Di ff erential gene expression analysis using RNA-seq Applied - PowerPoint PPT Presentation

Di ff erential gene expression analysis using RNA-seq Applied Bioinformatics Core https://abc.med.cornell.edu/

Day 1: Introduction 1. RNA isolation & library preparation 2. Illumina’s sequencing by synthesis 3. experimental design 4. raw sequencing reads • download • quality control

RNA-seq is popular, but still developing “RNA%seq)is) not$a$mature$technology .) It$is$ undergoing$rapid$evolution$ of)biochemistry) of)sample)preparation;)of)sequencing) platforms;)of)computational)pipelines;)and) of$ subsequent$analysis$methods$that$include$ statistical$treatments )and)transcript)model) building.)“) ENCODE&consortium& Reuter et al. (Mol Cell, 2015). High-Throughput Sequencing Technologies. doi:10.1016/j.molcel.2015.05.004

What to expect from the class Experimental design Sample type • Controls & quality • No. of replicates NOT COVERED: • Randomization • novel transcript discovery Library preparation Biological question • transcriptome • Poly-A enrichment • Expression quantification vs. ribo minus assembly • Alternative splicing • Strand information • alternative splicing • De novo assembly needed analysis • mRNAs, small RNAs • …. (see the course notes for references to useful Bioinformatics reviews) • Aligner • Normalization Sequencing • DE analysis strategy • Read length • PE vs. SR • Sequencing errors

RNA-seq workflow overview Total RNA e otal RNA extr xtraction action cells RNA • Fragmentation • mRNA enrichment fragments • Library preparation cDNA with adapters Sequencing Sequencing • Cluster generation • Sequencing by synthesis • Image acquisition Bioinformatics Bioinformatics

Quality control of RNA extraction 28S:18S ratio avoid degraded RNA junk

RNA-seq library preparation RNA extraction poly(A) enrichment or rRNA depletion/mRNA QC! enrichment ribo-depletion fragmentation random priming and reverse transcription 3’ adapter ligation second strand synthesis 5’ adapter ligation U U U U U U U reverse end repair, A- end repair, A- transcription addition, adapter addition, adapter ligation ligation U U PCR PCR PCR classical Illumina protocol dUTP stranded library sequential ligation of two (unstranded) preparation di ff erent adapters Van Dijk et al. (2014).. Experimental Cell Research, 322(1), 12–20. doi:10.1016/j.yexcr.2014.01.008

RNA-seq workflow overview Total RNA e otal RNA extr xtraction action cells RNA fragments cDNA with adapters Sequencing Sequencing flowcell with primers http://informatics.fas.harvard.edu/test-tutorial-page/

Cluster generation bridge amplification denaturation cluster generation removal of complementary strands ! identical fragment copies remain http://informatics.fas.harvard.edu/test-tutorial-page/

Sequencing by synthesis labelled dNTP 1. extend 1 st base 2. read generate base calls repeat for 50 – 100 bp 3. deblock Image from Illumina

How deep is deep enough? for DGE (logFC~ 2) in mammals: 20 – 50 20 – 50 mio mio SR, 75 SR, 75 bp bp Goals that require more , longer and possibly paired- end reads: • quantification of lowly expressed genes • identification of genes with small changes between conditions • investigation of alternative splicing /isoform quantification • identification of novel transcripts , chimeric transcripts • de novo transcriptome assembly

Many sources for biased signals Comparison of the transcriptional landscapes between human and mouse tissues Shin Lin a,b,1 , Yiing Lin c,1 , Joseph R. Nery d , Mark A. Urich d , Alessandra Breschi e,f , Carrie A. Davis g , Alexander Dobin g , Christopher Zaleski g , Michael A. Beer h , William C. Chapman c , Thomas R. Gingeras g,i , Joseph R. Ecker d,j,2 , and Michael P. Snyder a,2 a Department of Genetics, Stanford University, Stanford, CA 94305; b Division of Cardiovascular Medicine, Stanford University, Stanford, CA 94305; c Department of Surgery, Washington University School of Medicine, St. Louis, MO 63110; d Genomic Analysis Laboratory, The Salk Institute for Biological Studies, La Jolla, CA 92037; e Centre for Genomic Regulation and UPF, Catalonia, 08003 Barcelona, Spain; f Departament de Ciències Experimentals i de la Salut, Universitat Pompeu Fabra, 08003 Barcelona, Spain; g Functional Genomics, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY 11742; h McKusick-Nathans Institute of Genetic Medicine and the Department of Biomedical Engineering, Johns Hopkins University, Baltimore, MD 21205; i Affymetrix, Inc., Santa Clara, CA 95051; and j Howard Hughes Medical Institute, The Salk Institute for Biological Studies, La Jolla, CA 92037 Contributed by Joseph R. Ecker, July 23, 2014 (sent for review May 23, 2014) major projects. Thirteen of the mouse and human orthologous “Overall,)our)results)indicate)that) there)is) considerable$RNA$ expression$diversity$between$ humans$and$mice ,)well)beyond) what)was)described)previously,) likely)reBlecting)the)fundamental) physiological)differences)between) these)two)organisms.)“)

Batch e ff ect deluxe Leek et al. (2010). Nature Reviews. Genetics, 11(10), 733– 739. doi:10.1038/nrg2825 Once)we)accounted)for)the)batch)effect) Gilad Y and Mizrahi-Man O 2015. (…),)the)comparative)gene)expression) F1000Research 2015, 4:121 (doi: 10.12688/ data)no)longer)clustered)by)species,)and) f1000research.6536.1) instead,)we)observed)a)clear)tendency) for)clustering)by)tissue.)

