Introduction to transcriptome analysis using High Throughput Sequencing technologies (HTS)

Transcription

1 Introduction to transcriptome analysis using High Throughput Sequencing technologies (HTS)

2

3 A typical RNA Seq experiment Library construction

4 Protocol variations Fragmentation methods RNA: nebulization, hydrolysis cdna: sonication, Dnase I treatment Depletion of highly abundant transcripts Positive selection of mrna. Poly(A) selection. Negative selection. (RiboMinusTM) Strand specificity Most RNA sequencing is not strand-specific Single-end or Paired-end sequencing

5 Single and paire end sequencing

6 Microarrays versus RNA seq RNA-seq Counting Absolute abundance of transcrits All transcrits are present and can be analyzed mrna / ncrna (snorna, lncrna, erna,mirna,...) Analysis of allele-specific gene expression Detection of fusions and other structural variations Microarrays Indirect record of expression level (complementary probe) Relative abundance Cross-hybridization Content limited (can only show you what you're already looking for)

7 High reproducibility and dynamic range (a) Comparison of two brain technical replicate RNASeq determinations for all mouse gene models (from the UCSC genome database), measured in reads per kilobase of exon per million mapped sequence reads (RPKM), which is a normalized measure of exonic read density; R2 = (c) Six in vitro synthesized reference transcripts of lengths kb were added to the liver RNA sample ( to transcripts per sample; R2 > 0.99).

8 RNA Seq drawback Current disadvantages More expensive than standard expression arrays More time consuming than any microarray technology Some (lots of) data analysis issues Computing accurate transcript models Mapping reads to splice junctions Contribution of high-abundance RNAs (eg ribosomal) could dilute the remaining transcript population; sequencing depth is important

9 Do arrays and RNA Seq tell a consistent story? Do arrays and RNA-Seq tell a consistent story? The relationship is not quite linear but the vast majority of the expression values are similar between the methods. Scatter increases at low expression as background correction methods for arrays are complicated when signal levels approach noise levels. Similarly, RNASeq is a sampling method and stochastic events become a source of error in the quantification of rare transcripts Given the substantial agreement between the two methods, the array data in the literature should be durable Comparison of array and RNA-Seq data for measuring differential gene expression in the heads of male and female D. pseudoobscura

10 Sequencing technologies RNA-Seq has been performed using SOLiD, Illumina and 454 platform However, RNA-Seq needs high dynamic range SOLiD and Illumina sequencers Need PCR amplification Low coverage of GC rich transcrits Future? Helicos No PCR amplification Direct RNA sequencing seem possible High error rates Pacific Biosciences

11 Current technologies Roche 454 (pyrosequencing) Genome Analyzer IIx Illumina (sequencing by synthesis) ABI SOLiD (Seq. by Oligo Ligation/Detection, sequencing by ligation

12 Illumina

13 Illumina Genome Analyzer (I)

14 Illumina Genome Analyzer (II)

15 Illumina Genome Analyzer (III)

16 Illumina Genome Analyzer

17 ABI Next Gen Sequencing: SOLiD

18 ABI Next Gen Sequencing: SOLiD Use empcr on magnetic beads Sequencing by ligation using fluorescent probes

19 Sequencing by ligation (SOLiD) We have a flow cell (basically a microscope slide that can be serially exposed to any liquids desired) whose surface is coated with thousands of beads each containing a single genomic DNA species, with unique adapters on either end. Each microbead can be considered a separate sequencing reaction which is monitored simultaneously via sequential digital imaging. Up to this point all next-gen sequencing technologies are very similar, this is where ABI/SOLiD diverges dramatically (see figure 4)

20 Sequencing by ligation (SOLiD)

21 Sequencing by ligation (SOLiD)

22 Sequencing by ligation (SOLiD) According to colour code table four different dinucleotides may correspond to the same colour. This ambiguity is resolved using primer n-1 round The first ligation of this primer round analyses dinucleotide with one known nucleotide

23 Sequences are in color space format (cfasta) >853_64_861_F3 T Dibases: TT TT TG GG... Sequence: TTGG...

24 Advantage of 2 base encoding Higher accuracy for SNP detection

25 Typical analysis pipeline Raw sequence data Pre-processing Removing actifact poor quality reads, sequencing adaptors, low-complexity reads, near-identical reads (PCR biases) Assembly 1. Reference-based 2. De novo strategy 3. Combined strategy (1 & 2) Summarize to gene/transcrit Statistical analysis : Diff. Expressed genes

26 Overview of the reference based transcriptome assembly a Reads (grey) are first aligned to a reference genome using spliceaware aligner, such as TopHat or SpliceMap b A connectivity or splice graph is then constructed to represent all possible isoforms at a locus. c,d Finally, alternative paths through the graph (blue, red, yellow and green) are followed to join compatible reads together into isoforms.

27 Summarize to gene expression level Transcrits of different length have different read count Tag count is normalized for transcrit length and total read number in the measurement (RPKM, Reads Per Kilobase of exon model per Million mapped reads) 1 RPKM corresponds to approximately one transcript per cell FPKM, Fragments Per Kilobase of exon model per Million mapped reads (paired-end sequencing)