Bioinformatics for nextgeneration DNA sequencing Gabor T Marth

Bioinformatics for next-generation DNA sequencing Gabor T. Marth Boston College Biology Department BC Biology new graduate student orientation September 2, 2008

Genetic code (DNA) AGCGTGGTAGCGCGAGTTTGCGAGCTAGGCTCCGGATGCGA CCAGCTTTGATAGATGAATATAGTGTGCGCGACTAGCTGTGTGTT GAATATATAGTGTGTCTCTCGATATGTAGTCTGGATCTAGTGTTG GTGTAGATGGAGATCGCGTAGCGTGGTAGCGCGAGTTTGCGAGCTAGGCTCCGGATGCGACCAGCTTTGATAGATGAATATAGTGT GCGCGACTAGCTGTGTGTTGAATATATAGTGTGTCTCTCGATATGT AGTCTGGATCTAGTGTTGGTGTAGATGGAGATCGCGTGCTTGAG TCGTTTTTTTATGCTGATGATATAAATATATAGTGTTGGTG GGGGGTACTCTCTCTAGAGCCTCTCAAAAAGCT CGGGGATCGGGTTCGAAGAAGTGAGATGTACGCGCTAGXTAGTAT ATCTCTTTCTCTGTCGTGCTGCTTGAGATCGTTTTTTTATGCT GATGATATAAATATATAGTGTTGGTGGGGGGTACTCTCTCT AGAGCCTCTCAAAAAGCTCGGGGATCGGGTTCGAAGA AGTGAGATGTACGCGCTAGXTAGTATATCTCTTTCTCTGTCGTGCT

The genome

Genome sequencing ~1 Mb ~100 Mb >100 Mb ~3, 000 Mb

bases per machine run Next-generation sequencing machines Illumina, AB/SOLi. D short-read sequencers 1 Gb (1 Gb in 25 -50 bp reads) 100 Mb 454 pyrosequencer (20 -100 Mb in 100 -250 bp reads) 10 Mb ABI capillary sequencer 1 Mb 10 bp 100 bp 1, 000 bp read length

Individual human resequencing

Variations at every scale of genome organization Single-base substitutions (SNPs) Structural variations including largescale chromosomal rearrangements Insertion-deletion polymorphisms Epigenetic variations (e. g. changes in methylation / chromatic structure)

We care about genetic variations because… … they underlie phenotypic differences … cause heritable diseases and determine responses to drugs … allow tracking ancestral human history

Individual resequencing / SNP discovery (ii) micro-repeat analysis REF IND (iii) read mapping (iv) read assembly IND (v) SNP and short INDEL calling (i) base calling (vii) data validation, hypothesis generation

Tools

The variation discovery “toolbox” • base callers • read mappers • SNP callers • SV callers • assembly viewers

Base calling Quinlan et al. Nature Methods 2008

Read mapping is like doing a jigsaw puzzle… …you get the pieces… … and they give you the picture on the box Problem is, some pieces are easier to place than others…

Read mapping Michael Stromberg in prep.

SNP discovery Marth et al. Nature Genetics 1999 Quinlan et al. in prep.

Structural variation discovery Navigation bar Fragment lengths in selected region Depth of coverage in selected region Stewart et al. in prep.

Assembly viewers Huang and Marth Genome Research 2008

Data mining

SNP calling in single-read 454 coverage DNA courtesy of Chuck Langley, UC Davis • collaborative project with Andy Clark (Cornell) and Elaine Mardis (Wash. U. ) • goal was to assess polymorphism rates between 10 different African and American melanogaster isolates • 10 runs of 454 reads (~300, 000 reads per isolate) were collected

Mutational profiling in deep 454 data Pichia stipitis reference sequence Image from JGI web site • collaboration with Doug Smith at Agencourt • Pichia stipitis is a yeast that efficiently converts xylose to ethanol (bio-fuel production) • one specific mutagenized strain had especially high conversion efficiency • goal was to determine where the mutations were that caused this phenotype • we analyzed 10 runs (~3 million reads) of 454 reads (~20 x coverage of the 15 MB genome) • processed the sequences with our 454 pipeline • found 39 mutations (in as many reads in which we found 650 K SNP in melanogaster) • informatics analysis in < 24 hours (including manual checking of all candidates) Smith et al. Genome Research 2008

SNP calling in short-read coverage C. elegans reference genome (Bristol, N 2 strain) Bristol, N 2 strain (3 ½ machine runs) Pasadena, CB 4858 (1 ½ machine runs) • goal was to evaluate the Solexa/Illumina technology for the complete resequencing of large model-organism genomes • 5 runs (~120 million) Illumina reads from the Wash. U. Genome Center, as part of a collaborative project lead by Elaine Mardis, at Washington University • we found 45, 000 SNP with very high validation rate SNP Hillier et al. Nature Methods 2008

Current focus

1000 Genomes Project • data quality assessment • project design (# samples depth of read coverage) • read mapping • SNP calling • structural variation discovery

SV discovery in autism deletion amplification

Transcriptome sequencing (from: Mortazavi et al. Nature Methods 2008)

Lab

The team Michael Stromberg Michele Busby Eric Tsung Aaron Quinlan Derek Barnett Chip Stewart Damien Croteau-Chonka Weichun Huang

Resources • computer cluster • 128 GB RAM server • 20 TB disk space • 2 large R 01 grants from the NIH • a BC RIG grant

Collaborations Genome Canada Baylor HGSC Wash. U. GSC UC Davis UBC GSC UCSF UCLA Cornell NCBI @ NIH NCI @ NIH Marshfield Clinic Pfizer

Graduate student rotations • Looking for new graduate students • Spots are available for all three rotations • Lots or projects • Caveat: you need to be able to program… • Check us out at: http: //bioinformatics. bc. edu/marthlab/ • If you are interested, please talk to me