Professional Documents
Culture Documents
Real-time PCR
4000 3000
Ct (threshold cycle) is the intersection 1000e + 000 measurements associated with the
3500 2500
log (Rn)
Rn
Rn
2500 1500
2000
relative measure of the concentration
1000
1000e - 002 Therefore, the Ct values from PCR
1500
of target
0500
in the PCR reaction.
Threshold Many
Exponential phase reactions run under different conditions
Baseline
factors impact the absolute value
1000e - 003
3500
Cycle number the target. WeCycle
1000e + 001
will number
discuss the most Cycle number
3000
common template-independent factors The effect of master mix
2500
that can influence Ct and describe how
1000e + 000
components
Threshold
2000 to evaluate the performance of a real-
1000e - 001
The fluorescence emission of any
log (Rn)
1500
1000
1000e - 002
environmental factors such as pH and
0500 Threshold Exponential phase
Figure 1 shows several parameters
1000e - 003
salt concentration. Figure 2 shows the
0000
t
of the real-time reaction amplification raw fluorescence data of an Applied
2930 31 32 3334 35 36 3738 39 40
-0500
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 212223 24 2526 2728 2930 31 32 3334 35 36 3738 39 40 plot. The exponential phase in
1000e - 004
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 212223 24 2526 2728 2930 31 32 3334 35 36 3738 39 40 Biosystems TaqMan probe in the
Cycle number Cycle number
1000e + 001 Figure 1B corresponds to the linear background of two different master
phase in Figure 1C. The threshold mixes. Note that the fluorescence
must be set in the linear phase of the intensity is higher in master mix A even
1000e + 000
Threshold
1000e - 001 amplification plot in Figure 1C. The though the target, probe, and Applied
log (Rn)
1000e - 004
1 32 3334 35 36 3738 39 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 212223 24 2526 2728 2930 31 32 3334 35 36 3738 39 40
Cycle number
the baseline fluorescence signals, in as a different Ct value. The new Ct Master mix A
Master mix B
different for the two master mixes ROX dye has no bearing on the true 4.00 E + 3
(Figure 3A). Variations in the Ct value sensitivity of the reaction, but can have
Amplitude
do not reflect the overall performance other unintended consequences. Low 3.00 E + 3
VIC dye ROX dye
of the reaction system (Figure3B). concentrations of ROX dye can result 2.00 E + 3
Bin
ROX passive reference dye between small differences in target
The Rn value is calculated as the concentration (see the precision Figure 2. Raw fluorescence data obtained
ratio of the fluorescence of Applied section on the next page). from one assay using two master mixes
with the same ROX dye concentration. The
Biosystems FAM dye divided by the difference in signal is due to the master mix
fluorescence of ROX dye. Therefore, Efficiency of a PCR reaction composition. The reaction was performed
a lower amount of ROX dye would The efficiency of a PCR reaction on an Applied Biosystems 7900HT Fast
Real-Time PCR System using an Applied
produce a higher Rn value assuming can also affect Ct. A dilution series Biosystems VIC MGB probe. The x-axis
the fluorescence signal from the FAM amplified under low-efficiency shows the emission wavelength of the
dye is unchanged. This would lead conditions could yield a standard fluorophore, and the y-axis shows the intensity
of the emission.
Rn vs. cycle
A B
0.866 1.000
0.766
Master mix A
0.666
Rn
Master mix A
CtB
0.566
0.010
Master mix B
CtA
0.466
Baseline B
0.001
0.366
1 2 3 3 3 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 1 2 3 3 3 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40
Figure 3. Amplification of an RNase P target gene in equal amounts of human gDNA using master mixes A and B. (A) Rn is plotted against cycle
number; the baselines for both reactions are shown. (B) log (Rn) is plotted against cycle number. The threshold (green line) is set at the same level
for both master mixes. The Ct of master mix B (CtB) is earlier than that of master mix A (CtA) for identical concentrations of target, reflecting the lower
baseline of master mix B. All amplifications were performed using the Applied Biosystems 7500 Real-Time PCR System.
14.5
A B
0.30
14.0
0.20
Standard deviation
13.5
Ct (VIC probe)
13.0 0.10
12.5
0.00
Figure 4. Amplification of TGF-beta using master mixes containing 3 different concentrations of ROX dye. The variation of (A) Ct and (B) standard
deviation with ROX dye concentration is shown. Decreasing the ROX dye concentration gives an earlier Ct, but increases the standard deviation. All
amplifications were performed using the Applied Biosystems 7500 Real-Time PCR System.
40
30
100%
Ct
Ct
25
70%
20
5 logs 108%
100%
X Y 92%
15
1 10 100 1,000 10,000 100,000 Dilution
Quantity
Figure 5. Variation of Ct with PCR efficiency. The blue standard curve has an efficiency of Figure 6. Accurate calculation of PCR
100% (the slope is 3.3). The green standard curve has an efficiency of 78% (the slope is 4.0). efficiency depends on the range of template
Amplification of quantity Y gives an earlier Ct under low-efficiency conditions (green) compared to the amount used for the dilution series. For a
high-efficiency condition (blue). With a lower quantity (X) there is an inversion, and the low-efficiency 2-fold dilution with 5 points (green), the potential
condition (green) gives a later Ct than the high-efficiency condition (blue). artifact is higher than for the 10-fold dilution with
6 points (blue).
curve with a different slope from defined above. efficiency of a PCR reaction, a 5-log
one amplified under high-efficiency dilution series must be performed.
conditions. In Figure 5, two samples How to evaluate the performance A slope of 3.3 10% reflects an
(X and Y) amplified under low- and of a real-time PCR reaction efficiency of 100% 10%. A PCR
high-efficiency conditions show To compare two reactions where a reaction with lower efficiency will have
different Ct values for the same condition is changed (for example two lower sensitivity.
