Journal of Biotechnology 261 (2017) 137–141

Contents lists available at ScienceDirect

Journal of Biotechnology
journal homepage: www.elsevier.com/locate/jbiotec

Review

Bioinformatics can boost metabolomics research MARK

a a a a,b,⁎
René Meier , Christoph Ruttkies , Hendrik Treutler , Steffen Neumann
a
Leibniz Institute of Plant Biochemistry, Department of Stress and Developmental Biology, Weinberg 3, 06120 Halle, Germany
b
German Centre for Integrative Biodiversity Research (iDiv), Halle-Jena-Leipzig Deutscher, Platz 5e, 04103 Leipzig, Germany

A R T I C L E I N F O A B S T R A C T

Keywords: Metabolomics is the modern term for the field of small molecule research in biology and biochemistry. Currently,
Metabolomics metabolomics is undergoing a transition where the classic analytical chemistry is combined with modern
Mass spectrometry cheminformatics and bioinformatics methods, paving the way for large-scale data analysis. We give some
Metabolite profiling background on past developments, highlight current state-of-the-art approaches, and give a perspective on
Metabolite identification
future requirements.
Workflows
Cloud computing

1. Introduction
Research in the modern life-sciences aims at the understanding of
living organisms, where all processes between the genome and the
phenotype are of interest. The subject of studies include gene regula-
tion, protein synthesis, their post-translational modifications, and the
biochemistry of proteins and small molecules – metabolites.
Metabolomics is the modern term for the field of small molecule
analytical chemistry in biology and biomedical research, but the
underlying questions have been addressed already for hundreds of
years by physicians using the smell and colour of, e.g. urine for health
diagnosis. In 1971, Pauling et al. (1971) analysed more than 200
metabolites in breath and urine headspace, but the terms “metabolo-
mics” or “metabonomics” only appeared in the scientific literature more
than 25 years later (Oliver et al., 1998; Nicholson et al., 1999). In the
last two decades, a lot of progress has been made in the number of
metabolites that can be (simultaneously) detected, lowering of the
limits of detection with modern analytical technologies, and the
increased throughput of samples that can be processed.
2. Scaling up mass spectrometry based metabolomics
Today, mass spectrometry is a key technology for metabolomics
research. Due to immense technological advances in mass spectrometry
over the last years, the amount and complexity of the data produced has
been growing rapidly. These advances would not have been possible
without the extensive use of computers throughout the data processing
and analysis steps of the experiments.
While the first mass spectrometers used photo platters to record
spectra, the computer became an integral part of the instruments under
the term “data system” already in the 1970s. The rapid digital recording
of mass spectra also allowed to couple the MS instruments to
chromatographic separation, such as liquid or gas chromatography.
These separation processes greatly reduce the complexity of the
individual mass spectra, which in turn allows to measure more complex
samples, such as full methanolic extracts of plants or human body
fluids. The advent of ion mobility further increases the separation
power, but also increases the run time per sample.
The amount of data from raw spectra in LC/MS measurements is
overwhelming, hence a feature detection step is typically applied to
extract the chromatographic and spectral peaks into so called feature
lists. These feature lists can be used as metabolic fingerprints, which
represent a molecular phenotype. Typical metabolomics experimental
designs include the comparison of different genotypes, perform inter-
vention studies and time-series experiments. These setups require the
processing of dozens to hundreds, even thousands of samples. With
microarrays it is possible to quantify the abundance of RNA of specified
(short) sequences and directly compare the gene expression across
samples. In LC/MS and GC/MS however, the feature lists need to be
matched across samples, and both chromatographic shifts and mass
deviations have to be considered or even compensated for. Several
academic software packages have been developed in the past years,
with MetAlign being one of the first tools, initially developed for GC/
MS processing, and later also adapted for high-resolution LC/MS data
(Tikunov et al., 2005; Lommen and Kools, 2012). Now, both the XCMS
package (Smith et al., 2006), developed as part of the Bioconductor
project (Gentleman et al., 2003), and the OpenMS (Sturm et al., 2008;
Röst et al., 2016) framework are celebrating their tenth anniversary.

