For use with probe-based real-time PCR applications
on all real-time PCR instruments
Catalog # 172-5280
172-5281
172-5282
172-5284
172-5285
Bio-Rad Technical Support
For help and technical advice, please contact the Bio-Rad Technical Support department. In the United
States, the Technical Support department is open Monday–Friday, 5:00 AM–5:00 PM, Pacific time.
http://www.bio-rad.com
Bio-Rad Laboratories
Life Science Research
2000 Alfred Nobel Drive
Hercules, CA 94547
Online technical support and worldwide contact information are available at www.consult.bio-rad.com.
Legal Notices
LightCycler is a trademark of Roche Diagnostics GmbH. Rotor-Gene is a trademark of Corbett Research. Mastercycler is a
trademark of Eppendorf AG. Mx is a trademark of Stratagene Corporation. Bioanalyzer is a trademark of Agilent Technologies.
FAM, StepOne, and StepOnePlus are trademarks of Applera Corporation.
SYBR is a trademark of Life Technologies Corp. Bio-Rad Laboratories, Inc. is licensed by Life Technologies Corporation
to sell reagents containing SYBR Green I for use in real-time PCR, for research purposes only.
Bio-Rad’s real-time thermal cyclers are licensed real-time thermal cyclers under Applera’s U.S. Patent Number 6,814,934 B1 for
use in research, human in vitro diagnostics, and all other fields except veterinary diagnostics.
These products are covered by one or more of the following U.S. patents or their foreign counterparts owned by Eppendorf AG:
U.S. Patent Numbers 6,767,512 and 7,074,367.
NOTICE TO PURCHASER: LIMITED LICENSE
Use of this product is covered by one or more of the following U.S. patents and corresponding patent claims outside the U.S.:
5,804,375; 5,538,848; 5,723,591; 5,876,930; 6,030,787; and 6,258,569. The purchase of this product includes a limited,
non-transferable immunity from suit under the foregoing patent claims for using only this amount of product for the purchaser’s
own internal research. No right under any other patent claim and no right to perform commercial services of any kind, including
without limitation reporting the results of purchaser’s activities for a fee or other commercial consideration, is conveyed expressly,
by implication, or by estoppel. This product is for research use only. Diagnostic uses under Roche patents require a separate
license from Roche. Further information on purchasing licenses may be obtained from the Director of Licensing, Applied
Biosystems, 850 Lincoln Centre Drive, Foster City, California 94404, USA.
Sso7d Fusion Enzyme Technology iii
Educational Resources iv
Reagent Evaluation and Comparison Tutorials iv
Protocol 1
Sample Preparation Considerations 1
RNA Samples 1
RNA Integrity and Purity 1
DNA Samples 2
Plasmid Samples 2
Assay Design Considerations 3
Some Key Design Considerations 3
Multiplex Assays Design Considerations 4
Procedure 6
Reaction Mix Preparation and Thermal Cycling Protocol 6
Real-Time PCR Validation for Gene Expression Experiments 7
Determining the Optimal Reference Gene 7
Determining the Dynamic Range of the Reverse Transcription Reaction 9
Determining the PCR Efficiency 11
Troubleshooting Guide 14
Ordering Information 22
SsoAdvanced™ Universal Probes Supermix Instruction Manual| i
SsoAdvanced™ Universal Probes Supermix
Catalog # Supermix Volume Kit Size
172-5280 2 ml (2 x 1 ml vials) 200 x 20 μl reactions
172-5281 5 ml (5 x 1 ml vials) 500 x 20 μl reactions
172-5282 10 ml (10 x 1 ml vials) 1,000 x 20 μl reactions
172-5284 25 ml (5 x 5 ml vials) 2,500 x 20 μl reactions
172-5285 50 ml (10 x 5 ml vials) 5,000 x 20 µl reactions
Shipping and Storage
The SsoAdvanced universal probes supermix is shipped on dry ice. Upon receipt, the supermix
should be stored at –20ºC in a constant temperature freezer and protected from light. When
stored in these conditions, the supermix is guaranteed for one year. When stored at 4ºC, the
supermix is guaranteed for three months. To avoid excess freeze-thaw cycles, we recommend
preparing aliquots for storage.
Kit Contents
SsoAdvanced universal probes supermix is a 2x concentrated, ready-to-use reaction master
mix optimized for dye-based real-time PCR on any real-time PCR instrument (ROX-independent
and ROX-dependent). It contains antibody-mediated hot-start Sso7d fusion polymerase, dNTPs,
MgCl2, probes, enhancers, stabilizers, and a blend of passive reference dyes (including ROX
and fluorescein).
Instrument Compatibility
This supermix is compatible with all Bio-Rad and ROX-dependent Applied Biosystems real-time
PCR instruments, and with the Roche LightCycler LC480, Qiagen Rotor-Gene Q, Eppendorf
Mastercycler ep realplex, and Stratagene Mx real-time PCR systems.
Product Use Limitations
The SsoAdvanced universal probes supermix is intended for research use only, and is not
intended for clinical or diagnostic use.
Technical Assistance
Bio-Rad Laboratories takes great pride in providing best-in-class technical support through
our online, telephone, and field support. To obtain support, please visit www.bio-rad.com, call
1.800.4.BIORAD, or contact your local field applications scientist.
