Sample & Assay Technologies Advanced miRNA Expression Analysis: From Experimental Design through Data Analysis Jonathan Shaffer, Ph.D. [email protected]microRNA Technologies, R&D Americas The products described in this webinar are intended for molecular biology applications. These products are not intended for the diagnosis, prevention or treatment of disease.
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
The products described in this webinar are intended for molecular biology applications. These products are not intended for the diagnosis, prevention or treatment of disease.
Sample & Assay Technologies Three part webinar series
Webinar 2: Advanced miRNA Expression Analysis 2
miRNA and its role in human disease
Webinar 2 : Advanced microRNA expression analysis: From experimental design through data analysis
Speaker: Jonathan Shaffer, Ph.D.
Webinar 3 : Profiling miRNA expression in Cells, FFP E, and serum:On the road to biomarker development
Speaker: Jonathan Shaffer, Ph.D.
Webinar 1 : Meeting the challenges of miRNA research :An introduction to microRNA biogenesis, function, a nd analysis
� Overview� miRNA Background� miRNA expression profiling using a miScript miRNA PCR Array
� How to calculate fold-change using the ∆∆CT method of relative quantification � Setting the Baseline and Threshold� Data analysis example 1: Basic Experiment� Using the free miScript miRNA PCR Array data analysis tools� Data analysis example 2: Serum miRNA Experiment
� Summary of QIAGEN’s miRNA detection portfolio� Current Promotion
Sample & Assay Technologies Canonical pathway of miRNA biogenesis
Webinar 2: Advanced miRNA Expression Analysis 4
� Transcribed by RNA Polymerase II as a long primary transcript (pri-miRNAs), which may contain more than one miRNA.
� In the nucleus, pri-miRNAs are processed to hairpin-like pre-miRNAs by RNAse III-like enzyme Drosha.
� Pre-miRNAs are then exported to the cytosol by exportin 5.
� In the cytosol RNAse III-like Dicer processes these precursors to mature miRNAs.
� These miRNAs are incorporated in RISC.
� miRNAs with high homology to the target mRNA lead to mRNA cleavage.
� miRNAs with imperfect base pairing to the target mRNA lead to translational repression and/or mRNA degradation.
Sample & Assay Technologies Why quantify miRNAs?
Webinar 2: Advanced miRNA Expression Analysis 5
� Virtually every publication includes characterization by quantification.
� Changes in miRNA can be correlated with gene expression changes in development, differentiation, signal transduction, infection, aging, and disease.
Sample & Assay Technologies miRNA quantification by real-time PCR
Webinar 2: Advanced miRNA Expression Analysis 8
� Real-time PCR quantification of miRNAs
� Has been the gold standard for gene quantification
� Is the method of choice to confirm next-generation sequencing and microarray results
� Simple and easy to carry out
� High sensitivity, specificity
� High throughput compatible, automatable
� Needs very low amounts of template
Sample & Assay Technologies miRNA expression profiling using real-time PCR
Webinar 2: Advanced miRNA Expression Analysis 9
Key considerations
Scientific question� Well-defined and testable
Experimental sample set� Statistically meaningful number of replicates (biological replicates)
� A minimum of three replicates recommended� Additional replicates may enhance statistical power
� Inclusion of proper controls
Experimental testing platform� Simple, straightforward workflow� High sensitivity and specificity� When profiling expression: a variety of PCR arrays to meet your
experimental needs� Easy and simple data analysis tools� Customizable solutions� Availability of companion research tools
Sample & Assay Technologies miScript PCR System
Webinar 2: Advanced miRNA Expression Analysis 10
Fully integrated, complete miRNA quantification sys tem
1. miScript II RT Kit� HiFlex Buffer: Unparalleled flexibility for quantification of
miRNA and mRNA from a single cDNA preparation� HiSpec Buffer: Unmatched specificity for mature
� Cel-miR-39� Alternative data normalization using exogenously spiked Syn-cel-miR-39 miScript miRNA Mimic
� miScript PCR Controls� Data normalization using the ∆∆CT method of relative quantification
� miRNA reverse-transcription control (miRTC)� Assessment of reverse transcription performance
� Positive PCR control (PPC)� Assessment of PCR performance
Sample & Assay Technologies miScript PCR Controls
Webinar 2: Advanced miRNA Expression Analysis 17
Stable expression and excellent amplification effic iencies
SNORD61
SNORD68
SNORD72
SNORD95
SNORD96A
RNU6-2
� Highly conserved across multiple species� Human, Mouse, Rat, Dog, Rhesus macaque
