause any failures and downtime. Life Technologies has invested significantly in the development of this tool in order to best serve our customers. Without access to the RSM agent, issues may not be detected until the system goes down. Furthermore without the remote access capabilities diagnosing and implementing a solution can take much longer and will require significant back and forth over telephone and email with your onsite personal. By using the agent, downtime can be reduced to hours instead of days and require minimal onsite resources. Can the agent be disabled?
Version 2.2
Life Technologies has invested significantly in the development of this tool in order to best serve our customers. By using the agent, potential issues can be detected before they cause downtime. And in case troubleshooting is required, remote access through the agent can reduce resolution time to hours instead of days and require minimal onsite resources. Also keep in mind that no remote troubleshooting can be provided with this agent. Therefore, disabling this agent is not recommended. Despite these advantages the agent needs to be disabled, the site firewall can block outbound port 22 and 443.
Version 2.2
Administration
Version 2.2
Administration In the Torrent Browser user interface, select the Config tab to perform the following administrative tasks: Add customer support contacts Change the server name displayed in the Torrent Browser View and enable installed plugins Postfix email configuration Database administration You must use your ionadmin account for steps described in the Torrent Server Administration Guide. (Do not use your ionuser account.)
Add Customer Support Contacts Change the Displayed Server Name View Installed Plugins
Version 2.2
Postfix Email Configuration Database Administration

Add Customer Support Contacts
Fill in the information for a lab contact and an IT contact at your organization, and click the Save Changes button.
Change the Displayed Server Name
This section describes how to change the server name displayed in the Torrent Browser. The default name is Torre nt Server. This change only affects the server name shown in the Torrent Browser, and the default bookmark entry if you create a browser bookmark for the server. 1. Go to the Config tabs. On this tab only, the server name field is editable:
2. Replace Torrent Server with the name of your choice.
3. Click the Save Changes button on the top right. The server display name is now changed.
View Installed Plugins
The top panel lists the plugins currently installed in the /results/plugins directory, and plugin attributes:
Version 2.2
See Installing Analysis Plugins for plugin installation procedures. It is important to note the plugin version number to be sure you are working with the correct version.
Table Heading Enabled
Description Run flag: checked = Plugin is enabled to run. unchecked = Plugin is disabled. Auto-run flag: Enabled = Plugin is run automatically following analysis. Disabled = Plugin can only be run manually or the plugin is disabled. Plugin name. Plugin version number. Plugin installation date. The Manage link allows you to configure or uninstall a plugin. Not all plugins support the configuration option.
Auto-Run
Name Version Date Manage
The Enabled and Auto-Run checkboxes govern plugin behavior.

Enabled
Check the Enabled checkbox to enable a plugin. When a plugin is enabled, The plugin can be selected and run manually from the Default Analysis Report page. If the plugin is enabled and the Auto-Run checkbox is checked, the plugin runs at the completion of every analysis.
Auto-Run
Version 2.2
The Auto-Run checkbox is present only if the plugin author has not configured the plugin to not be run automatically. A plugin runs automatically, if both the Enabled and Auto-Run checkboxes are checked. Using the Admin Interface to manage Plugins, you can delete one or more plugins. However, only the plugin settings are deleted. The next time you load the page, the plugin directory is scanned and plugins are listed in the Config tab with their default values.
Postfix Email Configuration
The Torrent Server archiving system attempts to send email notifications of archive space problems through Postfix. 1. In the email address configuration panel, click Add to add an email address. This displays the Add dialog:
2. Enter an Email address in the edit box. 3. Click the Selected checkbox to enable this email address for sending email. 4. Click Save to save the email address and exit the session. The new email address is displayed in the Email p anel:
Click Cancel to end the session without saving the email address.
Database Administration
In the Database Administration panel, click Admin Interface to access the administrative configuration dialog, which displays in a new window: Where configuration options are directly available through an interface tab, such as Templates, the interface tab mechanism should be used to configure the item.
Version 2.2
Version 2.2
The following links describe configuration items only available using Admin Interface: Configure Chips Configure Experiments Configure Global Configs Adding or Changing a PGM Configure Users
Version 2.2
Configure Chips
Version 2.2
Configure Chips
The Chips dialog controls default analysis processing for a particular Ion Torrent chip. Please, check with Ion Torrent before modifying this information.
When an analysis job is started, the Torrent Browser looks up the chip type in the database and appends the command line argument (Args) to handle the chip in a particular way. When starting analysis, the command line parameters can be modified manually using the Runs Tab and selecting the Advanced option. This permits analysis processing behavior be modified on a run-by-run basis. When you check the takeover checkbox in the Runs Tab Advanced options, parameters specified in the database for the takeover chip are applied to the run.
To modify the default behavior: 1. Click Chips, or Change, on the Chips line:
Version 2.2
2. In the Name column, click the chip you want to change:
3. Modify the chip Name, Slots and Args, as needed:
Version 2.2
4. Click Save to save your change.
Version 2.2
Configure Global Configs