Quality Control
SsoAdvanced universal probes supermix demonstrates high PCR efficiency and linear
resolution over a wide linear dynamic range. Stringent specifications are maintained to ensure
lot-to-lot consistency. This product is free of detectable DNase and RNase activities.
Bio-Rad introduced our next generation of real-time PCR supermixes using our patented
Sso7d fusion protein technology, delivering a reagent that provides effective performance
in a wide range of qPCR applications. The dsDNA-binding protein, Sso7d, stabilizes the
polymerase-template complex, increases processivity, and provides greater speed and
reduced reaction times compared to conventional DNA polymerases, without affecting PCR
sensitivity, efficiency, or reproducibility.
Key Features and Benefits
Fast qPCR results and high performance — the Sso7d fusion polymerase and optimized
buffer deliver fast reaction times via instant antibody hot-start polymerase activation and
rapid polymerization kinetics to generate exceptional qPCR results in less than 30 min
Minimal inhibition of PCR — the polymerase’s increased resistance to PCR inhibitors
ensures maximum efficiency, sensitivity, and reproducibility
Single copy detection — data illustrate high sensitivity with amplification and detection from
a single copy of target gene
Robust discrimination and reproducibility — efficient discrimination and reliable
quantification can be obtained from 1.33-fold serial dilutions of input template
GC-rich targets — ability to amplify targets where other Taq-based supermixes
may be challenged
To learn more about similarities and differences between PCR and real-time PCR, understand
how SYBR® Green and probe-based chemistries function, and see how data are collected and
interpreted, please view our interactive tutorial Understanding Real-Time PCR.
Reagent Evaluation and Comparison Tutorials
Reverse Transcription
When comparing two different reverse transcription kits, often not all characteristics of the
reverse transcription (RT) reaction are tested. The end result is that a decision is made using
a limited set of data and criteria. The following protocol and exercise have been written in an
effort to create a more robust, reliable, and reproducible method of testing sensitivity, efficiency,
and other critical characteristics when comparing reagent providers for reverse transcription
kits. Reagent Comparison Guide for Real-Time PCR
To view an interactive tutorial and learn about reverse transcription chemistry, enzymes,
and priming methods, as well as how to perform a reagent comparison, please click here.
Understanding Reverse Transcription
Supermixes
When comparing two different supermixes, often not all characteristics of the PCR reaction
are tested. The end result is that a decision is made using a limited set of data and criteria.
The Reagent Comparison Guide for Real-Time PCR was written in an effort to create a more
robust, reliable, and reproducible method of testing sensitivity, efficiency, and other critical
assay characteristics when comparing reagent providers for use on real-time PCR systems.
To view an interactive tutorial and learn about supermix chemistry and enzymes, as well as how
to perform a reagent comparison, please click here. Understanding Real-Time PCR Supermixes
This manual is intended for use with probe-based assays on all real-time PCR systems using
a broad range of cycling conditions, template and primer input concentrations, and fast or
standard run times.
Sample Preparation Considerations
RNA Samples
Isolate RNA using the appropriate method for the given sample type (Aurum™ total RNA mini
kit for cell lines, Aurum total RNA fatty and fibrous tissue kit for tissue samples)
Compare the expected yield to the actual yield to ensure the isolation method yielded the
appropriate RNA concentrations (5–30 pg per cell, 0.1–4 µg per mg of tissue). When the yield
is less than expected, this may lead to suboptimal qPCR data results, due to less than ideal
quality samples resulting from suboptimal sample prep workflow
When the RNA will be used for RT-qPCR, it is recommended that you treat the sample
with DNase to remove residual contaminating DNA. DNase treatment is also a good idea
when isolating RNA from tissues that are high in DNA, as the excess DNA may affect
downstream applications
Store the RNA in an appropriate solution
– 0.1 mM EDTA (in DEPC-treated ultrapure water)
– TE buffer (10 mM Tris-HCl, 1 mM EDTA, pH 7.0)
Store the RNA at –80ºC in single-use aliquots
RNA Integrity and Purity
Use the Experion™ automated electrophoresis system or the Agilent Bioanalyzer to evaluate
the integrity of the RNA sample. When using multiple samples in the comparison, ensure that
the RQI/RIN numbers are similar to ensure accurate qPCR results
Use an agarose gel to assess RNA integrity if the above systems are not available. Apply the
same analysis concepts. High quality RNA will yield two clean peaks, 18s and 28s. Degraded
RNA will appear as a smear on the gel
To assess purity, evaluate the following spectrophotometer readings:
– A260/A280 >2.0 for pure RNA
– A260/A230 ~2.0 for pure RNA
• Lower ratios are indicative of contaminants from salts, carbohydrates, peptides, proteins,
phenols, and guanidine thiocyanate
Isolate DNA using the appropriate method for the given sample type (for example, column
purification for cell lines, phenol/chloroform or column purification for tissue samples)
Store the DNA in an appropriate solution
– 0.1 mM EDTA (in DEPC-treated ultrapure water)
– TE Buffer (10 mM Tris-HCl, 1 mM EDTA, pH 7.0)
Store the DNA at –80ºC in single-use aliquots
Assess DNA quality with an agarose gel; a single band indicates high integrity DNA, whereas
a smear indicates degraded DNA
Assess the DNA purity using a spectrophotometer for the following:
– A260/A230 >1.5 (lower ratios may be attributed to carryover guanidine, and/or inhibitors
– A260/A280 1.7–2.0 (lower ratios are indicative of contaminants from salts, carbohydrates,
– Higher ratios may be indicative of RNA contamination
Tips:
like humic acid and organics)
peptides, proteins, phenols, and guanidine thiocyanate)
Heat treating DNA may be required prior to qPCR to relax strong secondary structure
Using a restriction digest enzyme may be required for select qPCR applications, such as copy
number variation, to reduce signal-to-noise ratio.
