Illumina, Inc. · NASDAQ:ILMN
Illumina — Technical Presentation (October 2006)
39 slides from Illumina's technical presentation, Oct 30, 2006
Slide 1: Capturing Common, Multi-Ethnic Human Variation On A Single Microarray
making sense out of life Capturing Common, Multi-Ethnic Human Variation On A Single Microarray Sarah Shaw Murray, Ph.D. © 2006 Illumina, Inc. Illumina, Sentrix, Array of Arrays, BeadArray, DASL, Infinium, GoldenGate, BeadXpress, VeraCode, iSelect, IntelliHyb, CSPro, and Making Sense Out of Life are
Slide 2: Talk Overview
Talk Overview Characteristics of complex disease traits Whole genome association as a tool for detecting genetic components of complex disease traits Requirements for whole-genome association studies – Access to well-phenotyped case and matched control populations – Ability to assay tag SNPs; access
Slide 3: Question: How Genetic Variation Contributes to Complex Disease?
Question: How Genetic Variation Contributes to Complex Disease? CANCER CARDIO- VASCULAR OBESITY ALZHEIMER’S ARTHRITIS 3
Slide 4: Genetics or Environment?
Genetics or Environment? Over 1,000 genes are known where a mutation is both necessary and sufficient to cause a Mendelian disease The common diseases are not so simple – Genetics is involved, because family history is strong predictor of risk – No one gene “causes” the disease For type 1 diabetes
Slide 5: Common Disease, Common Variant Hypothesis
Common Disease, Common Variant Hypothesis The genetic component of common diseases results from the simultaneous occurrence of several contributing gene variants These gene variants are relatively common in the human population WORKING DEFINITION THE CHALLENGE THE HOPE “Common” means ≥ 5% Discover t
Slide 6: The Road to Whole Genome Association
The Road to Whole Genome Association Risch N and Merikangas K. Science. 1996 273:1516- 1517. More power to detect small gene effects in Illumina SNP HumanHap300 HumanHap650Y complex disease traits Linkage Panel 317,503 PhaseI Tag 655,352 PhaseI+II Tag using association study launched SNPs launched S
Slide 7: Whole Genome Association Study Design for Genetic Studies
Whole Genome Association Study Design for Genetic Studies Advantages – Detects smaller gene effects in complex disease traits – Association defines relatively small region (1 – few genes) – Does not require a priori knowledge for genes involved in disease – Can test for interactions of many genes Di
Slide 8: Key Needs for Whole Genome Association Studies
Key Needs for Whole Genome Association Studies Access to well-phenotyped case and matched control populations Ability to assay tag SNPs; access to vast majority of genome to provide high genomic coverage and power Genotype 100,000’s of loci accurately providing complete and accurate data – High locu
Slide 9: Genome-Wide Association Studies
Genome-Wide Association Studies SNP PANEL OF TEST EVERY SNP FOR ASSOCIATION ~100,000 - >600,000 IN STUDY POPULATION MARKERS PATIENT CONTROL P-VALUE n=1000 n=1000 GENOTYPE PATIENTS 25% AF 15% AF 2 X 10-8 WITH SNP PANEL ASSOCIATION INDICATES THAT A IDENTIFY CAUSATIVE NEARBY CANDIDATE GENE MAY PATIENT
Slide 10: www.hapmap.org
www.hapmap.org International I International HapMap Project PY | English | Francais | 82885 | Yoruba [Genotype Access] [Register] 3 The International HapMap Project is a partnership of scientists and funding agencies from Canada, China, Japan, Nigeria, the United Kingdom and the United States to dev
Slide 11: Objectives of the HapMap Project
Objectives of the HapMap Project Describe the common patterns of sequence variation in the human genome Include multiple populations with ancestry from parts of Africa, Asia and Europe Make this information freely available in the public domain Provide tools to aid discovery of genetic variants that
Slide 12: A haplotype map of the human genome
ARTICLES A haplotype map of the human genome The International HapMap Consortium" Inherited genetic variation has a critical but as yet largely uncharacterized role in human disease. Here we report a public database of common variation in the human genome: more than one million single nucleotide pol