Version 2.2
Configure Global Configs
The Configure Global Configs dialog permits you to specify run parameters. These are advanced configuration options, check with Ion Torrent if you are unsure of the effect of modifying this information.
Adding a Configuration
Version 2.2
There should only be one Global Config record. Do not create multiple Global Configs! 1. Click Add on the Global configs line to set the values of the configuration parameters: - Plugin folder - Default command line - FASTA path - Reference path - Records to display - Default Test Fragment key - Default library key - Default flow order - Plugin output folder - Default plugin script - Web root - Site name - Default storage options - Auto-acknowledge archive - Disable SFF trim - Args to use for SFF trim - Barcode args The sequence entered in the Args to use for SFF trim field overrides the adapter sequence that is coded into the analysis binary. To override the current setting to use the release 1.1.0 and 1.2.0 settings, enter the following arguments (on a single line): --qual-cutoff 9 --qual-window-size 30 --adapter-cutoff 5 --adapter CTGAGACTGCCAAGGCACACAGGGGATAGG
This parameter will immediately affect all new Analysis Reports generated by the Torrent Server. Check with your FAS if you have questions about this feature.
2. Click the Save button when you are done making changes.
Changing an Existing Configuration
1. Click Change, to the right of Global configs, to modify an existing configuration. 2. Click the Name of the configuration you want to change.
Version 2.2
3. Enter the values for the items you want to change. 4. Click the Save button to save your change(s).
Version 2.2
Shutdown Server
Version 2.2
Shutdown Server
The Management Actions panel displays the Shutdown Server dialog:
See Updating Torrent Server Software for a description of the Update Server feature.
1. In the Management Actions panel, click Shutdown Server to start the shutdown dialog.:
2. Click OK to confirm your shutdown request. Click Cancel to exit without shutting down the server.
You will see a confirmation message indicating the server is shutting down:
Version 2.2
Management Actions
Version 2.2
Management Actions
The Config tab Managment Actions section allows you to perform the following actions:
These procedures require your ionadmin account. (Do not use your ionuser account.)
View Network Settings
Click the View Network Settings link to see the following information about the Torrent Server:
Update the Torrent Server software
Updating your Torrent Server software causes the Torrent web services to restart. Make sure no analysis jobs are running on the server or are queued to run. When upgrading your Torrent Server to release 2.2, the following are required: Your current version of the Torrent Suite must be 2.0.1. Your PGM sequencer must be updated to release 2.0. After the upgrade, you must update your reference indices. You can use the Rebuild All Now button in the References tab. Depending on the size of your references, this process can take several hours. Analysis runs
Version 2.2
are not recommended during the reference indices update. See Updating Reference Library Indexes for more information.
To get the full instructions for updating your Torrent Server please refer to the latest Release Notes on the Ion Community. There may be additional steps and procedures required, depending on the type of software upgrade.
Follow these step to update your Torrent Server software: 1. Log in with your ionadmin account. 2. Click the Config tab Admin Interface link, and scroll down on the Management Actions section. 3. Click the Update Server link
4. Click the Update button.
5. On the confirmation screen, click OK.
Version 2.2
5.
6. When you see the following text, the update has finished. Verifying ion-gpu Verifying ion-analysis Verifying ion-alignment Verifying ion-pipeline Verifying tmap Verifying samtools Verifying ion-rsmts Verifying ion-samita Verifying ion-torrentr Verifying ion-plugins Verifying ion-pgmupdates Verifying ion-docs Verifying ion-referencelibrary Verifying ion-sampledata Verifying ion-publishers Verifying ion-onetouchupdater Verifying ion-dbreports ================================================================== TSconfig software update process completed successfully ================================================================== =============================================== Torrent Suite Update is done =============================================== 7. After the Torrent Suite update is done, please verify the version number in About tab. The version number should now be 2.2.
Update the Ion OneTouch Device
This procedure requires actions on both the Ion OneTouch device and in the Torrent Browser Management Actions section. Follow these steps to update the Ion OneTouch device software: 1. Connect the Ion OneTouch device and the Torrent Server with an Ethernet connection. 2. Get the updated IP address of the Ion OneTouch device. Follow either one of the following steps: Power cycle the Ion OneTouch device, or Wait for the IP address to update (takes one or two minutes). To check for the IP address, press the About button on the Ion OneTouch device.
Version 2.2
This page does not refresh. To refresh, go to a different screen and then go back.
3. As ionadmin, in the Torrent Browser Config tab Management Actions section, click the link Update OneTouch Device.
4. Click Update. 5. On the Ion OneTouch device, a splash screen appears with update progress. After update is complete, the Ion OneTouch device reboots itself.
Version 2.2
Configure Users
Version 2.2
Configure Users
The Users dialog provides for creating or modifying user accounts to access Torrent Server using the Torrent Browser:
Adding a User