Plasmid Samples
Prepare plasmids using an appropriate method
Store the stock plasmid in an appropriate solution
– TE buffer (10 mM Tris-HCl, 1 mM EDTA, pH 8.0)
Store the plasmid at –80ºC in single-use aliquots
Assess plasmid quality with an agarose gel; a single band indicates high integrity plasmid,
whereas a smear indicates degraded plasmid or excess enzymatic activity
Assess the plasmid purity using a spectrophotometer for the following:
– A260/A280 1.7–1.9 (lower ratios are indicative of contaminants from salts, carbohydrates,
peptides, proteins, phenols, and guanidine thiocyanate)
– Higher ratios may be indicative of RNA contamination
Sequence quality and secondary structure — evaluate using web-based tools to understand
the complexity of the structure, as it can impact the reaction performance
Sequence length — use the entire gene sequence, or a specific region of interest, to
optimally design an assay
Sequence masking — use web-based masking tools to mask low complexity and repetitive
regions to avoid assay design in these regions
Uniqueness of the sequence — use BLAST or BLAT to ensure no homology exists and help
avoid mispriming events
Uniqueness of the assay — use in silico PCR, or Primer-BLAST, to “blast” the primers against
the genome of interest to validate primer design specificity
Default settings in the software — ensure they are set correctly (for example, salt conditions,
oligo and amplicon sizes). The SsoAdvanced™ universal probes supermix and the qPCR
cycling protocols have been optimized for assays with a primer melting temperature (Tm)
of 60ºC designed using the open source Primer3, Primer3Plus or Primer-BLAST, default
settings. For assays designed using other tools, the primer T
Primer3. Suggested settings: 50 mM Na+, 3 mM Mg++, 1.2 mM dNTPs, 250 nM annealing
oligo, SantaLucia/SantaLucia
should be recalculated using
m
Some Key Design Considerations
For optimal PCR efficiency, design the amplicon size between 70 and 150 bp (<70 bp may be
needed for degraded/FFPE templates)
Maintain primer lengths between 18 and 25 bp for good specificity and binding abilities
Annealing temperatures between 58 and 62ºC are optimal (greater range can be obtained
using Bio-Rad’s Sso7d-based supermixes); temperatures >60ºC may result in less binding
efficiency and <58ºC may result in less specificity
The optimal amplicon GC content should be within 40–60% (greater range can be obtained
using Bio-Rad’s Sso7d-based supermixes)
Avoiding primer secondary structures reduces potential primer-dimer issues
Avoid mispriming by ensuring there are no more than 2 Gs or Cs in the last 5 bases on the 3'
end of the primer
Design your assay such that at least one primer or the probe spans an exon:exon junction
site to avoid gDNA amplification
Alternatively, design the assay such that the primers are in separate exons and the intron
size is >1 kb
Probe annealing temperature should be 8–10ºC higher than the primers to
ensure binding to the template prior to extension
Avoid placing Gs on the 5' end of the probe to avoid quenching of the fluorophore even after
probe cleavage
Probe lengths typically range from 18–30 bp, and vary depending on the type of probe
chemistry used and the target sequence
To ensure data generated in a multiplex reaction are equivalent to data generated in a singleplex
reaction, it is imperative to evaluate the assay performance in both singleplex and multiplex
reactions. It is also important to understand the expression level of your target sequences, as
this will impact the multiplex optimization method.
A B C
Fig. 1. Graphs show the three modes for expression in a duplex reaction. A, both genes express relatively equally;
B, one gene always expresses more than the other; C, one gene varies in expression levels depending on the sample.
1. Determine the expression levels of the genes prior to optimizing a multiplex approach. This
can be accomplished through the use of standard curves derived from template serial
dilutions (for example 100 ng to 1 pg).
2. Assign the reporter dyes based on the expression levels; brighter fluorophores should be
reserved for lower expressing targets.
