class: center, middle, inverse, title-slide .title[ # RNA-seq experimental considerations ] .author[ ### Mikhail Dozmorov ] .institute[ ### Virginia Commonwealth University ] .date[ ### 2025-09-22 ] --- <!-- HTML style block --> <style> .large { font-size: 130%; } .small { font-size: 70%; } .tiny { font-size: 40%; } </style> ## Library preparation steps - **RNA isolation and QC**, to extract RNA relevant to the experimental question - **Fragmentation**, to recover short reads across full length of long genes - **Size selection**, suitable for RNA sequencing. 300-500bp - mRNA, 20-150bp - small/miRNA - **Amplification**, typically by PCR. Up to `\(0.5-10ng\)` of RNA - **Library normalization/Exome capture** - **Barcoding** and **multiplexing** - Optionally, add **External RNA Control Consortium (ERCC) spike-in controls** - **Single** or **paired end** sequencing. The latter is preferrable for the _de novo_ transcript discovery or isoform expression analysis .small[ Sample preparation and library construction strategies: http://journals.plos.org/ploscompbiol/article/file?type=supplementary&id=info:doi/10.1371/journal.pcbi.1004393.s005 ] --- ## RNA isolation **Ribosomal RNA (rRNA) depletion** - `\(0.1-1\mu g\)` original total RNA (One cell contains ~10 picogram of total RNA) - rRNAs constitute over 90 % of total RNA in the cell, leaving the 1–2 % comprising messenger RNA (mRNA) that we are normally interested in (One cell contains ~0.1 picogram mRNA) - Enriches for mRNA + long noncoding RNA. - Hybridization to bead-bound rRNA probes --- ## RNA isolation - **Poly(A) selection (for eukaryotes only)** - Enrich for mRNA. - Hybridization to oligo-dT beads - **Small RNA extraction** - Specific kits required to retain small RNAs - Optionally, size-selection by gel &nbsp; .small[ Description of RNA-seq library enrichment strategies: http://journals.plos.org/ploscompbiol/article/file?type=supplementary&id=info:doi/10.1371/journal.pcbi.1004393.s006 ] --- ## Poly-A selection or ribosome depletion protocol? - Poly-A excels at gene quantification for classification/prediction purposes, better represents total RNA content - Ribosomal depletion - more noncoding RNAs, better alignment of reads, more gene fusion events - Overall, comparable performance &nbsp; .small[ Detailed comparison of RNA-seq library construction protocols: https://bmcgenomics.biomedcentral.com/articles/10.1186/s12864-017-4039-1 ] --- ## RNA quality Agilent 2100 bioanalyzer. RIN - RNA integrity number (should be >7) <img src="img/agilent.png" width="650px" style="display: block; margin: auto;" /> --- ## Unstranded vs. Strand-Specific Libraries * **Unstranded** * Random hexamer priming to reverse-transcribe mRNA. * Cannot distinguish which DNA strand the transcript originated from. * **Strand-specific** * **dUTP method:** During second-strand cDNA synthesis, dTTP is replaced with **dUTP (deoxyuridine triphosphate)**. * The strand containing dUTP is selectively degraded by Uracil-DNA Glycosylase (UDG). * **RT-qPCR (Reverse Transcription quantitative PCR):** Uses tagged primers during reverse transcription to capture strand orientation. .small[ Strand-related settings for RNA-seq tools: http://journals.plos.org/ploscompbiol/article/file?type=supplementary&id=info:doi/10.1371/journal.pcbi.1004393.s007 ] --- ## Unstranded vs. Strand-specific library <img src="img/strand.png" width="800px" style="display: block; margin: auto;" /> <!-- ## External RNA Control Consortium (ERCC) spike-in controls - A set of RNA standards for RNA-seq - 92 polyadenylated transcripts that mimic natural eukaryotic mRNAs - Designed to have a wide range of lengths (250–2,000 nucleotides) and GC-contents (5–51%) and can be spiked into RNA samples before library preparation at various concentrations (106-fold range) - ERCC spike-in controls can be used for normalization in the context of a global expression shift - Count the number of cells in each sample - Add the ERCC spike-in sequences to each sample in proportion to the number of cells - Normalize read counts based on cyclic loess robust local regression on the spike-in counts .small[ Baker, S.C. et al. The external RNA controls consortium: a progress report. Nat. Methods 2, 731–734 (2005). Jiang, L. et al. Synthetic spike-in standards for RNA-seq experiments. Genome Res. 21, 1543–1551 (2011). Loven, J. et al. Revisiting global gene expression analysis. Cell 151, 476–482 (2012). --> --- class: center,middle # Experimental design --- ## Sources of variability in RNA-seq measures In RNA-seq, we have multiple levels of randomness: - Biological variability in samples - Stochasticity of RNA content - Randomness of fragments being sequenced - Technical variability &nbsp; .small[ Auer, P.,RW Doerge. "Statistical Design and Analysis of RNA Sequencing Data." Genetics, 2010 https://doi.org/10.1534/genetics.110.114983 ] --- ## RNA-seq consideraations - **Replication**. It allows the experimenter to obtain an estimate of the experimental error - **Randomization**. It requires the experimenter to use a random choice of every factor that is not of interest but might influence the outcome of the experiment. Such factors are called nuisance factors - **Blocking**. Creating homogeneous blocks of data in which a nuisance factor is kept constant while the factor of interest is allowed to vary. Used to increase the accuracy with which the influence of the various factors is assessed in a given experiment - **Block what you can, randomize what you cannot** --- ## Experimental design: Multiplexing balances technical variability <img src="img/rna-seq_design.jpg" width="600px" style="display: block; margin: auto;" /> --- ## Sequencing length/depth - Longer reads improve mappability and transcript quantification - More transcripts will be detected and their quantification will be more precise as the sample is sequenced to a deeper level - Up to 100 million reads is needed to precisely quantify low expressed transcripts. - In reality, 20-30 million reads is OK for human genome. <!-- ## Power: effect size Statistical power to detect differential expression increases with increased effect size, sequencing depth and number of replicates | Replicates per group | 3 | 5 | 10 | |--------------------------------------|------|------|-------| | **Effect size (fold change)** | | | | | 1.25 | 17 % | 25 % | 44 % | | 1.5 | 43 % | 64 % | 91 % | | 2 | 87 % | 98 % | 100 % | &nbsp; .small[ Source: http://genomebiology.biomedcentral.com/articles/10.1186/s13059-016-0881-8 ## Power: Sequencing depth Statistical power to detect differential expression increases with increased effect size, sequencing depth and number of replicates | Replicates per group | 3 | 5 | 10 | |--------------------------------------|------|------|-------| | **Sequencing depth (millions of reads)** | | | | | 3 | 19 % | 29 % | 52 % | | 10 | 33 % | 51 % | 80 % | | 15 | 38 % | 57 % | 85 % | &nbsp; .small[ Source: http://genomebiology.biomedcentral.com/articles/10.1186/s13059-016-0881-8</div> --> --- ## RNASeqPower: Sample Size & Power Analysis for RNA-seq - `RNASeqPower` is an R package designed to **calculate statistical power and sample size** for RNA-seq experiments. - Compute **power for a given sample size**, effect size, and sequencing depth. - Estimate **minimum sample size required** to achieve desired power. - Accounts for **overdispersion** and variance typical of RNA-seq count data. .small[ Bioconductor: https://bioconductor.org/packages/RNASeqPower/ Svensson, V. et.al. "Power Analysis of Single-Cell RNA-Sequencing Experiments." _Nature Methods_ 2017 https://doi.org/10.1038/nmeth.4220 ]