1. Click Add on the Site administration menu for Users. 2. Enter a Username and Password; enter the password twice to confirm: If the user already exists or the password is invalid, you will be prompted to enter the correct information before continuing.
Version 2.2
3. Click the wanted save option. Clicking Save permits you to provide additional user information using the Cha nge user dialog, as follows. 4. In the Personal info dialog, optionally, enter a First name, Last name and E-mail address:
5. (Optional) In the Permissions dialog, check the Active, Staff status and Superuser status checkbox(es), as needed, and select the wanted User permissions:
Checkbox
Description
Version 2.2
Active
Designates whether or not this user is treated as active. The recommended method is to uncheck this item rather than delete this account. Designates whether or not this user can login to this administration site. Designates that this user has all permissions without explicitly assigning them.
Staff status
Superuser status
Selecting User permissions There are three ways to specify User permissions: a. Enter a string in the search window. All permissions matching the string are displayed, from which you may select permissions by highlighting the permission(s) and clicking the right arrow, in the center. b. Scroll through the permissions list. Highlight the wanted permission and click the right arrow to select the highlighted permission. Also hold down the control key to select more than one permission. c. Click Choose all at the bottom of the dialog, to highlight all available permissions, and click the right arrow to select all permissions. To deselect any permission, highlight selected permission(s), in the right window, or click Clear all foll owed by clicking the left arrow. 6. (Optional) Set the Last login and Date joined times, manually or using the calendar and clock icons. Click T oday and Now to set the values to the current date and time:
7. (Optional) Click the plus sign to display the Add group dialog.
8. (Optional) Currently, the Add group dialog is the same as the User permissions dialog in step 5, above. Add the user to the group in the same way as described in step 5. 9. Choose one of the following Save options to complete adding the new user.
Version 2.2
9.
#Save and add another #Save and continue editing #Save Save and add another Choose this option to complete adding the new user, and return to the Add user page to another user. Save and continue editing Choose this option to complete adding the new user, and return to the Change user page. Save Choose this option to complete adding the new user. Clicking Save takes you to the Select user to change page.
Modifying a User Entry
Use the following procedure to modify the information and permissions for an existing user: 1. On the Users line of the main Site administration menu, click Change. 2. On the Select user to change page, click the Username of the user you want to change. Usernames can be filtered, selected to the right, according to: By staff status, By superuser status or By active status.
3. Use the Change user dialog to modify user information in the same way as described for adding a user, beginning in step 4 above.
To be able to login to the server, it is important that you remember to check the Staff status checkbox in the Permissions dialog, shown in the following figure.
Version 2.2
4. Choose one of the Save options at the bottom of the page to save your changes.
Deleting a User Entry
Use one of the following procedures to delete an existing user. Delete a Single User 1. In the Users line of the main Site administration menu, click Change. 2. On the Select user to change page, click the Username of the user to be deleted. 3. At the bottom-left of the user information page, click Delete. 4. Confirm that you want to delete the user by clicking Yes, I'm sure:
Delete Multiple Users 1. In the Users line of the main Site administration menu, click Change. 2. On the Select user to change page, check the checkbox for each of the users you want to delete. 3. Click the drop-down menu and select Delete selected users:
4. Click Go:
Version 2.2
5. Confirm the list of users you want to delete by clicking Yes, I'm sure. 6. On the Select user to change page, the list of users confirms your deletion(s).
Version 2.2
Configure Experiments
Version 2.2
Configure Experiments
Use the Experiments (runs) dialog to modify run information stored in the database. Please, check with Ion Torrent before modifying this information. Runs are automatically added to the database by the Crawler process, so you should not add a run to the database using this dialog.
Modifying an Experiment
1. Click Experiments, or Change on the Experiments line, to modify an existing experiment description:
2. Click the ExpName of the run you want to change:
Version 2.2
2.
3. Edit the experiment description fields as necessary. Typically, you are only need to modify items that were not entered correctly at the time the information was captured on the PGM sequencer. 4. Click Save options to save your changes.
Deleting an Experiment
These procedures delete the experiment and any child records generated, such as reports. If the raw data are still present in the file system, the Crawler re-discovers the data and adds it to the database. To completely remove the run, you must delete the raw data before deleting the experiment from the database. Deleting a Single Experiment 1. Click the ExpName you want to delete. 2. Click Delete, on the lower left 3. Confirm the deletion. Deleting Multiple Experiments 1. Check the checkbox of the experiment(s) you want to delete, or check the top checkbox to select all experiments for deletion:
2. In the drop-down menu, select Delete selected experiments:
3. Click the Go button to perform the requested Action:
Version 2.2
4. When prompted to confirm the deletion, click Yes, I'm sure. The list of experiments confirms your deletion.
Version 2.2
Installing Analysis Plugins