⁎
Corresponding author.
E-mail address: sneumann@ipb-halle.de (S. Neumann).

http://dx.doi.org/10.1016/j.jbiotec.2017.05.018
Received 18 February 2017; Received in revised form 10 May 2017; Accepted 16 May 2017
Available online 26 May 2017
0168-1656/ © 2017 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/BY/4.0/).
R. Meier et al.
While previously experimental analysis comprised comparing just,
say, two measurements, these new paradigms automate the data
processing steps and allow the use of statistical analysis, giving
confidence values for the discovery and interpretation of, e.g. biomar-
kers. An example is the study conducted by Thévenot et al. (2015),
where more than 200 samples were processed with an advanced
workflow comprising preprocessing, signal drift correction and batch
effect removal, and uni- and multivariate statistics. In addition to the
huge number of R packages available for statistics and modelling, user-
friendly and advanced data analysis tools exist in the form of, e.g.
MetaboAnalyst (Xia et al., 2015) or MetExplore (Cottret et al., 2010).
The feature detection step is followed by the annotation of the
molecular structures, which are required for the biochemical inter-
pretation. A main advantage here is that mass spectrometry is
independent of the availability of the genome sequence, and can be
applied to any organism and tissue type. On the other hand, both
analytical limitations and the chemical diversity of metabolites, bio-
chemical processes, and external influences prevent that all possible
molecular structures present in an organism are known a priori.
Thus, metabolite identification is an important task in computa-
tional metabolomics. For several organisms, including human and
model organisms such as Arabidopsis thaliana, metabolite databases
have been established (Wishart et al., 2013; Mueller et al., 2003). If the
compound is assumed to be known in these databases, it will be
returned with a rather simple search for metabolites having a mass
within an instrument-dependent error window. However, all com-
pounds with a similar mass (and of course all with the same molecular
formula) will be retrieved as false positive hits. Their number can be
reduced if the molecular formula itself can be deduced from the
accurate mass, isotopic pattern and further hints. Generic chemical
databases like PubChem or ChemSpider (Wang et al., 2009; Pence and
Williams, 2010) contain orders of magnitudes more structures, but also
contain many, if not the majority, of compounds usually not relevant to
the experimental question.
3. Metabolite identification
Complementing to GC/MS and LC/MS, more structural information
is available from higher-order mass spectra, such as tandem MS or MSn.
Here the analyte ions undergo fragmentation in the instrument, and the
fragmentation spectra provide a fingerprint of the molecular structure.
Those spectra can be compared against reference data to identify the
metabolite, but require that the compounds are available and reference
data have been deposited in, e.g., MassBank (Horai et al., 2010), HMDB
(Wishart et al., 2007), and METLIN (Smith et al., 2005), or one of the
commercial offerings. An overview of spectral libraries and comparison
of their chemical coverage can be found in Vinaixa et al. (2016).
While mass spectral libraries are growing for tandem MS or MSn
spectra, the coverage is still relatively small compared to the number of
compounds that could potentially be present in typical samples.
Especially if no reference data are available, the spectra previously
have been interpreted manually, and structural hints constrained the set
of possible molecular structures.
One of the aims in computational mass spectrometry is to fill this
gap by proposing tentative identifications for unknown compounds.
Different approaches exist for the identification process. Besides the
prediction of molecular properties from the spectral information by rule
based expert knowledge or by machine learning approaches and their
matching with candidates from compound databases it is possible to
reproduce the process of fragmentation in silico.
MetFrag, launched in 2010, was one of the first approaches to
address this problem for accurate tandem mass spectra for hundreds of
candidate structures from chemical databases (Wolf et al., 2010).
MetFusion combines in silico fragmentation with spectral similarity
search in MassBank (Horai et al., 2010) and has shown to be an
excellent way to benefit from two different information sources.