3. Consider these factors when designing primers and probes:
All assays in a multiplex reaction should have the same or nearly the sameannealing temperature
Analyze all oligos for primer-dimer stability with all other oligos in the reaction
Account for nonspecific primer-probe annealing and cross reaction between assays
Amplicons of 50–150 bases are preferred. Shorter amplicons often have better
PCR efficiencies
4. Optimize each primer set.
Not all primer sets/concentrations perform the same. Empirical testing using a standard
curve is imperative
Prepare two standard curves for each assay
– One singleplex for each assay (if not completed already)
– One duplex (or multiplex if applicable)
PCR efficiencies must be similar to minimize amplification bias
When combined in a reaction, be sure each assay is used within the appropriate template/
target dynamic range. This range is often reduced when used in a multiplex application
5. Reduce the primer concentration of the higher expressing target.
Reducing the concentration, often times to 100–150 nM final, enables the lower expressing
target to amplify sufficiently. Use a primer matrix to determine the optimal concentration,
at which the chosen concentration would yield no shift in Cq values while exhibiting the
lowest flwescence signal
6. Reduce the primer concentration of both or all targets when the expression levels are
unknown or vary from sample to sample.
40
35
30
25
Cq
20
15
Log Starting Quantity
Fig. 2. Overlay of singleplex and duplex standard curves demonstrating no
significant performance differences, thus a well-optimized duplex exists.
Duplex Singleplex
Notes: Preferential amplification of targets:
Target abundance should not vary too much between assays; limiting variance guards
against domination of the reaction by a more abundant target
Verify that the Cq values have not changed between the singleplex and multiplex
standard curves
– If unavoidable, it may be permissible if all assays in the reaction shift equally
Always evaluate the performance of the supermix following the recommended reaction and
cycling conditions prior to modification
Be sure to set the activation time to 30 sec for cDNA and 2–3 min for genomic DNA
The 2x supermix has been optimized for 20 µl reactions in 96-well plates and 10 µl reactions
in 384-well plates
Procedure
Reaction Mix Preparation and Thermal Cycling Protocol
1. Thaw SsoAdvanced™ universal probes supermix and other frozen reaction components to
room temperature. Mix thoroughly, centrifuge briefly to collect solutions at the bottom of
tubes, and then store on ice protected from light.
2. Prepare (on ice or at room temperature) enough reaction setup for all qPCR reactions
by adding all required components except the template according to the following
recommendations (Table 1).
Table 1. Reaction setup.*
Volume per Volume per
Component 20 μl Reaction 10 μl Reaction Final Concentration
* Scale all components proportionally according to sample number and reaction volumes.
3. Mix the assay master mix thoroughly to ensure homogeneity and dispense equal aliquots
into each PCR tube or into the wells of a PCR plate. Good pipetting practice must be
employed to ensure assay precision and accuracy.
4. Add samples (and nuclease-free H2O if needed) to the PCR tubes or wells containing the
reaction setup (Table 1), seal tubes or wells with flat caps or optically transparent film, and
vortex 30 sec or more to ensure thorough mixing of the reaction components. Spin the tubes
or plate to remove any air bubbles and collect the reaction mixture in the vessel bottom.
5. Program thermal cycling protocol on the real-time PCR instrument according to Table 2.
6. Load the PCR tubes or plate onto the real-time PCR instrument and start the PCR run.
7. Perform data analysis according to the instrument-specific instructions.
Polymerase Annealing/
Setting/Scan Activation and Denaturation Extension + Plate
Real-Time PCR System Mode DNA Denaturation at 95/98°C Read at 60°C** Cycles
Roche LightCycler 480 Fast 10–30 sec
Standard 60 sec
Qiagen Rotor-Gene and
Stratagene Mx series
* 2–3 min denaturation at 95°C is highly recommended for genomic DNA template to ensure complete denaturation.
** Shorter annealing/extension times (1–10 sec) can be used for amplicons <100 bp. Longer annealing/extension
®
CFX96™, CFX384™,
®
iQ™5, MiniOpticon™,
times (30–60 sec or more) can be used for amplicons >250 bp, GC- or AT-rich targets, low expressing targets,
crude samples, or for higher input amounts (for example, 100 ng of cDNA or 500 ng of genomic DNA).
Standard 15–30 sec
Fast 5–15 sec 10–30 sec
Standard 60 sec
Fast
30 sec at 95°C for
cDNA
or 5–15 sec
2–3 min at 95°C
for genomic DNA*
10–30 sec
35–40
35–40
Real-Time PCR Validation for Gene Expression Experiments
The following validation experiments are critical for obtaining valid and publishable real-time
PCR data following the MIQE guidelines. These simple-to-follow experiments should be
completed prior to starting a new real-time PCR project.
Determining the Optimal Reference Gene
To properly perform a gene expression experiment, it is imperative that an optimal
reference gene(s) is used. The reference gene(s) must maintain a consistent expression
level across all samples in the project regardless of treatment, source, or extraction
method. The variation in reference gene expression is somewhat dependent on the level
of fold change discrimination desired. For example, if a twofold change in expression is
important, then the reference gene should have little to no variation in expression. However,
if a 20-fold change in expression is important, then the reference gene expression can have
some variability. To validate a reference gene(s), follow the steps on the next page.
1. Begin searching for a candidate list of reference genes by searching publications, speaking
with researchers using similar model systems, and mining microarray data, if available.
Minimally, five reference genes should be selected for evaluation. For your convenience,
Bio-Rad offers pre-plated reference gene panels using our highly validated and optimized
2. From your experiment, randomly select a few samples from each group (for example,
treatments, time courses, sources) ensuring that you evaluate all variable sample groups.