Slide 13: Some Lessons from the HapMap Project
Some Lessons from the HapMap Project Variation in local recombination rates is a major determinant of linkage disequilibrium (LD) Breakdown of LD is also variable and can appear as “block-like” structure A typical SNP is highly correlated with many neighboring SNPs in any given population Correlatio
Slide 14: Key Needs for Whole Genome Association Studies
Key Needs for Whole Genome Association Studies Access to well-phenotyped case and matched control populations Ability to assay tag SNPs; access to vast majority of genome to provide high genomic coverage and power Genotype 100,000’s of loci accurately providing complete and accurate data – High locu
Slide 15: Tag SNP Selection for Illumina HumanHap BeadChips
Tag SNP Selection for Illumina HumanHap BeadChips Maximize coverage of common variation by choosing tag SNPs – Tag SNP is a representative SNP – One serves as a proxy for another Tag SNP selection process: 1. Examine HapMap SNPs with MAF ≥ 0.05 in populations 2. Bin SNPs in high LD with one another
Slide 16: Genotyping Tools from Illumina
Genotyping Tools from Illumina # SNPs 650000 HumanHap650Y CUSTOM CONTENT 558K-650K HumanHap550+ 550,000 HumanHap550 325K–377K HumanHap300-Duo+ HumanHap300-Duo 317,000 240,000 HumanHap240S 100,000 Human-1 iSelect 7,600-60,800 GoldenGate 96 – 6,000 1 2 12 96 # SAMPLES 16
Slide 17: Content Strategy for Whole-Genome Genotyping BeadChips
Content Strategy for Whole-Genome Genotyping BeadChips Content HumanHap300 HumanHap550 HumanHap650Y Number of SNPs 317,503 555,352 655,352 HapMap data for tag Release 16c Release 20 Release 20 SNP selection >1 million common SNPs >2 million common SNPs >2 million common SNPs MAF threshold (≥) 0.05 0
Slide 18: HumanHap550 HapMap Coverage by Population
HumanHap550 HapMap Coverage by Population PAIRWISE TESTS 1 HumanHap550 CEU (mean 0.93 median 1.0) 0.9 COVERAGE OF HAPMAP RELEASE 21 HumanHap550 CHB+JPT 0.8 (mean 0.91 median 1) 0.7 HumanHap550 YRI (mean 0.74 median 0.87) 0.6 0.5 0.4 0.3 0.2 0.1 0 >0 >0.1 >0.2 >0.3 >0.4 >0.5 >0.6 >0.7 >0.8 >0.9 MAX r
Slide 19: HapMap Coverage by Population (HumanHap650Y)
HumanHap550 HapMap Coverage by Population PAIRWISE TESTS 1 HumanHap550 CEU (mean 0.93 median 1.0) 0.9 COVERAGE OF HAPMAP RELEASE 21 HumanHap550 CHB+JPT 0.8 (mean 0.91 median 1) 0.7 HumanHap550 YRI (mean 0.74 median 0.87) 0.6 HumanHap650Y YRI 0.5 (mean 0.81 median 0.93) 0.4 0.3 0.2 0.1 0 >0 >0.1 >0.2
Slide 20: HumanHap550 Coverage by Bin Size
HumanHap550 Coverage by Bin Size RELEASE 21 PAIRWISE TESTS, r2 = 0.7, MAF ≥ 0.05 CEU POPULATION 250,000 Non tagged Tagged (1 SNP) 200,000 Tagged (>1 SNP) NUMBER OF BINS 150,000 100,000 50,000 0 1 2 3 4 5 6 7 8 9 10+ BIN SIZE 20
Slide 21: Genome Coverage: HumanHap550 and Simulated Random 1.1M
Genome Coverage: HumanHap550 and Simulated Random 1.1M SNPs selected from 6 MAF ≥ 0.05, PAIRWISE TESTS representative 1 chromosomes: (26% of COVERAGE OF HAPMAP RELEASE 21 genome) 0.9 HumanHap550 Random 1.1M Loci selected based on 0.8 HumanHap650Y YRI even-spacing and minor allele frequency 0.7 (MAF)
Slide 22: Testing Coverage Using Power
Testing Coverage Using Power p1 p2 p3 p4 1 N Ptot = ∑ Pi p = p3 > p1 > p2 > p4 N i POWER TO DETECT ASSOCIATION IS A FUNCTION OF: Sample size (# cases and controls) – 1000 cases and 1000 controls Disease prevalence – 0.01 in general population Frequency of risk allele – Common variation (MAF > 0.05)
Slide 23: Total Power in Europeans
Total Power in Europeans HumanHap550 1.0 Maximum HumanHap550 TOTAL POWER 80% power 0.5 0.0 1.1 1.2 1.3 1.4 1.5 RELATIVE RISK 23
Slide 24: Total Power in Yoruba
Total Power in Yoruba HumanHap550 HumanHap650Y 1.0 Maximum HumanHap650Y TOTAL POWER HumanHap550 80% power 0.5 0.0 1.1 1.2 1.3 1.4 1.5 RELATIVE RISK 24
Slide 25: HumanHap300 vs Random 500k in Europeans