138
Journal of Biotechnology 261 (2017) 137–141
Compound identification is a time-consuming task as it requires the
look-up and combination of many different information sources to
collect as many evidences as possible to support a putative identifica-
tion. Especially for high-throughput analysis where hundreds of
metabolites need to be identified the workload for analysts is notably
high. Therefore, MetFrag was further enhanced to provide a methodo-
logical interface to query different databases and combine the informa-
tion drawn into the identification process. These new functions greatly
reduce the burden on users to collect and merge ever increasing
amounts of information available for substances present in different
compound databases, thus enabling them to consider more evidence.
With parallelizing the MetFrag analysis the processing of hundreds of
tandem MS spectra is performed within minutes and the addition of
different information types has shown to improve identification rates
from 6% up to 70% depending on the dataset and information sources
used (Ruttkies et al., 2016).
The usage of computational approaches for the identification of
metabolites has been shown in several studies (McMillan et al., 2016;
Van Meulebroek et al., 2016; Narduzzi et al., 2015). So far, results from
in silico methods alone are not sufficient to count as full identification
and additional validation in the lab, preferably with an authentic
standard, is required. However, using these tools augmented with the
experimental context and additional meta data reduces the workload by
providing high quality putative metabolite annotations.
With untargeted MS measurements from modern MS instruments it
is possible to measure thousands of features along with their fragment
spectra. Hence, the manual inspection and identification of these data
sets is no longer an option. Recent developments aim at relating
hundreds or thousands of unidentified features by spectral similarity.
This approach results in clusters of biochemically related metabolites
offering a bird's eye view on all feature classes in a sample. Molecular
networking implements this idea by using a graph with features as
nodes which are connected by edges if the spectral similarity is above a
certain threshold (Watrous et al., 2012). This allows the quick inspec-
tion of classes of metabolites with many members and has been
demonstrated on data from different organisms. MS2LDA follows a
different strategy by identifying patterns in fragment spectra as
fingerprints of chemical substructures (van der Hooft et al., 2016).
This allows the decomposition of fragment spectra and the assignment
of fragments to certain parts of the molecule. MetFamily performs a
hierarchical clustering of features based on spectral similarity com-
bined with principal component analysis of MS abundances (Treutler
et al., 2016). This allows the discovery of biochemically related features
with regulated behaviour under different conditions called regulated
metabolite families.
4. Reproducible and shareable research
The scientific discourse through letters among researchers and
articles in scientific journals has a long history going back centuries,
but electronic data publications have emerged only in the last few
years. The amounts of data recorded in the life sciences mandate that
data is available and enriched with experimental metadata. In meta-
bolomics, public repositories are available for several years now: The
NIH funded Metabolomics Workbench (Sud et al., 2016) and the
European MetaboLights (Haug et al., 2013) hosted at EMBL-EBI.
Even though various analysis tools exist to tackle these problems a
standardization of the computational workflows is hardly implemented.
The community's need for standardization becomes even more obvious
when realizing that even today the reproducibility of results available
in peer-reviewed publications is not always possible. Besides the
experimental conditions, making computational analysis pipelines
reproducible with standardization is a first step to solve this issue
(Sandve et al., 2013).
Especially in metabolomics, the large number of samples and
features in the experimental results rises the need for unattended data
R. Meier et al.
Fig. 1. Inclusion of MetFrag as an independent node for metabolite identification into KNIME.
processing. Workflow or pipeline tools are visual programming lan-
guages that enable the average biologist or bio-medical researcher to
apply state-of-the-art algorithms and data analysis to big data sets. They
are already widespread in commercial data mining, but also in scientific
areas such as pharmaceutical research or genomics.
Proteomics tools from the OpenMS package have found their way
into the two most wide-spread workflow-systems in bioinformatics,
KNIME (Berthold et al., 2007) and Galaxy (Afgan et al., 2016), and
prototypes exist for the integration of MetFrag (fig. 1). Galaxy has also
been used as workflow engine to integrate other proteomics tools
(Sheynkman et al., 2014), and for metabolomics applications in Galaxy,
tool integration and workflows are being addressed by the Work-
flow4Metabolomics team (Giacomoni et al., 2015) and at the University
of Birmingham (Davidson et al., 2016).