Many more sources for biased signals • sequencing errors • miscalled bases • PCR artifacts no. of reads • length bias • GC bias • duplicates • issues with the reference • CNV • mappability GC content • inappropriate data processing inclusion of multi-mapped reads exclusion of multi-mapped reads

Invest in replicates! • recommended: 6 biological replicates per condition for DGE of strongly changing genes (logFC >= 2) (Gierlinski et al., 2015) Gene X 10.26 10.24 log2 Counts 10.22 10.20 10.18 from all “clean” 10.16 “clean” condition 1 condition 2 “bad” gene’s expression in an individual replicate with the trimmed mean across all repli- 𝑜 𝑢 𝑦̅ 𝑕;𝑜 𝑢 𝑡 𝑕;𝑜 𝑢 |𝑦 𝑕𝑗 − 𝑦̅ 𝑕;𝑜 𝑢 | > 𝑜 𝑡 𝑡 𝑕;𝑜 𝑢 𝑜 𝑡 𝑔 𝑗 𝑗 𝑜 𝑢 = 3 𝑜 𝑡 = 5 𝑜 𝑢 𝑜 𝑡 𝑜 𝑡 = 3 – the read depth profiles between clean and “bad” replicates is shown in “bad” replicate “clean” replicates “ ”

Replicates Technical replicates sequencing lane library prep sequencing lane RNA extraction sequencing lane library prep sequencing lane Biological replicates RNA from an independent growth of cells/tissue sequencing lane RNA extraction sequencing lane library prep sequencing lane RNA extraction sequencing lane

Don’t put all your eggs into the same basket! ! multiplex all samples! Auer, P. L., & Doerge, R. W. (2010). Genetics, 185(2), 405–16.

RNA-seq workflow overview Total RNA e otal RNA extr xtraction action cells RNA • Fragmentation • mRNA enrichment fragments • Library preparation cDNA with adapters Sequencing Sequencing • Cluster generation • Sequencing by synthesis • Image acquisition Bioinformatics Bioinformatics

Bioinformatics workflow of RNA-seq analysis Images Raw reads List of fold changes & statistical values

Bioinformatics workflow of RNA-seq analysis Images Base calling & demultiplexing .tif Bustard/RTA/OLB, CASAVA FASTQC Raw reads Raw reads Mapping .fastq STAR Aligned reads Counting .sam/.bam featureCounts Read count table Normalizing .txt DESeq2, edgeR Normalized read count table DE test & multiple testing correction .Robj DESeq2, edgeR, limma List of fold changes & statistical values Filtering .Robj, .txt Customized scripts Downstream analyses on DE genes

Where are all the reads? Sequence Read Archive GenBank DDBJ http://www.ncbi.nlm.nih.gov/genbank/ http://www.ddbj.nig.ac.jp/intro-e.html ENA https://www.ebi.ac.uk/ena/

Let’s download! • Gierlinski et al. have submitted their data set to SRA – we will retrieve it via ENA ( https://www.ebi.ac.uk/ena/ ) • accession number: ERP004763 ls mkdir wget cut grep awk

FASTQ file format = FASTA + q uality scores 1 read " 4 lines! ⌥ 1 @ERR459145 .1 DHKW5DQ1 :219: D0PT7ACXX :2:1101:1590:2149/1 2 GATCGGAAGAGCGGTTCAGCAGGAATGCCGAGATCGGAAGAGCGGTTCAGC 3 + 4 @7 <DBADDDBH?DHHI@DH > HHHEGHIIIGGIFFGIBFAAGAFHA ’5? B@D 1. @Read ID and sequencing run information 2. sequence 3. + (additional description possible) 4. quality scores

Base quality score ⌥ @ERR459145 .1 DHKW5DQ1 :219: D0PT7ACXX :2:1101:1590:2149/1 GATCGGAAGAGCGGTTCAGCAGGAATGCCGAGATCGGAAGAGCGGTTCAGC + @7 <DBADDDBH?DHHI@DH > HHHEGHIIIGGIFFGIBFAAGAFHA ’5? B@D base error probability p, e.g. 10e-4 ! -10 x log10(p) Phred score, e.g.: 40 turn score into ASCII symbol “FASTQ score”, e.g.: ( http://www.ascii-code.com/ ?