To sequence or not to sequence is not a question anymore. BUT

Transcription

1 To sequence or not to sequence is not a question anymore. BUT Vladimír Beneš 21 June

2 More data on their way to you! High GC High GC Low GC Low GC v3 flowcell imaged area larger by 50 %! v3 sequencing chemistry Some GC-rich clusters poorly resolved/not detected at very high densities Old Cluster Amplification Larger, brighter GC-rich clusters are well resolved and detected at very high densities New Cluster Amplification

3 Increasing yield Sequence output from Illumina reads per lane Millions of reads PF Illumina Arrives 0.3 GAII GAIIx SCS 2.8, cbot, v5 kits SCS2.6, v4 kits May 08 Jun 08 Jul 08 Aug 08 Sep 08 Oct 08 Nov 08 Dec 08 Jan 09 Feb 09 Mar 09 Apr 09 May 09 Jun 09 Jul 09 Aug 09 Sept 09 Oct 09 Nov 09 Dec 09 Jan 10 Feb 10 Mar 10 Apr 10 May 10 June 10 July 10 Aug 10 Sept 10 Oct 10 Nov 10 Dec 10 Jan 11 Feb 11 Mar 11 April 11 May 11 June 11 GA HiSeq2000

4 Improving quality of called bases Average quality per cycle, RNA-Seq Quality score Cycle 36 Base mrnaseq 76 Base mrnaseq 76 Base V4 mrnaseq 105 Base V4 mrnaseq 76 Base V5 mrnaseq v5105 s_6 s7 V5 s8 v5 TruSeq 150 R1 TruSeq 150 R2

5 Challenges The only thing constant in life is change Distorted expectations of users Data ( massive amounts, formats ) Interpretation of results (suboptimal experimental design; is everything relevant?) Incomplete understanding of sources of error and bias in MPS data

6 Bias is never good

7

8 Hype/hope curve

9 MPS space PNAS (2010) Kahvejian, Nat Biotech (2008)

10 Available MPS applications transcriptome RNA-Seq, Tag-Seq mirnome smallrna-seq protein-na interactions ChIP-Seq, CLIP-Seq epigenome Methyl-Seq de novo & re-sequencing Metagenomics Genome capture, multiplexing yes yes yes yes yes yes yes

11 MPS methods used in epigenomics Rada-Iglesias & Wysocka, Genome Medicine (2011)

12 Portela & Esteller, Nature Biotechnology (2011)

13 Importance of experimental design What do I want to study? "Would you tell me, please, which way I ought to go from here?" "That depends a good deal on where you want to get to," said the Cat. "I don t much care where--" said Alice. "Then it doesn t matter which way you go," said the Cat. "--so long as I get SOMEWHERE," Alice added as an explanation. "Oh, you re sure to do that," said the Cat, "if you only walk long enough. Lewis Carroll s Alice in Wonderland

14 Which sequencing mode to use? Sequencing type Exon capture Whole genome sequencing Recommendation 50Mb Kit, human: 105b SR to get sufficient coverage Large rearrangements: Mate-pairs large insert Resequencing: SNPs/indels: Coverage is good 100+ PE. If you don t get the coverage at the start you ll regret it. Coverage is the key! RNA-Seq Chip-Seq Multiplexing Tag counting: large number of mappable tags 36-50b SR should suffice. Longer reads may need to be trimmed to match exons. Transcriptome assembly or exon usage: 75+ single or 75+ PE depending on a de novo/spliced read mapping approach or map pairs to detect also alternative splicing. Strand-specific libraries: complex insight into transcriptome 36b SR unless you have real concerns about alignability of your target (i.e. some strange looking enhancer region) Coverage is the key!

15 MPS workflow 1) Library preparation 2) Cluster generation on a flow cell SE, PE reads, bases 3) Sequencing & imaging 4) Data processing & analysis

16 MPS library preparation 5 AATGATACGGCGACCACCGA-ACACTCTTTCCCTACACGACGCTCTTCCGATCT--INSERT--TCGTATGCCGTCTTCTGCTTG TTACTATGCCGCTGGTGGCT-TGTGAGAAAGGGATGTGCTGCGAGAAGGCTAGA--INSERT--AGCATACGGCAGAAGACGAAC5 where 5 AATGATACGGCGACCACCGA 5 CAAGCAGAAGACGGCATACGA is the P5 attachment/amplification primer sequence is the P7 attachment/amplification primer sequence 5 ACACTCTTTCCCTACACGACGCTCTTCCGATCT is the SBS3 sequencing primer sequence INSERT is a complex mix of DNA fragments

17 Forked adapters A Insert Insert A Adapter ACACTCTTTCCCTACACGAC TCGTATGCCGTCTTCTGCTTG GCTCTTCCGAT C T P- GATCGGAAGAGC CGAGAAGGCTA G -P T CTAGCCTTCTCG GTTCGTCTTCTGCCGTATGCT CAGCACATCCCTTTCTCACA Adapter

18 Library preparation II. A conventional protocol A nextera protocol at least 500 ng of gdna <50 ng of gdna! Adey et al., Genome Biology (2010)

19 Library preparation III. Strict QC of starting material (GiGo) Qubit quantification gel images, bioanalyzer/experion traces

20 Library preparation IV. Bioruptor, probe (ChIP-Seq) Covaris vs nebulization Kits (proprietary, home-brewed, NEB!) Size selection using gel extractor, E- gel, Pippin prep, SPRIworks, Lo-bind tubes! Covaris Hydroshear

21 RNA-Seq libraries rrna depletion (oligo-dt beads, Ribo-Minus, Ribo-Zero ) BUT mitochondria-derived rrna mostly ignored!!? 18S 28S 10: RiboMinus (UHRR) prep 2 3. RiboZero (UHRR) Anu 10/12/09 #3 Both from Experiment 357;run 10/13/09 strand-specific library, Levine et al., Nat Meth (2010)

22

23

24 Library quantification & QC Qubit Bioanalyzer HS DNA Chip DNA 1000 Chip

25 Methyl-Seq Carey et al., Drug Discovery Today (2011) Zilbermann & Henikoff, Development (2007)

26 Pacific Biosciences Detection volume Modified bases By courtesy of Pacific Bioscience

27 Direct detection of methylated bases Flusberg et al. Nature Methods (2010)

28 Data integration Pepke et al., Nature Methods (2009)

29 Where are we heading?

30 Nucleic acids detection and sequencing techniques Kahvejian et al., Nat Biotech (2008)

31 Caution required available information (clues) amount state of our knowledge (answers) time We are drowning in information and starving for knowledge. Rutherford D. Roger

32 MPS features Unprecedented discovery power Hypothesis-free Almost unbiased results Sensitivity & specificity For tag-counting applications truly wholegenome, -transcriptome, -methylome view Only one source of technology noise

33 Acknowledgments Bettina Haase Dinko Pavlinic Jens Stolte Jonathon Blake Tobias Rausch Jürgen Have Zimmermann a nice day! All our users and former colleagues Science is built with facts as a house is with stones, but a collection of facts is no more a science than a heap of stones is a house. Jules Henri Poincare