Building on these efforts, the H2020 project PhenoMeNal is devel-
oping a standardized e-infrastructure that will support the data proces-
sing and analysis pipelines for molecular phenotype data generated by
metabolomics applications. Virtualization and containerization techni-
ques are easy ways to share workflows with researchers where the
whole setup can be shipped with one or several containers to another
computer. Time-consuming tasks can be run on cloud infrastructures
providing enough compute resources without manually installing
additional software. With the arise of the de.NBI cloud infrastructure
it becomes easier to run and develop containerized metabolomics
workflows.
5. Reaching beyond metabolomics
Notably, many of the challenges described here for metabolomics
also apply to other – seemingly unrelated – disciplines. One task in
environmental research is the monitoring of water quality (which
requires the profiling or comparison of samples across sites, time, and
in response to water treatment) and the identification of unknown
contaminants (Altenburger et al., 2015; Wölz et al., 2010). Similarly, in
food authenticity control, samples can be screened for their biochemical
characteristics. These and other areas will benefit from progress
139
Journal of Biotechnology 261 (2017) 137–141
described here as well (Gómez-Romero et al., 2010; Chong et al., 2013).
6. Training in bioinformatics for metabolomics
As evident from the above description, the data analysis in
metabolomics is a highly interdisciplinary and challenging task,
requiring a broad range of expertises, which are in general not covered
by university programs in biology, chemistry, nor bioinformatics.
Several training courses have been organized in the past, and similar
ones will be held in the future, including many OpenMS courses at both
bioinformatics and mass spectrometry conferences, or the metabolo-
mics workshops at the Leibniz Institute of Plant Biochemistry.
Internationally, several courses supported by EMBO have been orga-
nized at the EMBL-EBI, other courses are regularly given by the
Phenome centre at Imperial College in London, and in 2016 the
European Metabolomics Training Coordination Group (EmTraG) was
formed.
7. Conclusion
The de.NBI network is covering different aspects of metabolomics in
several of its service centres, including the Center for Integrative
Bioinformatics (CIBI). With the advent of the de.NBI cloud, the data
management and processing of even large studies with many samples
can be handled.
Bio- and cheminformatics have advanced tremendously in the last
years, and will certainly develop further in the future. Future develop-
ments will include more robust algorithms and the estimation of
parameters from input data. For high-throughput data analysis, less
human intervention will be possible. Gone are the days in high-
throughput sequencing where trace files or Illumina images were stored
just in case, through quality control we know how far we trust the
resulting sequences. Similarly, we need to work towards high-through-
put metabolomics data analysis, adding confidence intervals and
probabilities along the pipeline. For sure, there are enough challenges
to be solved.
R. Meier et al.
Acknowledgements
RM, SN, and HT acknowledge BMBF funding under grant number
031L0107, CR acknowledges European Commission PhenoMeNal Grant
EC654241. SN acknowledges basic institutional funding by the Leibniz
Association.
References
Afgan, E., Baker, D., van den Beek, M., Blankenberg, D., Bouvier, D., Cech, M., Chilton, J.,
Clements, D., Coraor, N., Eberhard, C., Grüning, B., Guerler, A., Hillman-Jackson, J.,
Von Kuster, G., Rasche, E., Soranzo, N., Turaga, N., Taylor, J., Nekrutenko, A.,
Goecks, J., 2016. The galaxy platform for accessible, reproducible and collaborative
biomedical analyses: 2016 update. Nucleic Acids Res. 44 (W1), W3. http://dx.doi.
org/10.1093/nar/gkw343. http://arxiv.org/abs//oup/backfile/Content_public/
Journal/nar/44/W1/10.1093_nar_gkw343/3/gkw343.pdf.