Version 2.2
Installing Analysis Plugins The Plugin Store Enabling an Installed Plugin Enabling and Disabling Automatic Plugin Execution Deleting a Plugin Analysis plugins are customized scripts, invoked off at the end of the analysis pipeline, which can provide secondary data analysis or other custom functionality. Plugins may be distributed by Ion Torrent or can be developed and customized by end users. Care should be taken installing and running 3rd-party plugins that are not supported by Ion Torrent.
The Plugin Store
In the Torrent Browser Plugin Store, you can search for plugins of interest, download them to your Torrent Server, and run them on your analyses. These plugins are not developed by Ion Torrent. You can be install a Plugin Store plugin automatically or manually. Each plugin page provides an XML link for automatic installation and a zipped offlline installer for manual installation. These links are shown in the example plugin page below.
For more information on the Torrent Browser Plugin Store, visit the store at this link, Torrent Browser Plugin Store, or see The Plugin Store User Guide.
Automatic Installation From the Plugin Store
Follow these steps for automatic installation of a plugin from the Plugin Store: 1. In the Torrent Browser Plugin Store, go to the page for the plugin to be installed. 2. Scroll to the bottom of the page, and find the XML link with these directions: "Copy and paste this link into your Torrent Browser under the 'Config' tab after clicking the 'Add' button in the plugin section to auto install the plugin." An example XML link is the following: http://torrentcircuit.iontorrent.com/warehouse/download/feedfile/Ex.xml Copy this link. 3. In your Torrent Browser, go to the Config tab Plugin section, and click the Add button:
Version 2.2
3.
4. In the Install via URL tab, paste the plugin's XML link into the empty field, and click Download.
After installation is complete, you must click the Enabled checkbox to make the plugin available to users.
Manual Installation of a Plugin Store package
Follow these steps to download a plugin's installation package for manual installation: 1. In the Torrent Browser Plugin Store, go to the page for the plugin to be installed. 2. Scroll to the bottom of the page, and find the table with this explanation: "Links to the plugin archives are provided below for manual installation." Click the Offline Installer link.
3. Save the download to your machine. 4. In your Torrent Browser, go to the Config tab Plugin section, and click the Add button:
Version 2.2
4.
5. In the Install via Zip upload tab, click the Select file button. Browse to the Offline Installer you just downloaded, and select it.
6. Click the Upload file button. The installation begins.
7. After installation is complete, you must click the Enabled checkbox to make the plugin available to users.
Enabling an Installed Plugin
Follow these steps to enable an installed plugin: 1. Copy plugins to the /results/plugins directory. Each plugin is in a separate subdirectory having the plugin name, by convention. 2. Open the Torrent Browser Config tab. The installed plugins are displayed:
Version 2.2
3. Click the Enabled checkbox for new plugins, to make them available to users. Click the Auto-Run checkbox for plugins that you want to be run on every analysis run.
You must click the Enabled checkbox for new plugins, to make them available to users.
Enabling and Disabling Automatic Plugin Execution
Click the Enabled checkbox for the plugin to enable the plugin to run automatically after pipeline analysis processing. Uncheck the checkbox to disable automatic plugin exection. When the plugin Auto-Run status is enabled, the plugin is run at the end of every analysis job. Plugins that require manual input, for instance for options or user information, cannot be Auto-Run.
Deleting a Plugin
To delete a plugin: 1. Go to the /results/plugins directory and delete the sub-directories containing the plugins you want to delete. 2. Open the Torrent Browser Config tab, click Admin Interface and login. 3. Select Plugins. 4. Click the checkbox for the plugin(s) you removed from the plugins directory. 5. In the Action drop-down menu, select Delete selected plugins. 6. Click Go to delete the plugins and confirm your action.
Version 2.2
Troubleshooting
Version 2.2
Troubleshooting The following topics are common tasks to start with when investigating problems:
Checking Crawler and Job Server Status
The following table lists the background processes that run on the Torrent Server: Process Crawler Program crawler.py Startup Script ionCrawler Description Searches for new runs from the PGM sequencer and puts run information into the database so that they appear on the browser Ru ns tab. Sends analysis jobs to the Sun Grid Engine (SGE). Sends plugin jobs to the Sun Grid Engine (SGE). Celery is a background job processor for Django. When space is needed, cleans raw data off the disk based on the retention setting for each run, as defined on the Ru ns tab.
Job Server
serve.py
ionJobServer
Plugin Server
ionPlugin.py
ionPlugin
Celeryd
manage.py
celeryd
Archive Manager
backup.py
ionArchive
Startup scripts for each process can be found in the /etc/init.d directory
Log file for each process can be found in the /var/log/ion directory. They are: crawl.log iarchive.log celery_w1.log ionPlugin.log
If these processes are not running, run information is not updated and analysis reports are not generated. If this occurs, there is no risk of data loss but the Crawler and Jobs Server processes should always be running. The Arc hive process only runs if archiving has been configured. Process status is displayed in the browser Services tab, as shown in the following figure:
Version 2.2
If a process is not running, a Down or Offline reason is displayed in the Services tab.
Use the status command to verify that the three processes are running:
ps ax | grep py
This displays the running Torrent Server processes:
Version 2.2