3. Isolate the RNA and DNase-treat using the same protocol for all samples. Quantify and
normalize the RNA to the same concentration.
4. Perform a reverse transcription reaction for each sample using the same kit, volume, and
concentration. Dilute the cDNA, as needed, treating each sample the same to ensure there
are no differences from sample to sample in terms of volume and concentration from the
initial RNA input.
5. Perform a real-time PCR experiment using the samples and the candidate reference genes
using technical triplicates for each sample.
6. Evaluate the data for each reference gene by calculating a standard deviation for all
samples. For example, if you evaluated eight samples and seven reference genes, simply
calculate the standard deviation of those eight samples’ Cq values for each reference
gene. Thus, you will end up with seven standard deviation values. Compare the values
to determine which reference gene(s) have the lowest value. Although there is no precise
threshold for determining a good reference gene, a good rule of thumb is to ignore any
reference gene with a standard deviation higher than 0.5. If you are using a Bio-Rad CFX
real-time PCR system, you can utilize the software to automatically calculate an M-value to
assist in determining the optimal reference gene.
In this data set (Figure 3), TBP and PPIA are both below 0.5 and may be suitable reference
genes for the given project. Keep in mind there is no one good reference gene for all
projects, so the reference gene must be validated for every project.
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0
Fig. 3. Seven reference genes evaluated using random samples from untreated and
treated sample groups. TBP and PPIA exhibited the lowest standard deviations with ~0.4 and
0.2, respectively. Note that GAPDH and ACTB exhibited the highest standard deviations, thus
would be unacceptable reference genes. If you are unable to find a single stable reference gene,
consider using multiple reference genes. This method involves calculating a geometric mean of
the reference gene quantities (not Cq values) prior to performing the normalization.
Determining the Dynamic Range of the Reverse Transcription Reaction
An optimal reverse transcription reaction is expected to generate a true representation of the
RNA converted into cDNA. However, it is imperative to determine the dynamic range of the
reaction to ensure that the initial RNA loaded does not fall outside the dynamic range. If it
does, then the downstream real-time PCR data may be invalid. To validate the dynamic range,
perform the following :
1. Preparation of a serial dilution using a single RNA source (or a pooled RNA sample) is
required to prepare the cDNA synthesis reactions for the experiment. Ensure an adequate
amount of RNA is available; adjust concentrations and volumes accordingly.
2. Start with 1 µg of total RNA and perform a tenfold serial dilution covering at least 5 or 6 logs
of dynamic range.
3. Perform RT using 20 μl reactions. Transfer the RNA, as shown in Figure 4, to the respective
reaction tubes. For example, transfer 1 μg of RNA to Reaction 1 tube. Repeat transferring
RNA to the remaining reaction tubes.
Serial dilution
of the RNA
0
10
–1
10
–2
10
–3
10
–4
10
–5
10
–6
10
1 µg RNA
100 ng RNA
10 ng RNA
1 ng RNA
100 pg RNA
10 pg RNA
1 pg RNA
Bio-Rad® iScript™ cDNA
Synthesis Kit
Reaction 1
Reaction 2
Reaction 3
Reaction 4
Reaction 5
Reaction 6
Reaction 7
Fig. 4. Tenfold serial dilution of RNA starting at 1 µg down to 1 pg, thus covering six logs of dynamic
range. Each RNA dilution was transferred to the respective cDNA reaction tube for cDNA synthesis.
4. Dilute the cDNA as needed to perform real-time PCR reactions using a minimum of two
genes — reference and low expressing. However, it is recommended to evaluate four genes
— reference, low, medium, and high expressing.
5. Prepare the real-time PCR plate (Figure 5) and cycle according to the recommended protocol.
Reference
gene
Fig. 5. A recommended plate layout.
Low
expressor
Medium
expressor
High
expressor
6. Evaluate the data. Follow the guidelines in this manual (page 14–15) for setting the baseline
and threshold prior to analyzing the data. Figure 6 illustrates the most common results from
the experiment and how to interpret the data.
Fig. 6. The blue standard curve represents the target gene and the green standard curve
represents the reference gene. A, both assays demonstrate equivalent performance in linearity
and dynamic range covering 1 µg to 1 pg. Thus, any RNA input going forward within this range will
be acceptable; B, both assays are either saturated at the 1 µg data point or the reverse transcription
reaction is inhibited due to carryover inhibitors from the RNA sample. Consider using less RNA (≤100 ng)
or re-purifying the RNA; C, the reference assay has a broader dynamic range than the target assay,
therefore, the dynamic range is limited. Consider reevaluating the target assay design, using less RNA
(≤100 ng), or re-purifying the RNA; D, the target assay exhibits a high standard deviation at the lowest
concentration (1 pg) and should not be considered part of the dynamic range. This is due to a lack of
sensitivity or reproducibility, and may be alleviated by using a carrier in the RNA sample such as
glycogen or non-target gDNA carrier; E, after considering all the data, the concentration points that
define the dynamic range from rejecting the variant 1 pg data and the saturated/inhibited 1 µg data point
results in an effective dynamic range (RNA loading) is 1–100 ng.