� Relatively stable expression in many tissues� Amplification efficiencies of these assays are 100%� Consistent performance in both HiSpec and HiFlex Buffers� Ideal normalizers for ∆∆CT method of relative quantification
� Total HeLa S3 (miRNeasy)� Pellet 1: Frozen June 2010� Pellet 2: Frozen April 2011
� HiSpec Buffered cDNA� miScript real-time PCR
� miFinder miScript miRNA PCR Array
� 1 hour� HiSpec Buffer
� 2 minutes
� 2 hours
� 15 minutes
Sample & Assay Technologies
Webinar 2: Advanced miRNA Expression Analysis 22
Real-time PCR data analysis
(1) Schmittgen TD, Livak KJ.(2008):Analyzing real-time PCR data by the comparative C(T) method. Nat Protoc.;3(6):1101-8
(2) Livak, KJ, and Schmittgen, TD.(2001): Analysis of Relative Gene Expression Data Using Real-Time Quantitative PCR and the 2-∆∆CT Method METHODS 25, 402–408
(3) www.Gene-Quantification.info
CT = 23.8
� Absolute quantification� Absolute input copies, based on a standard curve
� Relative quantification� Comparative CT method: also known as the 2-∆∆CT method� Selection of internal control� Selection of calibrator (e.g. untreated control or normal
sample)� Assumes that the PCR efficiency of the target gene is
similar to the internal control gene (and that the efficiency of the PCR is close to 100%)
� Fold change = 2-∆∆CT
– ∆CT = CTGene - CT
Normalizer
– ∆∆CT = ∆CT (sample 2) – ∆CT (sample 1) where sample 1 is the control sample and sample 2 is the experimental sample
Sample & Assay Technologies
Webinar 2: Advanced miRNA Expression Analysis 23
Data analysis workflow
Steps 1 & 2:Set Baseline and Threshold to determine C T values
Step 3:Export C T values
Step 4:Analyze data using ∆∆CT method of relative quantification
Sample & Assay Technologies
Webinar 2: Advanced miRNA Expression Analysis 24
Set Baseline and Threshold to determine C T values
Steps 1 & 2
� Baseline� Definition: Noise level in early cycles where there is no detectable increase in
fluorescence due to PCR products.� How to Set:
– Observe amplification plot using the “Linear View”– Determine the earliest visible amplification– Set the baseline from cycle 2 to 2 cycles before the earliest visible amplification– Note: The number of cycles used to calculate the baseline can be changed and should be
reduced if high template amounts are used
Sample & Assay Technologies
Webinar 2: Advanced miRNA Expression Analysis 25
Set Baseline and Threshold to determine C T values
Steps 1 & 2
� Threshold� Purpose: Used to determine the CT (threshold cycle) value. The point at which
the amplification curve intersects with the threshold line is called the CT.� How to Set:
– Observe amplification plot using the “Log View”– Place the threshold in the lower half of the log-linear range of the amplification plot, above
the background signal– Note: Never set the threshold in the plateau phase
Sample & Assay Technologies
Webinar 2: Advanced miRNA Expression Analysis 26
Applied Biosystems ® 7900HT
Setting the Baseline and Threshold
Threshold: 0.2
� Baseline: From cycle 2 (or 3) to 2 cycles before the earliest visible amplification.� Threshold: Place in the lower half of the log-linear range of the amplification plot,
miScript miRNA PCR Arrays on Roche ® LightCycler ® 480
Select Second Derivative Maximumanalysis method
Obtain C T Values
Sample & Assay Technologies
Webinar 2: Advanced miRNA Expression Analysis 29
Step 3: Export C T values
� One 5 µM FFPE section used per FFPE isolation� Each isolation is from a different section� On average, each isolation provided enough total RNA for:
– Two full human miRNome profiles– Ten pathway-focused PCR arrays
� RT: 125 ng total RNA, HiSpec Buffer� qPCR: Human miFinder miScript miRNA PCR Array (0.5 ng cDNA per well)
4
8
12
16
20
24
28
32
36
40
1 7 13 19 25 31 37 43 49 55 61 67 73
CT Value
FFPE Isolation 1
FFPE Isolation 2
FFPE Isolation 34
8
12
16
20
24
28
32
36
40
1 7 13 19 25 31 37 43 49 55 61 67 73
CT Value
FFPE Isolation 1
FFPE Isolation 2
FFPE Isolation 3
Normal Lung Lung Tumor
Sample & Assay Technologies
Webinar 2: Advanced miRNA Expression Analysis 30
∆∆CT method of relative quantification
Step 4: Analyze data
Normal (N) Lung Total RNA Lung Tumor (T) total RNA
N cDNA (Iso. 1) N cDNA (Iso. 2) N cDNA (Iso. 3) T cDNA (Iso. 1) T cDNA (Iso. 2) T cDNA (Iso. 3)
Exported C T values Exported C T valuesCalculate ∆CTfor each miRNAon each array
∆CT = CTmiRNA – AVG CT
SN1/2/3/4/5/6 ∆CT = CTmiRNA – AVG CT
SN1/2/3/4/5/6
� Tip for choosing an appropriate snoRNA/snRNA control s for normalization� Make sure that the selected controls are not influenced by the experimental conditions
Sample & Assay Technologies
Webinar 2: Advanced miRNA Expression Analysis 31
∆∆CT method of relative quantification
Step 4: Analyze data (cont.)