ionadmin@ion-torrent-server:/etc/init.d$ ps ax | grep py 9279 ? Sl 0:06 /usr/bin/python /opt/ion/iondb/bin/crawler.py 9303 ? Sl 0:02 /usr/bin/python /opt/ion/iondb/anaserve/serve.py 9317 ? Sl 3:23 /usr/bin/python /opt/ion/iondb/backup/backup.py 9338 ? Sl 0:02 /usr/bin/python /opt/ion/iondb/bin/ionPlugin.py 29852 ? Sl 0:03 /usr/bin/python /opt/ion/iondb/manage.py celeryd --concurrency=4 -n w1.oldwest --loglevel=INFO --logfile=/var/log/ion/celery_w1.log --pidfile=/var/run/celery/celery_w1.pid 29861 ? S 0:00 /usr/bin/python /opt/ion/iondb/manage.py celeryd --concurrency=4 -n w1.oldwest --loglevel=INFO --logfile=/var/log/ion/celery_w1.log --pidfile=/var/run/celery/celery_w1.pid 29862 ? S 0:00 /usr/bin/python /opt/ion/iondb/manage.py celeryd --concurrency=4 -n w1.oldwest --loglevel=INFO --logfile=/var/log/ion/celery_w1.log --pidfile=/var/run/celery/celery_w1.pid 29863 ? S 0:00 /usr/bin/python /opt/ion/iondb/manage.py celeryd --concurrency=4 -n w1.oldwest --loglevel=INFO --logfile=/var/log/ion/celery_w1.log --pidfile=/var/run/celery/celery_w1.pid 29864 ? S 0:00 /usr/bin/python /opt/ion/iondb/manage.py celeryd --concurrency=4 -n w1.oldwest --loglevel=INFO --logfile=/var/log/ion/celery_w1.log --pidfile=/var/run/celery/celery_w1.pid
If a process (crawler.py, backup.py, manage.py, ionPlugin.py, or serve.py) is not running, you must manually restart it because processes do not start automatically.
Restarting Services
Currently, there is no method to restart a process using the Torrent Browser UI. The easiest approach is to shutdown and restart the server. Before restarting the server, make sure that no PGM sequencers are uploading data to the server,otherwise the file transfer will be interrupted. You can also restart the processes using the scripts located in the /etc/init.d directory. For example, use the following command to restart the Crawler:
user@svr:/etc/init.d$ sudo /etc/init.d/ionCrawler restart Stopping crawler Starting crawler pid = 26025
Use the ps ax | grep py command or the Torrent Browser UI to verify that the processes are running. After restarting a process, it continues from the point where it was interrupted, and no more user interaction is needed. If the processes do not continue to run after being restarted, contact Ion Torrent for assistance.
Verifying Network Connectivity and Name Resolution
There can be many reasons for network connectivity or name resolution to fail. Use the following procedure to try to resolve connectivity and name resolution problems: 1.
Version 2.2
1. Verify that the Torrent Server is configured correctly by reviewing the Torrent Server deployment instructions. 2. Find the IP address of the Torrent Server as described in #Verifying Torrent Server IP Address. If you cannot reach the Torrent Server an IP address, you are likely to need help from the site IT Administrator who understands how the local network is configured.
Verifying Torrent Server IP Address
The Torrent Server is configured out-of- the-box to automatically get an IP address from the DHCP server on the network. Unless the local IT Administrator has specifically assigned an IP address in advance, you will not know what the current IP address is. The Torrent Server has several Ethernet ports on the back. Make sure your site network is connected to the port labeled LAN, called eth0 in Linux terminology. The Ethernet port are identified as eth0, eth1, ..., for as many ports as are available. On Torrent Server, eth0 is the only port connected to your network and is configured by DHCP. To determine the IP address assigned to eth0, login and type: ifconfig eth0. This displays the following output: ionadmin@ion-torrent-server:~$ ifconfig eth0 eth0 Link encap:Ethernet HWaddr 00:1b:21:5b:bb:44 inet addr:192.168.1.123 Bcast:192.169.4.255 Mask:255.255.255.0 inet6 addr: fe80::21b:21ff:fe5b:bb44/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:209970726 errors:0 dropped:0 overruns:0 frame:0 TX packets:419252947 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:14131928595 (14.1 GB) TX bytes:607398487997 (607.3 GB) Memory:fbea0000-fbec0000 Your IP address is the inet addr: inet addr:192.168.1.1 Bcast:192.169.4.255 Mask:255.255.255.0 Another useful check is the line beginning with UP, which indicated the interface is active and working: UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 If the eth0 port is not available, it is possible the Ethernet cable is connected to a network, so you will not see the word UP: BROADCAST MULTICAST MTU:1500 Metric:1 If an IP address is assigned, the interface is likely to work. If no IP address is assigned and the interface is not UP, you may need to get help from your site IT Administrator. If you are still concerned about network connectivity, you can test that different desktops are able to successfully pi ng the server IP address. If you are not able to ping the server from the desktops that need to access the Torrent Browser running on the server, contact your site IT Administrator.
Troubleshooting and Configuring the Time Service
The Torrent Server uses the Linux Network Time Protocol (NTP) program to synchronize its time with another time
Version 2.2
server. By default, the Torrent Server is configured to synchronize its time service to a trusted time service on the Internet. This requires that the network configuration permits the NTP network protocol to connect to that time service on the Internet. The Torrent Server can also act as a time server for PGM sequencers. However, if the server is not able to synchronize with the trusted time service, it will not act as a time server for PGM sequencers (Torrent Server does not forward potentially incorrect information to other machines). If the network configuration is blocking the NTP protocol from reaching the Internet, the Torrent Server and PGM sequencers will not be able to synchronize time. Your site network administrator is probably aware of this connectivity restriction, and it is likely that IT has a time server in the network.
Changing the Time Server used by the Torrent Server
After the site IT administrator has identified the local time server, use the following procedure to update the Torrent Server to use the new time server: 1. Edit the NTP configuration file a. Edit /etc/ntp.conf (using the nano or vi text editor, for example). b. Change the line from server ntp.ubuntu.com to server <hostname>. 2. Restart the time service on the Torrent Server. a. Restart NTP, using the command: sudo /etc/init.d/ntp restart. When a time server is configured, you can also set the PGM equipment to get time from either the Torrent Server or the IT-provided time server.