Determining the PCR Efficiency
Determining the PCR efficiencies of your reference gene and target gene(s) is critical before
starting any real-time PCR experiment. Knowing the PCR efficiency determines the appropriate
relative gene expression math model. Not knowing may affect and invalidate the results. To
determine the PCR efficiency among other key characteristics, prepare standard curves to
evaluate the following:
1. A serial dilution of the cDNA, gDNA, or plasmid template is required to prepare the standard
curve. Ensure an adequate supply of template and an adequate volume are available to
evaluate all the assays used in the experiment.
Serial dilution of the template
11:101:10 01:1,0001:10,0001:100,0001:1,000,000
Fig. 7. Tenfold serial dilution covering 6 logs of dynamic range is prepared using a starting template
of your choice based on target expression levels.
2. Prepare the real-time PCR reactions using a fresh bottle of supermix, nuclease-free water,
and primer sets. Figure 8 is an example of a plate layout.
Reference
gene
Fig. 8. Example of a plate layout with four seven-point
standard curves with NTCs in technical triplicates —
one for each gene of interest and the reference gene.
Low
expressor
Medium
expressor
High
expressor
3. Cycle according to the recommended protocol.
4. Analyze the data. Follow the guidelines in this manual for setting the baseline and threshold
prior to analyzing the data.
5. To determine which math model should be applied, simply subtract the slope value of the
reference gene from each target gene. If the ∆slope is ≤0.1, then the PCR efficiencies are
within accepted limits and the ∆∆CT math model can be used. If the ∆slope is ≥0.1, then the
efficiency correction math model (Pfaffl method) must be applied.
Pipet a minimum of 5 µl for each sample. This ensures greater precision and a smaller standard
deviation for technical replicates. If the samples are too concentrated, simply dilute accordingly.
Use a calibrated pipet of the appropriate volume range and never plunge the tip more than several
millimeters below the surface of the sample. Pipet slowly and use the pipet tip demarcations to
visualize accuracy
Prepare individual master mixes for each sample by combining the real-time PCR supermix,
nuclease-free water, and primers along with the template and mix thoroughly. Then, pipet 20 µl
into the respective wells on the plate
A tenfold dilution series is recommended to cover the most logs of dynamic range; however,
depending on the expression level of the gene(s) evaluated and the total template amount
available, this can be reduced to a fivefold dilution series
Efficiency
Calculate efficiency using the software or the following equation:
[–1/m]
E = 10
30
25
Cq
20
15
–1. A PCR efficiency from 90–110% (slope values from –3.6 to –3.1) is preferred.
2345
Log Star ting Quantity
Fig. 9. Reference gene has a PCR efficiency of 97.59% (–3.381) and six logs of
dynamic range. The target gene has a PCR efficiency of 99.17% (–3.342) with six logs of
dynamic range. Subtracting the slope values, 3.381 – 3.342 = 0.039, which is <0.1.
678
Linearity
Calculate the R2 statistic for each standard curve using the qPCR analysis software; the R2
should be ≥–0.980. However, if the R2 is <0.980, remove outliers. If there are too many outliers,
then reevaluate the experiment to determine the cause of the lower R2 value.
Dynamic Range
Determine the general trend of the slope where linearity (R2) and efficiency are within acceptable
ranges, as specified above.
Sensitivity
Determine the lowest concentration of the serial dilution where replicate reproducibility is high
and the R2 of the standard curve is ≥0.980.
Specificity
For probe-based assays, achieving optimal specificity requires observing a single band (PCR
product) following a gel analysis.
Review Tables 3 and 4 to determine if you are within an acceptable range of nucleic acid yield.
If your yields of RNA are considerably less than is typical for your sample type, reevaluate
your isolation method. For reference, typical yields from some mammalian tissues are listed in
tables 3 and 4.
Table 3. RNA yields.
Total RNA per Cell Total DNA per Cell
5–30 pg Varies by genome
Table 4. RNA yields per mg of tissue.
Sample Type Yield
Liver 4 µg
Spleen 4 µg
Heart 3 µg
Kidney 2 µg
Lung 2 µg
Brain 1.5 µg
Bone 50 ng
Adipose <10 ng
PCR Inhibitors/Oversaturation
If you suspect that your sample(s) contains PCR inhibitors, consider the following
corrective actions:
1. Evaluate your sample type to determine if any of the common inhibitors listed in the
following list may be present in your sample as carryover. If you suspect contamination,
re-purify the samples using a commercially available post-isolation cleanup kit.
2. Evaluate the A260/280 and A260/230 ratios. Refer to the RNA/DNA isolation section
(page 1–2) in this manual for further information.
Melanin EtOH >1% v/v
Polysaccharides Proteinase K
Polyphenolics DMSO >5%
Hemoglobin EDTA >50 mM
Chlorophyll SDS >0.01% w/v
Heparin Sodium Acetate >5 mM
Humic acid Mercaptoethanol
Hematin Guanidinium
Phenol >0.2% v/v
DTT >1 mM
* Not an inclusive list.