Altenburger, R., Ait-Aissa, S., Antczak, P., Backhaus, T., Barceló, D., Seiler, T.-B., Brion,
F., Busch, W., Chipman, K., de Alda, M.L., de Aragão Umbuzeiro, G., Escher, B.I.,
Falciani, F., Faust, M., Focks, A., Hilscherova, K., Hollender, J., Hollert, H., Jäger, F.,
Jahnke, A., Kortenkamp, A., Krauss, M., Lemkine, G.F., Munthe, J., Neumann, S.,
Schymanski, E.L., Scrimshaw, M., Segner, H., Slobodnik, J., Smedes, F., Kughathas,
S., Teodorovic, I., Tindall, A.J., Tollefsen, K.E., Walz, K.-H., Williams, T.D., den Brink,
P.J.V., van Gils, J., Vrana, B., Zhang, X., Brack, W., 2015. Future water quality
monitoring – adapting tools to deal with mixtures of pollutants in water resource
management. Sci. Total Environ. 512–513, 540–551. http://dx.doi.org/10.1016/j.
scitotenv.2014.12.057. http://www.sciencedirect.com/science/article/pii/
S0048969714017598.
Berthold, M.R., Cebron, N., Dill, F., Gabriel, T.R., Kötter, T., Meinl, T., Ohl, P., Sieb, C.,
Thiel, K., Wiswedel, B., 2007. KNIME: The Konstanz Information Miner. Studies in
Classification, Data Analysis, and Knowledge Organization (GfKL 2007).
Springerhttp://dx.doi.org/10.1007/978-3-540-78246-9_38.
Chong, E.S.L., McGhie, T.K., Heyes, J.A., Stowell, K.M., 2013. Metabolite profiling and
quantification of phytochemicals in potato extracts using ultra-high-performance
liquid chromatography–mass spectrometry. J. Sci. Food Agric. 93 (15), 3801–3808.
http://dx.doi.org/10.1002/jsfa.6285.
Cottret, L., Wildridge, D., Vinson, F., Barrett, M.P., Charles, H., Sagot, M.-F., Jourdan, F.,
2010. MetExplore: a web server to link metabolomic experiments and genome-scale
metabolic networks. Nucleic Acids Res. 38 (suppl 2), W132–W137. http://dx.doi.
org/10.1093/nar/gkq312.
Davidson, R.L., Weber, R.J.M., Liu, H., Sharma-Oates, A., Viant, M.R., 2016. Galaxy-m: a
galaxy workflow for processing and analyzing direct infusion and liquid
chromatography mass spectrometry-based metabolomics data. GigaScience 5, 10.
http://dx.doi.org/10.1186/s13742-016-0115-8.
Gómez-Romero, M., Segura-Carretero, A., Fernández-Gutiérrez, A., 2010. Metabolite
profiling and quantification of phenolic compounds in methanol extracts of tomato
fruit. Phytochemistry 71 (16), 1848–1864. http://dx.doi.org/10.1016/j.phytochem.
2010.08.002. http://www.sciencedirect.com/science/article/pii/
S003194221000302X.
Gentleman, Rossini, Dudoit, Hornik, 2003. The Bioconductor FAQ. http://www.
bioconductor.org.
Giacomoni, F., Le Corguillé, G., Monsoor, M., Landi, M., Pericard, P., Pétéra, M., Duperier,
C., Tremblay-Franco, M., Martin, J.-F., Jacob, D., Goulitquer, S., Thévenot, E.A.,
Caron, C., 2015. Workflow4metabolomics: a collaborative research infrastructure for
computational metabolomics. Bioinformatics 31 (9), 1493. http://dx.doi.org/10.
1093/bioinformatics/btu813. http://arxiv.org/abs//oup/backfile/Content_public/
Journal/bioinformatics/31/9/10.1093/bioinformatics/btu813/2/btu813.pdf.
Haug, K., Salek, R.M., Conesa, P., Hastings, J., de Matos, P., Rijnbeek, M., Mahendraker,
T., Williams, M., Neumann, S., Rocca-Serra, P., Maguire, E., González-Beltrán, A.,
Sansone, S.-A., Griffin, J.L., Steinbeck, C., 2013. MetaboLights – an open-access
general-purpose repository for metabolomics studies and associated meta-data.
Nucleic Acids Res. 41 (Database issue), D781–D786. http://dx.doi.org/10.1093/nar/
gks1004.
