Transcript Real-time qPCR Experimental Design Considerations

DNA Analysis Facility
User Educational Series
December 11, 2009
 Basic
Experimental Design for real-time
qPCR experiments
 Identify
the sources of variation in these
experiments
 Make
recommendations
Presented at qPCR Symposium 2009
San Francisco CA Nov 9-10, 2009
Tichopad A, Kitchen R, Riedmaier I, Becker C, Stahlberg A, Kubista M.
Design and Optimization of Reverse-Transcription Quantitative PCR
Experiments. Clinical Chemistry 2009;55: 1816-1823


Real-time qPCR has many applications:
• Viral Load detection
• Genotyping
• SNP detection
• ChIP Assays
• miRNA analysis
• Gene expression studies
Gene Expression experiments are typically designed to test
a hypothesis that a difference in gene expression exists
between groups of biological subjects exposed to different
treatments.

Sampling
• Collection of samples
• Storage of samples prior to extraction

Nucleic Acid Extraction
• Method of extraction
• Presence of inhibitors
• Storage of RNA prior to RT Reaction

Nucleic Acid Quality and Quantification
• Check RNA Quality
• Good Quantification in order to balance RT Rxn.

Reverse Transcription
• Selection of enzyme and priming strategy
• gDNA contamination?
• Presence of inhibitors?

Real-time qPCR
• Assay validation





 All
Choice of Chemistry
Choice of primers/probes
PCR efficiency
Dynamic Range of Assay
Choice of Endogenous Control
of these steps impact the end result of the
qPCR measurement, and they all have the
potential to add noise to the experimental data.
 Studied Variance
• The treatment effect can only be resolved if it is
larger than the random noise within the groups
due to the confounding noise.
 Confounding Variance
• Biological or Inter-subject Variance
 This is the random difference between individuals
• Processing Variance
 These are technical variances due to processing of
samples, extractions, RT and qPCR reactions.
 Goal
of Experimental Design is to optimize
your treatment effect relative to the
confounding effect of your biological and
processing noise.
 This
requires knowing where your sources of
variation are likely to occur and accounting for
these with you data analysis.
 Being
cost effective with your choices.
Kubista’s group designed an experiment to look at the sources of
variation that are found in a typical qPCR experiment.
Tichopad et al. Design and Optimization of Reverse-Transcription Quantitative PCR Experiments.
Clinical Chemistry 2009;55: 1816-1823

Liver Tissue
• qPCR Assays: ACTB, IL1B, CASP3, FGF7

Blood
• qPCR Assays : ACTB, IL1B, CASP3, IFNG

Cell Cultures
• qPCR Assays : ACTB, H3F3A, BCL2, IL8

Single Cells: individual astrocytes from mouse brain
• qPCR Assays: 18s
Tichopad et al. Design and Optimization of Reverse-Transcription Quantitative PCR Experiments.
Clinical Chemistry 2009;55: 1816-1823
Total Variance
Variance contribution
from processing steps:
Subject
Sample/Extraction
RT
qPCR
Tichopad et al. Design and Optimization of Reverse-Transcription Quantitative PCR Experiments.
Clinical Chemistry 2009;55: 1816-1823
Total noise SD: Cumulative variance which is
expressed as the SD of measured CT values.
Highlighted figure is the mean of all 4 genes.
Tichopad et al. Design and Optimization of Reverse-Transcription Quantitative PCR Experiments.
Clinical Chemistry 2009;55: 1816-1823

3 subjects x 3 samples x 3 RT’s x 3 qPCR’s (81 CT’s measured)
Subject Level:
SD was negligible at this step
Sampling Level:
Largest SD was estimated for
this step. Mean SD=1.2 Ct
which is >2 fold variation.
RT Level:
3 genes: mean SD=0.39CT’s
4th gene: SD= 0.9 CT’s
Total noise SD estimate ~1.5 CT’s
qPCR Level:
showed highest reproducibility.
Mean SD=0.09 CT’s
Tichopad et al. Design and Optimization of Reverse-Transcription Quantitative PCR Experiments.
Clinical Chemistry 2009;55: 1816-1823

3 subjects x 3 samples x 1 extractions x 3 RT’s x 3 qPCR’s
(81 CT’s measured)
Subject Level:
SD was negligible.
Sampling Level:
SD=1.9 CT’s
This is consistent with other
studies that show mRNA levels
vary greatly between individual
cells.
Total noise SD estimate ~2.0 CT’s
RT Level:
SD=0.30 CT’s
qPCR Level:
SD=0.51 CT’s
Tichopad et al. Design and Optimization of Reverse-Transcription Quantitative PCR Experiments.
Clinical Chemistry 2009;55: 1816-1823

3 subjects x 1 samples x 3 extractions x 3 RT’s x 3 qPCR’s
(81 CT’s measured)
Total noise SD estimate ~0.66 CT’s
Subject Level:
Negligible for 2 genes,
SD=1CT for other 2 genes.
Sampling Level:
Highest reproducibility
SD=0.12 CT’s
RT Level:
Similar for all genes
SD=0.24
qPCR Level:
3 higher expressor’s (CT’s 1625)
SD=0.17 CT’s
Low expressor SD=0.4 CT’s
Tichopad et al. Design and Optimization of Reverse-Transcription Quantitative PCR Experiments.
Clinical Chemistry 2009;55: 1816-1823


1 subject x 10 samples x 1 extraction x 3 RT’s x 3 qPCR’s
(90 CT’s measured)
Subject Level: Cell cultures are
unique at this level due to their
clonal nature.
Sampling Level: mean SD=0.27 CT’s
RT Level: mean SD=0.31 CT’s
Total noise SD estimate ~0.44 CT’s
qPCR Level: mean SD=0.14 CT’s
Tichopad et al. Design and Optimization of Reverse-Transcription Quantitative PCR Experiments.
Clinical Chemistry 2009;55: 1816-1823

qPCR variance (mean SD=0.13 CT’s) is lower than the variance of
other steps and does not depend on sample type.

qPCR variance will be higher in samples with CT’s > 30.

qPCR was done in duplicate in most publications, but without
justification as to why selected.

The use of single wells is indicated but does not insure against a
failed reaction.

If cDNA is limited, a single qPCR well is preferable because
splitting into two wells will further reduce the cDNA available in
the qPCR reaction.
Tichopad et al. Design and Optimization of Reverse-Transcription Quantitative PCR Experiments.
Clinical Chemistry 2009;55: 1816-1823
Supplemental Table 1
General
Recommendation
Solid tissue
Blood
Cell culture
Low copy
transcript
Upstream replicates are better than downstream replicates. Hence
generally, including more subjects is superior to any other
replicates and should be preferred as long as it is economically
feasible.
Several samples should be withdrawn from the same tissue and
processed separately (sampling replicates).
Other types or replicates are inferior
Producing RT replicates is superior to any other types of
replicates.
The number of cell culture wells should be maximized prior to any
other type of replicates. Secondarily, increasing the number of RT
replicates should be considered.
Replicates should be produced at the RT level rather than at any
other.
Tichopad A, Kitchen R, Riedmaier I, Becker C, Stahlberg A, Kubista M.
Design and Optimization of Reverse-Transcription Quantitative PCR Experiments. Clinical Chemistry
2009;55: 1816-1823
 Perform
a fully nested pilot study to
identify the sources of variation
associated with your experiment.
 Cost-optimize
the experimental design to
include the optimal number of subjects
and technical replicates you need to
strengthen the power of your
experiment.