HumanHap300 vs Random 500k in Europeans 22 21 20 Pilmn − P500 k 19 ΔP = 18 P500 k 17 16 15 100% increase/decrease CHROMOSOME 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 50 100 150 200 250 25 POSITION (Mb)
Slide 26: HumanHap550 vs Random 500k in Europeans
HumanHap550 vs Random 500k in Europeans 22 21 20 Pilmn − P500 k 19 ΔP = 18 P500 k 17 16 15 100% increase/decrease CHROMOSOME 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 50 100 150 200 250 26 POSITION (Mb)
Slide 27: Illumina Whole-Genome Genotyping BeadChips vs. Random 500k
Illumina Whole-Genome Genotyping BeadChips vs. Random 500k CEU JPT+CHB YRI 3000 2000 HumanHap650Y 1000 0 3000 NUMBER 2000 HumanHap550 1000 0 3000 2000 HumanHap300 1000 0 -1 0 1 -1 0 1 -1 0 1 POWER DIFFERENCES 27
Slide 28: HumanHap550 Coverage of RefSeq Transcripts and Exons
HumanHap550 Coverage of RefSeq Transcripts and Exons 100% RefSeq Transcripts +/- 10 kb (mean 15, median 7) 90% RefSeq Exons +/- 10 kb (mean 3.8, median 3) 80% 70% PERCENT COVERED 60% 50% 40% 30% 20% 10% 0% 1+ 2+ 3+ 4+ 5+ 6+ 7+ 8+ 9+ NUMBER OF SNPs 28
Slide 29: HumanHap550 Spacing by Population
HumanHap550 Spacing by Population MAF ≥ 0.05 35 CEU, Average spacing on autosomes 5.5 kb 30 CEU, 90% percentile 12 kb CHB+JPT, Average Spacing 25 Kb BETWEEN SNPs on Autosomes 6.5 kb CHB+JPT, 90% percentile 14 kb 20 YRI, Average Spacing on Autosomes 6.3 kb 15 YRI, 90% percentile 14 kb 10 5 0 rX r1 r2
Slide 30: HumanHap550 Minor Allele Frequency Distribution
HumanHap550 Minor Allele Frequency Distribution 0.25 CEU, mean 0.23, median 0.23 CHB+JPT mean 0.21, median 0.20 0.20 YRI mean 0.22, median 0.21 PROPORTION OF SNPs 0.15 0.10 0.05 0.00 0-0.05 0.05-0.10 0.10-0.20 0.20-0.30 0.30-0.40 0.40-0.50 MINOR ALLELE FREQUENCY 30
Slide 31: SNPs in Known Copy Number Polymorphic Genomic Regions
SNPs in Known Copy Number Polymorphic Genomic Regions 4,284 SNPs from 495 regions included in HumanHap550 Regions were identified in 3 recent publications – Conrad, et al. (Nature Genetics 38:75-81, 2006) – Hinds, et al. (Nature Genetics 38:82-85, 2006) – McCarroll, et al. (Nature Genetics 38:86-92,
Slide 32: Analysis of a Copy Number Polymorphic Region
Analysis of a Copy Number Polymorphic Region 120 HapMap SAMPLES Published polymorphic region HumanHap550 32
Slide 33: Key Needs for Whole Genome Association Studies
Key Needs for Whole Genome Association Studies Access to well-phenotyped case and matched control populations Ability to assay tag SNPs; access to vast majority of genome to provide high genomic coverage and power Genotype 100,000’s of loci accurately providing complete and accurate data – High locu
Slide 34: Infinium II Assay
Infinium II Assay GENOMIC DNA 750ng WGA TT TC CC FRAGMENT DNA UNLABELED DNA HYBRIDIZATION ALLELE DETECTION THROUGH ddNTP SINGLE BASE EXTENSION 34
Slide 35: Allele Detection Through Single-Base Extension
Allele Detection Through Single-Base Extension 50mer OLIGO SEQUENCE BEAD Polymerase T 1. SELECTIVITY 2. SPECIFICITY A C DNA SAMPLE G T A T T 35
Slide 36: HumanHap550 Performance
HumanHap550 Performance 120 samples; 28 trios; 8 replicates PARAMETER SNP COUNTS PERCENT PRODUCT SPECIFICATION >99% (average) Call Rate 66,498,377 / 66,642,240 99.78% >98% (minimum) Reproducibility 4,430,704 / 4,431,059 99.99% >99.9% Mendelian 5,534 / 15,452,032 0.035% <0.1% Inconsistencies HapMap G
Slide 37: Unprecedented Call Rates of >99%
Unprecedented Call Rates of >99% FROM FUNG AND SINGLETON ET AL. NEUROLOGY THE LANCET A total of 219,577,497 unique genotype calls were made and the average call rate across all samples was 99.6%. 37
Slide 38: Inflammatory Bowel Disease Gene Finding Using HumanHap300
Inflammatory Bowel Disease Gene Finding Using HumanHap300 38
Slide 39: Concluding Remarks
Concluding Remarks HumanHap550 utilizes genome-wide tag SNPs derived from the International HapMap Project, on a single BeadChip slide Provides comprehensive genomic coverage and power for association studies in all populations Demonstrates exceptionally high data quality (measured by both accuracy