Horai, H., Arita, M., Kanaya, S., Nihei, Y., Ikeda, T., Suwa, K., Ojima, Y., Tanaka, K.,
Tanaka, S., Aoshima, K., Oda, Y., Kakazu, Y., Kusano, M., Tohge, T., Matsuda, F.,
Sawada, Y., Hirai, M.Y., Nakanishi, H., Ikeda, K., Akimoto, N., Maoka, T., Takahashi,
H., Ara, T., Sakurai, N., Suzuki, H., Shibata, D., Neumann, S., Iida, T., Tanaka, K.,
Funatsu, K., Matsuura, F., Soga, T., Taguchi, R., Saito, K., Nishioka, T., 2010.
MassBank: a public repository for sharing mass spectral data for life sciences. J. Mass
Spectrom. 45 (7), 703–714. http://dx.doi.org/10.1002/jms.1777.
Lommen, A., Kools, H.J., 2012. Metalign 3.0: performance enhancement by efficient use
of advances in computer hardware. Metabolomics 8, 719–726. http://dx.doi.org/10.
1007/s11306-011-0369-1.
McMillan, A., Orimadegun, A.E., Sumarah, M.W., Renaud, J., da Encarnacao, M.M.,
Gloor, G.B., Akinyinka, O.O., Reid, G., Allen, S.J., 2016. Metabolic derangements
identified through untargeted metabolomics in a cross-sectional study of Nigerian
children with severe acute malnutrition. Metabolomics 13 (2), 13. http://dx.doi.org/
10.1007/s11306-016-1150-2.
Mueller, L.A., Zhang, P., Rhee, S.Y., 2003. AraCyc: a biochemical pathway database for
Arabidopsis. Plant Physiol. 2, 453–460. http://dx.doi.org/10.1104/pp.102.017236.
http://arxiv.org/abs/http://www.plantphysiol.org/cgi/reprint/132/2/453.pdf,
http://www.plantphysiol.org/cgi/content/abstract/132/2/453.
Narduzzi, L., Stanstrup, J., Mattivi, F., 2015. Comparing wild American grapes with Vitis
vinifera: a metabolomics study of grape composition. J. Agric. Food Chem. 63 (30),
140
Journal of Biotechnology 261 (2017) 137–141
6823–6834. http://dx.doi.org/10.1021/acs.jafc.5b01999. PMID: 26158394.
Nicholson, J.K., Lindon, J.C., Holmes, E., 1999. Metabonomics’: understanding the
metabolic responses of living systems to pathophysiological stimuli via multivariate
statistical analysis of biological NMR spectroscopic data. Xenobiotica 29 (11),
1181–1189. http://dx.doi.org/10.1080/004982599238047.
Oliver, S.G., Winson, M.K., Kell, D.B., Baganz, F., 1998. Systematic functional analysis of
the yeast genome. Trends Biotechnol. 16 (9), 373–378. http://dx.doi.org/10.1016/
S0167-7799(98)01214-1.
Pauling, L., Robinson, A.B., Teranishi, R., Cary, P., 1971. Quantitative analysis of urine
vapor and breath by gas–liquid partition chromatography. Proc. Natl. Acad. Sci.
U.S.A. 68 (10), 2374–2376.
Pence, H.E., Williams, A., 2010. ChemSpider: an online chemical information resource. J.
Chem. Educ. 87 (11), 1123–1124. http://dx.doi.org/10.1021/ed100697w.
Röst, H.L., Sachsenberg, T., Aiche, S., Bielow, C., Weisser, H., Aicheler, F., Andreotti, S.,
Ehrlich, H.-C., Gutenbrunner, P., Kenar, E., Liang, X., Nahnsen, S., Nilse, L., Pfeuffer,
J., Rosenberger, G., Rurik, M., Schmitt, U., Veit, J., Walzer, M., Wojnar, D., Wolski,
W.E., Schilling, O., Choudhary, J.S., Malmström, L., Aebersold, R., Reinert, K.,
Kohlbacher, O., 2016. Openms: a flexible open-source software platform for mass
spectrometry data analysis. Nat. Methods 13, 741–748. http://dx.doi.org/10.1038/
nmeth.3959.
Ruttkies, C., Schymanski, E.L., Wolf, S., Hollender, J., Neumann, S., 2016. MetFrag
relaunched: incorporating strategies beyond in silico fragmentation. J. Cheminform.