Normal (N) Lung Lung Tumor (T)Calculate ∆CT
for each miRNAon each array
∆CT ∆CT∆CT ∆CT∆CT ∆CT
Calculate ∆∆CT for each miRNA
between groups(T – N)
∆∆CT = Avg. ∆CT (T) – Avg. ∆CT (N)
Calculate fold-change for each miRNA (T vs. N)
2-(∆∆CT)
Calculate Average ∆CT for each miRNAwithin group (N or T)
∆CT + ∆CT + ∆CT
3
∆CT + ∆CT + ∆CT
3
Sample & Assay Technologies
Webinar 2: Advanced miRNA Expression Analysis 32
-14
-10
-6
-2
2
6
10
14
Log2 Fold-Regulation (Tumor vs. Norm
al)
Fold-Regulation: Tumor vs. Normal
� Significant differences exist between the mature miRNA expression levels of the two tissue types
∆∆CT method of relative quantification
Step 4: Analyze data (cont.)
Sample & Assay Technologies
Webinar 2: Advanced miRNA Expression Analysis 33
Incorporating the free miScript miRNA PCR Array Data Analysis Software
miScript’s straightforward, data analysis workflow
Steps 1 & 2:Set Baseline and Threshold to determine C T values
Step 3:Export C T values
Step 4:Access the free data analysis tools at
http://pcrdataanalysis.sabiosciences.com/mirna
Step 5 & on:Automatic data using ∆∆CT method of relative quantification
Sample & Assay Technologies
Webinar 2: Advanced miRNA Expression Analysis 34
Set Baseline and Threshold to determine C T values
Steps 1 & 2
� Baseline� Definition: Noise level in early cycles where there is no detectable increase in
fluorescence due to PCR products.� How to Set:
– Observe amplification plot using the “Linear View”– Determine the earliest visible amplification– Set the baseline from cycle 2 to 2 cycles before the earliest visible amplification– Note: The number of cycles used to calculate the baseline can be changed and should be
reduced if high template amounts are used
� Threshold� Purpose: Used to determine the CT (threshold cycle) value. The point at which
the amplification curve intersects with the threshold line is called the CT.� How to Set:
– Observe amplification plot using the “Log View”– Place the threshold in the lower half of the log-linear range of the amplification plot, above
the background signal– Note: Never set the threshold in the plateau phase
� Important: Ensure that baseline and threshold settings are the same across all PCR runs in the same analysis to allow comparison of results.
Sample & Assay Technologies
Webinar 2: Advanced miRNA Expression Analysis 35
Export C T values and access free analysis tools
Steps 3 & 4
� Export CT values according to the manual supplied with the real-time PCR instrument
� Access the free miScript miRNA PCR Array Data Analysis Tools� Website: http://sabiosciences.com/mirnaArrayDataAnalysis.php
– Web-based
– Excel-based
Sample & Assay Technologies
Webinar 2: Advanced miRNA Expression Analysis 36
miScript miRNA PCR Array Data Analysis Tools
Step 5: Automatic data analysis
� Web-based software� No installation needed� Tailored for each array
� Raw CT values to fold-change results� Using ∆∆CT Method
� This tab includes:� All miRNAs and controls found on
chosen array� Column where Normalization RNA
(control gene) can be selected� All CT data uploaded to software
� What should you do at this page?1. Click boxes next to desired ‘Control
Genes’ (miScript PCR Controls are pre-selected)