3. If the most concentrated sample in the dilution series is showing compression, as seen in
Figure 10, where the tenfold dilution series ∆Cq value is < 3.3 compared to the more diluted
points, then PCR inhibitors are most likely present in the sample. However, compression
may also be due to an overloaded amount of template, error in the dilution series, or
pipetting error.
a. Re-purify the sample(s) using a different isolation method, or post-isolation column cleanup
b. Remove the highest dilution point
c. Increase the annealing/extension time
Amplification
4
10
3
10
RFU
2
10
0
0
Fig. 10. Presence of PCR inhibition at the
highest dilution point, as indicated by
delayed amplification.
Low Template Input, Low Expression, High Cq Values
If your Cq values are higher than expected or you are concerned about Cq values >30,
consider the following corrective actions:
1. Confirm the expected expression level, if known, to ensure that the target of interest is
present in your given sample. Additionally, consider higher input concentrations of sample
for low expressing targets. Remember that for every twofold increase in starting sample
concentration, the Cq value shifts one cycle earlier (assuming 100% PCR efficiency).
2. Confirm the template input amount using a fluorescence-based quantification method to
ensure the cDNA. input range is 100 ng to 100 fg or the genomic DNA input range is 500 ng
to 5 pg. (cDNA will require purification prior to quantification analysis.)
3. Increase the volume of template pipetted into the PCR reaction. For the highest accuracy
and precision, pipet a minimum volume of 5 µl for each sample.
4. Consider adding a carrier to your sample stock to increase homogeneity — examples
include tRNA, glycogen, and unrelated gDNA.
5. Consider using nonstick polypropylene tubes for sample stock storage to prevent nucleic
acid from binding to the tube walls.
6. Confirm that the reverse transcription reaction was successful. A simple-to-follow protocol is
outlined in Reagent Comparison Guide for Real-Time PCR
Amplification Plots
If you notice that any data point(s) in your amplification plots exhibit a sigmoidal shape in
the log view (Figure 11, left), this is typically due to an incorrect baseline setting. Consider
the following corrective actions:
1. Deselect automatic baseline setting and assign manual baseline. Adjust the baseline begin
and end cycles so that the amplification plot matches the others on the plot. Sometimes this
takes a few tries, but a general rule of thumb is to set the end cycle about two cycles before
the start of true amplification, as seen in Figure 12.
6
10
4
10
3
10
RFU
2
10
1
10
Fig. 11. Incorrect baseline is exhibited in the left graph indicated by the arrow pointing to the first
dilution point where the amplification plot is more sigmoidal in shape. As a result, an artificially lower
Cq value is obtained. Corrected baseline is shown in the graph on the right.
Fig. 12. Baseline set ting is best completed in the linear view. In this example,
the amplification starts around cycle 8; therefore, setting the end baseline two
cycles prior at cycle 6 is best.
2. Either remove this data point or dilute your sample so that it does not show amplification earlier
than cycle 15. This ensures that the software’s algorithm has enough background to subtract
from the signal. Early amplification may cause the algorithm to fail due lack of background data.
If you notice high standard deviations for technical replicates or inconsistent gene
expression data, this could be due to the threshold being positioned either too high or
too low. Consider the following corrective action:
When setting the threshold, you should choose a position that is in the middle of the geometric
(exponential) phase of PCR. Setting the threshold too high or too low places the threshold in a less
than ideal region of amplification where greater noise is present and PCR is not 100% efficient.
A B
3
10
RFU
2
10
01020304050
C
3
10
RFU
2
10
0
10
Amplification
Cycles
Amplification
30
20
Cycles
Amplification
3
10
RFU
2
10
0
50
40
Cycles
20
10
30
4050
Fig. 13. Illustrations of baseline settings. A, when the threshold is set too high, the data collected are often
from the linear phase of PCR, where the reaction is not the most efficient; B, the threshold is set too low. When
set too low, the data collected are often within the background noise of the reaction; C, a correct threshold
setting where the data collected are within the geometric (exponential) phase of PCR.
If you have already ruled out your samples as a source for poor efficiency, then the assay may
be the cause of the problem. Please review the section on assay design in this manual for
further information (page 3).
Also, consider the following corrective action:
Perform a temperature gradient experiment to determine the optimal annealing temperature.
Set up the gradient as follows:
a. Use several representative samples in your project.
b. Set the temperature range 10ºC above and 6ºC below the calculated annealing temperature.
c. Choose the final annealing temperature based on overall performance related to specificity.
1. Evaluate the assay design by following the bioinformatics workflow outlined in the beginning
of this manual (page 3). This will help ensure that the primers are highly specific to your
target of interest and no other target region(s).
2. Perform a temperature gradient to determine the optimal annealing temperature of the
primers. Load your plate with the same reaction setup and sample for each primer set in a
column format so that you can evaluate the annealing temperatures. Set the gradient 10ºC
above and 6ºC below the calculated annealing temperature to ensure a proper temperature
range is covered. Choose the best temperature based on the overall PCR amplification,
keeping in mind that lower temperatures may reduce specificity and higher temperatures
may reduce primer binding efficiency.
If you suspect the standard curve and dilution points are not within the MIQE guidelines
of 90–110% PCR efficiency with an R2 of 0.99 or greater, consider the following
corrective actions:
1. Ensure that the standard curve covers at least 5–6 logs of dynamic range. When the
standard curve is too small, the variability of the true efficiency greatly increases.