Checking NTP Configuration on the Torrent Server
To test the NTP configuration on the Torrent Server, enter the following command on the Torrent Server command line: ntpq -p On a working server, the response looks similar to the following example: remote refid st t when poll reach delay offset jitter ============================================================= xxx.yyy 69.65.40.29 3 u 196 1024 377 0.175 14.235 0.9 On a server that cannot connect, the response looks similar to the following example: remote refid st t when poll reach delay offset jitter ============================================================= xxx.yyy .INIT. 16 u 64 }} {{0 0.000 0.000 0.000 Notice, for the server that cannot synchronize time, the following columns indicate an error condition: Column Value Meaning
Version 2.2
st
16
The server is not synchronized to another server. The server has never synchronized its time.
when
Also, the time stamps are all 0.0000. These usually show some deviation when operating typically. For a firewall to permit NTP access to the Internet, Port 123 (UDP) must be permitted, incoming and outgoing
Verifying PGM File Transfer
Verify that all files successfully transferred from the PGM sequencer to the Torrent Server. Manually transfer files, if needed, by going to the Data Management screen of the PGM sequencer and select the Re-transfer option for any of the runs in question. You can safely re-transfer data. Do not delete the data from the PGM sequencer until you are confident that the data is present on the Torrent Server, and the Analysis Report has been generated successfully.
Using QMON to Manage the SGE Cluster
QMON is a graphical tool to manage the SGE Cluster. With QMON you can do things like: see the status of the servers in the cluster check server CPU load and memory utilization see the status of jobs, are they running, completed, or still waiting in the queue disable servers from accepting new jobs, and re-enable them later delete jobs that are currently processing or waiting in queue and many other advanced configuration options SGE is typically used in a cluster of several servers to distribute data processing jobs. With the Torrent Server, SGE is used in the cluster configuration and also in the single-server configuration. This design makes it it easy for users to start with one server and add more "compute nodes," as needed, to ramp up the data processing capacity.
Client OS Support
A GUI is not installed on Torrent Server. You must start QMON, remotely, using ssh. To displayed QMON on your desktop, you need a client that supports the X Windows protocol; a Linux client automatically includes X Windows support. There are numerous software packages that add X Windows functionality to Windows, but the easiest and most economical solution is to run Linux in a virtual machine on a Windows system.
Starting QMON
To start QMON, login to the server with ssh, and include the -X option, this tells the server that you are going to be using an application that uses the X Windows display. ssh -X ionadmin@ion-torrent-server After you are logged into the server, just type QMON from the command line to launch the tool.
Version 2.2
ionadmin@ion-torrent-server:~$ qmon The server should be preconfigured to support QMON so you should not get any error message. The most common error is that you forgot the -X option, which causes the following error when you try to start QMON: ionadmin@ion-torrent-server:~$ qmon Error: Can't open display: Shortly after entering the QMON command, the "Grid Engine" splashscreen should appear, indicating your X Windows display is working. If you are on a slow connection, it may take a few seconds for the tool to launch. After a short time, the splashscreen will automatically go away.
When you see the following toolbar, QMON is ready to use:
There are many buttons here to access advanced options. Typically, only use the first two buttons on the upper-left: Job Control and Queue Control. To get started with QMON, view the information displayed in #Job Control and #Queue Control and observe your jobs as they are processed by SGE.
Job Control
The Job Control screen shows you the job status. The display does not automatically refresh, so remember to click Refresh to update the status.
Version 2.2
Tab Pending Jobs Running Jobs Finished Jobs
Description Jobs that are in the queue waiting to be process. Jobs currently being processed. Jobs that have completed processing.
You can delete jobs from this screen by highlighting the job then pressing the Delete button.
Queue Control
The Queue Control Screen shows the different servers that are available for processing jobs.
The used/total column shows the number of jobs currently running on the server (used), and the number of processing slots available on the server (total). This example shows seven processing slots available and six jobs currently being processed.
Version 2.2
There are four queues: tl.q for Top Level control jobs all.q for Analysis jobs plugin.q for Plugin jobs thumbnail.q for Proton Thumbnail Analysis jobs To show the status of all jobs, enter the command: qstat -f. For jobs indicating an error, where an "E" is displayed in the last column of the job status, clear the error using one of the following methods: Enter the command qmod -cq all.q on the command line. Login to QMON and, for any queues are marked with "E," highlight the entry and click the Clear Error button.
System Diagnostics
You can get system diagnostics information by entering the URL <yourTorrentServerName>/rundb/info/ in your browser address window. This displays the following dialog:
Click on the System Status Support Archive link to download a zip file of system logs from the /opt, /tmp, and / var directories.
Further Investigation and Problem Resolution
Version 2.2
After the root cause of a major problem is identified, the following more intrusive action may be needed: Replacing failed hard disk drive Downgrading software packages Reinstalling software Modifying config files Add/modify/delete database information Please contact Ion Torrent for assistance before trying any of these steps.
Version 2.2
Updating the IonReporterUploader Plugin