target concentration. In this example, different master mixes or two different
although the high-efficiency condition instruments), the following parameters R2 value
(the blue curve in Figure 5) gives a must be evaluated. Another critical parameter in evaluating
later Ct at high concentrations, it PCR efficiency is R 2 , which is a
results in better sensitivity at low target Dynamic range statistical term that indicates how
concentrations. The PCR efficiency is To properly evaluate PCR efficiency, good one value is at predicting
dependent on the assay, the master a minimum of 3 replicates and a another. When R 2 is 1, the value of Y
mix performance, and sample quality. minimum of 5 logs of template (Ct) can be used to accurately predict
Generally, an efficiency between 90% concentration are necessary. The the value of x (Figure 7A). If R 2 is 0, the
and 110% is considered acceptable. reason for this suggested level value of x cannot be predicted from
of rigor is illustrated in Figure 6, the value of y (Figure 7B). An R 2 value
The observation that the Ct value which demonstrates the possible >0.99 provides good confidence in
produced from one sample is higher mathematical variation of slope or correlating two values.
than that of the other, could be efficiency obtained when testing
valuable in concluding that there dilutions over 1 log vs. 5 logs. Thus, Precision
is less template in the first sample, even if the assay is 100% efficient, The standard deviation (square root
assuming all other factors such as a range from 70% to 170% can be of the variance) is the most common
instruments, reagents, and assays obtained when testing a dilution series measure of precision. If many data
are equal. However, this is not true of a single log, due to the standard points are close to the mean, the
when different instruments, reagents, deviation in one dilution. For the same standard deviation is small; if many
primers and probes, or reaction number of dilutions or replicates on data points are far from the mean, the
volumes are involved in producing the a 5-log range, the potential artifact standard deviation is large.
two Cts. Therefore, the absolute Ct is only 8%. That means for 94%
value comparison is only meaningful efficiency on a 5-log range, the assay In practice, a data set with a sufficient
when comparing experiments using would have a range of 88% to 100% number of replicates forms an
the same reaction conditions as efficiency. To accurately determine the approximately normal distribution. This
is frequently justified by the classic cases, the standard deviation has to a key factor in determining the
central limit theorem which states be 0.167. The greater the standard sensitivity of a reaction (Figure 5).
that sums of many independent, deviation, the lower the ability to Another important consideration when
identically distributed random variables distinguish between 2-fold dilutions. detecting very low copy numbers is
tend towards the normal distribution To be able to discriminate between that a normal distribution of template
as a limit. As shown in Figure 8A, a 2-fold dilution in more than 95% of is not expected. Instead, a Poisson
approximately 68% of the values are cases, the standard deviation has to distribution is followed, which predicts
within 1 standard deviation of the be 0.250 (Figure 8C). that in a large number of replicates
mean, 95% are within 2 standard containing an average of one copy of
deviations, and 99.7% lie within 3 Sensitivity starting template, approximately 37%
standard deviations. Any system capable of effectively should actually have no copies, only
amplifying and detecting one copy of 37% should contain one copy, and
If a PCR is 100% efficient, the Ct starting template has achieved the 18% should contain two copies (see
difference between two successive ultimate level of sensitivity, regardless Figure 9). Thus, for reliable low-copy
concentrations in a 2-fold dilution is of the absolute value of the Ct. detection, a large number of replicates
1 (Figure 8B). To be able to quantify a is necessary to provide statistical
2-fold dilution in more than 99.7% of As described earlier, efficiency is significance and to overcome the
A 7
B 7
R2 = 1 R2 = 0
6 6
5 5
Sample 2
Sample 2
4 4
3 3
2 2
1 1
0 0
0 1 2 3 4 5 6 0 1 2 3 4 5 6 7
Sample 1 Sample 1
Figure 7. Examples of R2 values calculated for 2 straight lines. (A) There is a direct relation between x and y values. (B) There is no linear relation
between x and y values.
A 99.6% B X Y
95.4%
68.2%
1 1 1 1 1 1 1
1 cycle
X Y
C
1 1 1 1
3 2 1 1 2 3 1 cycle
Figure 8. Normal distribution and standard deviation. (A) Normal distribution of data is shown. For a PCR efficiency of 100%, the difference in Ct
between the means of two successive samples in a 2-fold dilution series is 1 (sample X and sample Y). (B) To be able to quantify both samples in 99.7%
of cases, the standard deviation has to be less than 1 cycle divided by 6 standard deviations (1/6 = 0.167). (C) To be able to quantify both samples in 95%
of cases, the standard deviation has to be less than 1 cycle divided by 4 standard deviations (1/4 = 0.25
Poisson distribution limitation.
Frequency for 64 replicates 20
18
16 Conclusion
14
12 Efficiency, R 2 , precision, and sensitivity are used to
3.3 pg
10
6.6 pg determine performance of a PCR reaction when comparing
8
6 different reaction conditions. For a rigorous evaluation, all
4
2
factors listed in Table 1 must be evaluated together.
0
34 35 36 37 38 39 40
In addition to these factors, proper experimental controls
Ct
(such as a no-template control and a no-RT control) and
Figure 9. Poisson distribution for low copy number. The blue curve template quality must be evaluated and validated.
represents Poisson distribution for 3.3 pg of DNA (1 copy of DNA). The green
curve represents Poisson distribution for 6.6 pg of DNA (1 cell, 2 copies of DNA).
For Research Use Only. Not for use in diagnostic procedures. 2016 Thermo Fisher Scientific Inc. All rights reserved. All trademarks
are the property of Thermo Fisher Scientific and its subsidiaries unless otherwise specified. TaqMan is a trademark of Roche Molecular
Systems, Inc., used under permission and license. CO019879 0116