2
2. If the R
is <0.98, review the standard curve data points for outliers. Remove any outliers
where the ∆Cq is >0.5 for the group. For example, if your 100 pg dilution point has Cq
values of 29.2, 29.6, and 30.5, you should remove the Cq value of 30.5. If there are too
many outliers, it may be a sign of other technical issues.
Control Samples/Wells Are Not Performing as Expected
If your non-template control (NTC) wells indicate amplification, you need to determine
the source. Although the most likely cause is nucleic acid contamination, other possible
causes include:
Pipetting template into the NTC well
Sample from adjacent wells being aerosolized while pipetting or removing the plate seal
after samples have been loaded
Contaminated plate, water, primers, or supermix
Use of nonfiltered pipet tips
Degraded probe
1. Evaluate your current workflow and adjust as needed. If you suspect your reagents are
contaminated, the best method to determine the source is to replace them one at a time
starting with the water, which is a common source of contamination. Next, make a fresh
dilution of primers from the stock solution. And finally, use a new aliquot of the supermix.
Discard any identified contaminated reagent from the lab.
2. If the problem persists, evaluate the background noise for the entire real-time PCR run
across all wells. If the signal is unusually high compared to prior runs, your probe may
be degrading. When this occurs, the high temperatures cause the probe to cleave, thus
releasing the reporter dye into solution and allowing fluorescence. Probes should be
aliquoted upon receipt into amber tubes and should not be exposed to freeze/thaw
cycles > five times, as this causes premature degradation.
If your no-RT control wells indicate amplification, you need to determine the amount
of gDNA contamination present in your cDNA sample(s) to understand the impact on
your data.
1. Using Table 5, determine the percent of gDNA contamination present. For example, if the
∆Cq (no-RT control Cq – cDNA Cq) for a given sample is seven or greater, then less than 1%
of the DNA present in the sample is gDNA, which would be considered insignificant.
Table 5. Determining percent of gDNA contamination.
2. Evaluate the assay design and note the location of the primers. To avoid gDNA amplification,
at least one primer must span an exon:exon junction site. Alternatively, the primers can be
designed in two different exons that are separated by an intronic region >1 kb.
If you are using an internal positive control (IPC) and the standard deviation of the Cq values
across all samples is >0.167, then consider the following:
When the IPC for a given sample(s) is higher than the group, this is most likely due to the
presence of a PCR inhibitor. Review the sections on sample preparation for more information.
Multiplexing
Ideally, data between singleplex and multiplex should remain the same in terms of Cq values
and PCR efficiency. In addition, if your data exhibits relative Cq shifts for all data points
between singleplex and multiplex, then the final data output remains the same. However, if
you observe variable Cq shifts for respective data points between singleplex and multiplex,
consider the following:
The higher expressing assay may be using up the reaction components such as dNTPs
and enzyme, and thus causing a shift in the lower expressing assay to later Cq values than
observed in singleplex.
1. Primers and probes from different assays may be interacting. Make sure there are no stable
dimers formed between the oligos from different assays. This can be completed using
various open source tools online.
2. If the lower expressing assay has a longer amplicon, >150 bp, then consider redesigning the
assay to be shorter or equivalent in length to the higher expressing gene. Shorter amplicons
typically can have greater PCR efficiencies.
3. Choose assays with more similar expression levels, if possible, to avoid reagent
competition. If this strategy is not possible, optimize the assays using a primer-limiting
strategy to limit the available primer for the higher expressing gene. This in turn forces an
earlier plateau phase of PCR.
a. Construct a primer matrix (Table 6.) for the higher expressing assay ranging from 50 nM
to 150 nM while keeping the lower expressing assay constant.
b. Select the concentration that generates the lowest fluorescence signal without any effect
on the Cq compared to singleplex data.
c. Repeat the multiplex experiment to compare the newly optimized primer set.
172-5280 2 ml (2 x 1 ml vials), 200 x 20 μl reactions
172-5281 5 ml (5 x 1 ml vials), 500 x 20 μl reactions
172-5282 10 ml (10 x 1 ml vials), 1,000 x 20 μl reactions
172-5284 25 ml (5 x 5 ml vials), 2,500 x 20 μl reactions
172-5285 50 ml (10 x 5 ml vials), 5,000 x 20 µl reactions
Two-Step Reverse Transcription Reagents
170-8842 iScript Advanced cDNA Synthesis Kit for RT-qPCR, 50 x 20 μl reactions
170-8843 iScript Advanced cDNA Synthesis Kit for RT-qPCR, 250 x 20 μl reactions
170 - 8 89 0 iScript cDNA Synthesis Kit, 25 x 20 μl reactions
170-8891 iScript cDNA Synthesis Kit, 100 x 20 μl reactions
170-8840 iScript Reverse Transcription Supermix for RT-qPCR, 25 x 20 μl reactions
170-8841 iScript Reverse Transcription Supermix for RT-qPCR, 100 x 20 μl reactions
170 - 8 89 6 iScript Select cDNA Synthesis Kit, 25 x 20 μl reactions
170 - 8 897 iScript Select cDNA Synthesis Kit, 100 x 20 μl reactions