2. Designate columns of CT values as Control, Group 1, Group 2, etc.
3. Click ‘Update All Changes’
1. Choose ‘Control Genes’
2. Designate C T values column groups
3. Click Update All Changes
Sample & Assay Technologies
Webinar 2: Advanced miRNA Expression Analysis 39
Step 5C
miScript miRNA PCR Array Data Analysis
Readout Tab ���� View Housekeeping Genes, Data Overview, & Data QC
� View Housekeeping Genes Tab� Selected ‘Control Genes’ can be
visualized and method of normalization can be chosen
� Data Overview Tab� This tab provides an overview of all
∆∆CT calculations performed by the software
� Data QC Tab� This tab provides an overview of QC
data associated with the miRTC(reverse transcription control) and PPC (positive PCR control)
Readout Tab ���� View Housekeeping Genes
Readout Tab ���� Data QC
Choose ‘Normalized By’
Sample & Assay Technologies
Webinar 2: Advanced miRNA Expression Analysis 40
Step 5D
miScript miRNA PCR Array Data Analysis
Analysis Result Tab: this tab provides an overview of all ∆∆CT related calculations and provides a guide for you regarding the trust that you should place in your data (see red arrow)
Comments on data quality
Sample & Assay Technologies
Webinar 2: Advanced miRNA Expression Analysis 41
Step 5E
miScript miRNA PCR Array Data Analysis
Scatter Plot, Volcano Plot, Clustergram, and Multigroup Plot Tabs: When clicked, these tabs provide various statistical outputs that will open as new windows. The scatter plot is included as an example.
Scatter Plot
Sample & Assay Technologies
Webinar 2: Advanced miRNA Expression Analysis 42
Step 5F
miScript miRNA PCR Array Data Analysis
Export Data Tab: When clicked, results calculated by the miScript miRNA PCR Array Data Analysis software can be exported to Microsoft Excel.
Sample & Assay Technologies
Webinar 2: Advanced miRNA Expression Analysis 43
Serum Sample analysis using the miScript PCR System
5. (Optional) Preamplification: If the starting amount of serum or plasma is 50 µl (or less), perform preamplification using the miScript PreAMP PCR Kit
Phase I (determined expressed miRNAs):Pooled sample profiling,
Maximal Assays (miRNome or 384HC)
Phase II (determine differentially expressed miRNAs ):Individual profiling of samples (that went into poo ls),
Only Expressed miRNAs (< 384-well plate)
Phase III (statistical power):Individual profiling of all samples in study,
Differentially Expressed miRNAs from Phase II (multi ple samples per 384-well plate)
Sample & Assay Technologies
Webinar 2: Advanced miRNA Expression Analysis 46
Serum & Plasma Samples
Special data analysis case
Serum or plasma total RNA samples: The snoRNA/snRNA panel of targets does not exhibit robust expression and therefore should not be selected as Normalization Controls.
Control Serum Sample 1 Serum Sample 2 Serum Sample 3
SNORD61 36.3 34.3 35.8
SNORD68 34.6 35.0 35.3
SNORD72 35.0 35.0 35.0
SNORD95 31.1 39.3 33.5
SNORD96A 33.6 34.5 35.4
RNU6-2 37.9 39.1 35.0
Typical CT Values for miScript PCR Controls in Serum Samples
� Step 1: Calibrate samples using cel-miR-39 CT mean� Step 2: Normalize serum or plasma sample data
� Option 1: Normalize CT values to CT mean of all commonly expressed miRNAs� Option 2: Normalize CT values to CT mean of invariant miRNAs
Sample & Assay Technologies Anatomy of a miScript miRNA PCR Array
� Cel-miR-39� Alternative data normalization using exogenously spiked Syn-cel-miR-39 miScript miRNA Mimic
� miScript PCR Controls� Data normalization using the ∆∆CT method of relative quantification
� miRNA reverse-transcription control (miRTC)� Assessment of reverse transcription performance
� Positive PCR control (PPC)� Assessment of PCR performance