Version 2.2
Updating the IonReporterUploader Plugin

This page describes the procedure to manually update the IonReporterUploader module on your Torrent Server. This procedure is only for use if the IonReporterUploader version on your Torrent Serve is not compatible with the Io n Reporter server. If there is a mismatch, the plugin typically starts file transfer, but the transfer fails and the files are not available as samples in the Ion Reporter UI. The error appears in the plugin log on the Torrent Server (in the Torrent Browser Report tab, select the Run Report, scroll down to the Plugin Status Section, find the IonReporterUploader plugin, and and click on the More Detailed Information link).
Warning The procedure described in this section breaks a working IonReporterUploader plugin. Do not implement these procedures unless instructed to do so by the Torrent Suite or Ion Reporter rel ease notes or by Ion Reporter support.
Overview and requirements This list gives a summary of the steps to manually update the plugin: 1. Download the plugin tar ball from the Ion Reporter server. 2. On the Torrent Server, remove the plugin directory and replace it with files from the plugin tar ball. 3. Do one of the following with the authentication token: a. If the Ion Reporter server is already configured for token-based authentication, copy the existing token (and paste it into a text editor). This is the recommended option if the server already has a token configured. b. Configure the plugin on the Ion Reporter server for token-based authentication. 4. Configure the plugin on the Torrent Server with the authentication token. This procedure involves steps on both servers, and requires the following accounts: An Ion Reporter admin or user account to download the plugin module from the Ion Reporter server An Ion Reporter admin account to regenerate the authentication code in the Ion Reporter UI (not required if you copy the existing authentication token) An ionadmin account on the Torrent Server Linux host, to update the plugin
The Plugin Tar Ball This section describes how to download the plugin module from the Ion Reporter UI and install it on the Torrent Server host machine. 1. In Ion Reporter UI, go to the user name menu (at the top of the page, next to Welcome). 2. Click on Ion Reporter Plugin, and download the file IonreporterUploader.tar.gz. 3. If you are running your browser from the Torrent Server host machine, save the file to /results/plugins. Otherwise, save the file to your local machine and then copy it to the /results/plugins directory on the T
3.
Version 2.2
orrent Server host. 4. In you haven't already done so, log into the Torrent Server host with an ionadmin account. 5. Remove the current Uploader plugin directory. Example commands: cd /results/plugins /bin/rm rf ./IonReporterUploader
6. Still in the /results/plugins directory, untar the plugin module: tar xzf IonreporterUploader.tar.gz 7. Delete the tar ball: /bin/rm ./IonreporterUploader.tar.gz At this point, the plugin in installed, but is not operational until you configure it.
Copy an Existing Authentication Token Follow these steps to copy the existing authentication token. 1. Log into the Ion Reporter UI with an administrator account. 2. Open your profile, in the user name menu ("IR Admin" in this example):
3. In your profile popup, copy the contents of the Authentication Token field:
4. If you have a Torrent Browser open, copy the authentication key directly into the Uploader plugin configuration (described below). Or, copy the authentication token into a text editor or other application which does not add end-of-line characters, and distribute the token to the ionadmin.
Version 2.2
Do not click the Generate button. 5. Close your profile page. Configure the Ion Reporter Server for Token-Based Authentication Do not follow the instructions in this section if you copied the existing authentication token. These steps generate a new authentication token on the Ion Reporter side. Do not generate a new authentication token unless truly necessary. The existing token expires immediately, and any working plugin connections are broken. Any file transfers that are in progress fail. This expiration applies to any other Torrent Server configured with the authentication token from the same Ion Reporter account.
Follow these steps to generate the authentication token: 1. Log into the Ion Reporter UI with an administrator account. 2. Open your profile, in the user name menu ("IR Admin" in this example):
3. In your profile popup, click the Generate button:
4. If you have a Torrent Browser open, copy the authentication key directly into the Uploader plugin configuration (described below). Or, copy the authentication token into a text editor or other application which does not add end-of-line characters, and distribute the token to the ionadmin.
Configure the Uploader Plugin in the Torrent Browser This sections describes the last step in configuring the plugin. You configure the plugin in the Torrent Browser with the Ion Reporter authentication token. Follow these steps to configure your Ion Reporter Uploader plugin in your Torrent Browser:
Version 2.2
1. Log into the Torrent Browser as either an ionadmin or a regular user. 2. Go to the Config tab and scroll down to the plugin section. Find the entry for IonReporterUploader. 3. Click the Manage link for IonReporterUploader, and select Configure IonReporterUploader:
4. The Ion Reporter Uploader Configuration page opens:
5. Enter the authentication token from the Ion Reporter administrator in the authentication token field. IMPORTANT: Be careful not to add an extra space character before or after the token. 6. Click Save. 7. Enable the IonReporterUploader checkboxes for Enabled and Auto-Run:
The Uploader plugin is ready to transfer files.
Version 2.2
Version 2.2