8, 3. http://dx.doi.org/10.1186/s13321-016-0115-9.
Sandve, G.K., Nekrutenko, A., Taylor, J., Hovig, E., 2013. Ten simple rules for
reproducible computational research. PLOS Comput. Biol. 9 (10), 1–4. http://dx.doi.
org/10.1371/journal.pcbi.1003285.
Sheynkman, G.M., Johnson, J.E., Jagtap, P.D., Shortreed, M.R., Onsongo, G., Frey, B.L.,
Griffin, T.J., Smith, L.M., 2014. Using galaxy-p to leverage rna-seq for the discovery
of novel protein variations. BMC Genomics 15, 703. http://dx.doi.org/10.1186/
1471-2164-15-703.
Smith, C.A., Maille, G.O., Want, E.J., Qin, C., Trauger, S.A., Brandon, T.R., Custodio, D.E.,
Abagyan, R., Siuzdak, G., 2005. METLIN: a metabolite mass spectral database. In:
Proceedings of the 9th International Congress of Therapeutic Drug Monitoring and
Clinical Toxicology, vol. 27. Louisville, Kentucky. pp. 747–751. http://dx.doi.org/10.
1097/01.ftd.0000179845.53213.39.
Smith, C., Want, E., O’Maille, G., Abagyan, R., Siuzdak, G., XCMS:, 2006. Processing mass
spectrometry data for metabolite profiling using nonlinear peak alignment, matching
and identification. Anal. Chem. 78 (3), 779–787. http://dx.doi.org/10.1021/
ac051437y.
Sturm, M., Bertsch, A., Gröpl, C., Hildebrandt, A., Hussong, R., Lange, E., Pfeifer, N.,
Schulz-Trieglaff, O., Zerck, A., Reinert, K., Kohlbacher, O., 2008. OpenMS – an open-
source software framework for mass spectrometry. BMC Bioinformatics 9, 163.
http://dx.doi.org/10.1186/1471-2105-9-163.
Sud, M., Fahy, E., Cotter, D., Azam, K., Vadivelu, I., Burant, C., Edison, A., Fiehn, O.,
Higashi, R., Nair, K.S., Sumner, S., Subramaniam, S., 2016. Metabolomics
workbench: an international repository for metabolomics data and metadata,
metabolite standards, protocols, tutorials and training, and analysis tools. Nucleic
Acids Res. 44, D463–D470. http://dx.doi.org/10.1093/nar/gkv1042.
Thévenot, E.A., Roux, A., Xu, Y., Ezan, E., Junot, C., 2015. Analysis of the human adult
urinary metabolome variations with age, body mass index, and gender by
implementing a comprehensive workflow for univariate and opls statistical analyses.
J. Proteome Res. 14, 3322–3335. http://dx.doi.org/10.1021/acs.jproteome.
5b00354.
Tikunov, Y., Lommen, A., Vos, C.d., Verhoeven, H., Bino, R., Hall, R., Bovy, A., 2005. A
novel approach for nontargeted data analysis for metabolomics: large-scale profiling
of tomato fruit volatiles. Plant Physiol. 139 (3), 1125–1137. http://dx.doi.org/10.
1104/pp.105.068130.
Treutler, H., Tsugawa, H., Porzel, A., Gorzolka, K., Tissier, A., Neumann, S., Balcke, G.U.,
2016. Discovering regulated metabolite families in untargeted metabolomics studies.
Anal. Chem. 88 (16), 8082–8090. http://dx.doi.org/10.1021/acs.analchem.
6b01569.
van der Hooft, J.J.J., Wandy, J., Barrett, M.P., Burgess, K.E.V., Rogers, S., 2016. Topic
modeling for untargeted substructure exploration in metabolomics. Proc. Natl. Acad.
Sci. U.S.A. 113 (48), 13738–13743. http://dx.doi.org/10.1073/pnas.1608041113.
Van Meulebroek, L., Hanssens, J., Steppe, K., Vanhaecke, L., 2016. Metabolic
fingerprinting to assess the impact of salinity on carotenoid content in developing
tomato fruits. Int. J. Mol. Sci. 17 (6), 821. http://dx.doi.org/10.3390/ijms17060821.
http://www.mdpi.com/1422-0067/17/6/821.