Sample & Assay Technologies
Webinar 2: Advanced miRNA Expression Analysis 48
Calibrate samples using cel-miR-39 C T mean
Serum or plasma sample data normalization
� Uncalibrated
Assay Sample 1 Sample 2
hsa-miR-16 16.0 19.0
hsa-miR-21 20.0 24.0
hsa-miR-192 23.0 26.0
hsa-miR-103 23.0 23.0
hsa-miR-25 22.0 25.0
cel-miR-39 18.0 21.0
� Compared to sample 1, all assays in sample 2 appear to have delayed CT values� Compared to sample 1, cel-miR-39 in sample 2 also has a delayed CT value� Conclusion: calibrate samples (cel-miR-39 CT values indicate a differential recovery)
� Calibrated (Sample 1 C T values +3)
Assay Sample 1 Sample 2
hsa-miR-16 19.0 19.0
hsa-miR-21 23.0 24.0
hsa-miR-192 26.0 26.0
hsa-miR-103 26.0 23.0
hsa-miR-25 25.0 25.0
cel-miR-39 21.0 21.0
Sample & Assay Technologies
Webinar 2: Advanced miRNA Expression Analysis 49
Option 1: C T values normalized to C T mean of expressed miRNAs
Serum or plasma sample data normalization
-8
-4
0
4
8
12
Fold-Regulation
� Calculate the C T mean for commonly expressed miRNAs� Those miRNAs with C T values < 30 (or 32, or 35) in all assessed samples
Sample & Assay Technologies
Webinar 2: Advanced miRNA Expression Analysis 50
Option 2: C T values normalized to C T mean of invariant miRNAs
Serum or plasma sample data normalization
-8
-4
0
4
8
12
Fold-Regulation
Commonly Expressed miRNAs
hsa-let-7a hsa-miR-92a
hsa-let-7c hsa-miR-93
hsa-miR-21 hsa-miR-103a
hsa-miR-22 hsa-miR-126
hsa-miR-23a hsa-miR-145
hsa-miR-24 hsa-miR-146a
hsa-miR-25 hsa-miR-191
hsa-miR-26a hsa-miR-222
hsa-miR-26b hsa-miR-423-5p
� Calculate the C T mean for invariant miRNAs� Choose at least 4 to 6 miRNAs that exhibit little C T variation
Sample & Assay Technologies
Webinar 2: Advanced miRNA Expression Analysis 51
Comparison of normalization methods
Serum or plasma sample data normalization
Option 1:Commonly Expressed miRNAs
(miRNome, 384HC)
Option 2:Invariant Panel of miRNAs
(small panel screening)
-8
-4
0
4
8
12
Fold-Regulation
-8
-4
0
4
8
12
Fold-Regulation
� Note 1: In this example, fold-regulation looks highly similar, irrespective of the chosen normalization method. This is correct, as your results should be independent of the chosen normalization method.
� Note 2: For small panel screening, do not use a CT mean of all miRNAs, as this array is biased (miRNA assays included on this array are not random)
Sample & Assay Technologies
Webinar 2: Advanced miRNA Expression Analysis 52
miRNA expression profiling
Serum or plasma sample data normalization
1.E-05
1.E-04
1.E-03
1.E-02
1.E-01
1.E+00
1.E+01
1.E
-05
1.E
-04
1.E
-03
1.E
-02
1.E
-01
1.E
+0
0
1.E
+0
1
2-∆CT
: Normal Serum2-∆CT: Breast Cancer Serum
miR-34a
14
18
22
26
30
34
38
1 6 11 16 21 26 31 36 41 46 51 56 61
miRNA
CT Value
Serum Isolation 1
Serum Isolation 2
Serum Isolation 3
Normal Serum 2 -∆CT: Breast Cancer vs. Normal
� Workflow� 200 µl serum � 50 µl total RNA � 5 µl total RNA, HiSpec Buffer � one-half of cDNA used for 96-well
miFinder miScript miRNA PCR Array
� High reproducibility can be achieved using the miRNeasy Supplementary Protocol � Each isolation was from a different normal serum donor
� Significant differences exist between the mature miRNA expression levels of the two tissue types
� ± 2-fold [red lines] used as a cutoff for significance
Sample & Assay Technologies Where can I find miScript miRNA PCR Arrays?
Webinar 2: Advanced miRNA Expression Analysis 53
www.sabiosciences.com/mirna_pcr_array.php
� miRNA Overview
� miScript PCR System
� miScript miRNA PCR Arrays
� Products for functional studies
� miRNA purification options
Sample & Assay Technologies Where can I find miScript Primer Assays?