TSUserDocs 2.2

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

TSUserDocs 2.2

Uploaded by

Copyright:

Available Formats

USER DOCUMENTATION

Torrent Suite 2.2

Revision Date May 2012

Torrent Suite 2.2

Revision Date May 2012

Technical Notes and Whitepapers

Technical Note - Analysis Pipeline

- Page 3 Copyright 2012 Life Technologies Corporation

- Page 4 Copyright 2012 Life Technologies Corporation

- Page 5 Copyright 2012 Life Technologies Corporation

Nuc Flow: TACGTACG Library: 1010010 Test Frag: 0100101

- Page 6 Copyright 2012 Life Technologies Corporation

- Page 7 Copyright 2012 Life Technologies Corporation

- Page 8 Copyright 2012 Life Technologies Corporation

Read Filtering and Trimming

- Page 9 Copyright 2012 Life Technologies Corporation

Technical Note - Filtering and Trimming

- Page 11 Copyright 2012 Life Technologies Corporation

- Page 12 Copyright 2012 Life Technologies Corporation

- Page 13 Copyright 2012 Life Technologies Corporation

Technical Note - Quality Score

Technical Note - TMAP Alignment

- Page 16 Copyright 2012 Life Technologies Corporation

Molecular Biology, 147(1), 195 197.

White Paper - Torrent Variant Detection Algorithms

Torrent Variant Detection Algorithms

SNP Detection - DiBayes Algorithm

- Page 18 Copyright 2012 Life Technologies Corporation

- Page 19 Copyright 2012 Life Technologies Corporation

Indel Detection - Variant Hunter Algorithm

- Page 20 Copyright 2012 Life Technologies Corporation

- Page 21 Copyright 2012 Life Technologies Corporation

- Page 22 Copyright 2012 Life Technologies Corporation

Torrent Variant Caller 2.0 Workflow

- Page 24 Copyright 2012 Life Technologies Corporation

- Page 25 Copyright 2012 Life Technologies Corporation

Resulting File Type ---bfmask.bin DAT

Ion 318 Chip 520 81% 4994815 25 MB 232 GB

Ion 316 Chip 520 70% 2167130 14 MB 126 GB

Ion 314 Chip 520 71% 660938 2.9 MB 41 GB

- Page 26 Copyright 2012 Life Technologies Corporation

- Page 27 Copyright 2012 Life Technologies Corporation

- Page 28 Copyright 2012 Life Technologies Corporation

Manually Starting an Analysis

- Page 29 Copyright 2012 Life Technologies Corporation

2. Click the Advanced button:

See Working with Runs for a description of these arguments.

Stage Well Characterization

A run stage has the following color code legend:

- Page 31 Copyright 2012 Life Technologies Corporation

Color Green Yellow Gray

Meaning Complete Active processing Pending

- Page 32 Copyright 2012 Life Technologies Corporation

Torrent Variant Caller

Viewing Runs and Reports

- Page 33 Copyright 2012 Life Technologies Corporation

Viewing Runs and Reports

- Page 34 Copyright 2012 Life Technologies Corporation

- Page 35 Copyright 2012 Life Technologies Corporation

Enter a run/report text string to search for a run or report by name:

Viewing the Preview Report

- Page 36 Copyright 2012 Life Technologies Corporation