Vinaixa, M., Schymanski, E.L., Neumann, S., Navarro, M., Salek, R.M., Yanes, O., 2016.
Mass spectral databases for LC/MS- and GC/MS-based metabolomics: state of the
field and future prospects. TrAC Trends Anal. Chem. 78, 23–35. http://dx.doi.org/10.
1016/j.trac.2015.09.005.
Wölz, J., Brack, W., Moehlenkamp, C., Claus, E., Braunbeck, T., Hollert, H., 2010. Effect-
directed analysis of ah receptor-mediated activities caused by PAHs in suspended
particulate matter sampled in flood events. Sci. Total Environ. 408 (16), 3327–3333.
http://dx.doi.org/10.1016/j.scitotenv.2010.03.029. http://www.sciencedirect.com/
science/article/pii/S0048969710003001.
Wang, Y., Xiao, J., Suzek, T.O., Zhang, J., Wang, J., Bryant, S.H., 2009. PubChem: a
public information system for analyzing bioactivities of small molecules. Nucleic
Acids Res. 37 (Web Server Issue), W623–W633. http://dx.doi.org/10.1093/nar/
gkp456.
Watrous, J., Roach, P., Alexandrov, T., Heath, B.S., Yang, J.Y., Kersten, R.D., van der
Voort, M., Pogliano, K., Gross, H., Raaijmakers, J.M., Moore, B.S., Laskin, J.,
Bandeira, N., Dorrestein, P.C., 2012. Mass spectral molecular networking of living
microbial colonies. Proc. Natl. Acad. Sci. U.S.A. 109 (26), E1743–E1752. http://dx.
doi.org/10.1073/pnas.1203689109.
R. Meier et al.
Wishart, D.S., Tzur, D., Knox, C., Eisner, R., Guo, A.C., Young, N., Cheng, D., Jewell, K.,
Arndt, D., Sawhney, S., Fung, C., Nikolai, L., Lewis, M., Coutouly, M.-A., Forsythe, I.,
Tang, P., Shrivastava, S., Jeroncic, K., Stothard, P., Amegbey, G., Block, D., Hau, D.D.,
Wagner, J., Miniaci, J., Clements, M., Gebremedhin, M., Guo, N., Zhang, Y., Duggan,
G.E., MacInnis, G.D., Weljie, A.M., Dowlatabadi, R., Bamforth, F., Clive, D., Greiner,
R., Li, L., Marrie, T., Sykes, B.D., Vogel, H.J., Querengesser, L., 2007. HMDB: the
human metabolome database. Nucleic Acids Res. 35 (suppl 1), D521–526. http://dx.
doi.org/10.1093/nar/gkl923.
Wishart, D.S., Jewison, T., Guo, A.C., Wilson, M., Knox, C., Liu, Y., Djoumbou, Y., Mandal,
R., Aziat, F., Dong, E., Bouatra, S., Sinelnikov, I., Arndt, D., Xia, J., Liu, P., Yallou, F.,
141
Journal of Biotechnology 261 (2017) 137–141
Bjorndahl, T., Perez-Pineiro, R., Eisner, R., Allen, F., Neveu, V., Greiner, R., Scalbert,
A., 2013. HMDB 3.0 – the human metabolome database in 2013. Nucleic Acids Res.
41 (D1), D801–D807. http://dx.doi.org/10.1093/nar/gks1065.
Wolf, S., Schmidt, S., Müller-Hannemann, M., Neumann, S., 2010. In silico fragmentation
for computer assisted identification of metabolite mass spectra. BMC Bioinformatics
11 (1), 148. http://dx.doi.org/10.1186/1471-2105-11-148. http://www.
biomedcentral.com/1471-2105/11/148.
Xia, J., Sinelnikov, I.V., Han, B., Wishart, D.S., 2015. Metaboanalyst 3.0-making
metabolomics more meaningful. Nucleic Acids Res. 43, W251–W257. http://dx.doi.
org/10.1093/nar/gkv380.