Association mapping with high density marker panels

1
Association mapping with high density marker
panels

• Jeffrey Barrett

2
Outline

• Linkage disequilibrium and recombination
• HapMap
• Tag SNPs
• Basic association
• Practical

3
Linkage disequilibrium
4
Linkage disequilibrium
time
5
Indirect association
6
Measuring LD
locus 1
D ?11 - pq
p 1-p
q pq (1-p)q
1-q p(1-q) (1-p)(1-q)
p 1-p
q ?11 ?12
1-q ?21 ?22
locus 2
D D/DMAX
r2 D2/p(1-p)q(1-q)
7
Theoretical and empirical LD
Reich et al. Nature (2001)
8
LD analysis with Haploview
9
Genotypes vs haplotypes
Genotypes AA CT CC GA Haplotypes ACCG /
ATCA ACCA / ATCG ATCG / ACCA ATCA / ACCG
2n possible reconstructions n number of
heterozygous sites
10
Limited haplotype diversity
Daly et al, Nat Genet (2001)
11
Visualizing empirical LD
12
Haplotype blocks
13
Haplotype blocks
14
Haplotype blocks
15
Haplotype blocks
16
D and r2
17
D in 100kb
18
D in common SNPs, 100kb
19
r2 in 100kb
20
HapMap
21
HapMap samples
90 Yoruba individuals (30 parent-parent-offspring
trios) from Ibadan, Nigeria (YRI) 90 individuals
(30 trios) of European descent from Utah
(CEU) 45 Han Chinese individuals from Beijing
(CHB) 45 Japanese individuals from Tokyo (JPT)
22
Why multiple populations?
23
HapMap SNPs
PHASE I 1,000,000 successful SNPs across the
genome PHASE II 5,000,000 additional SNPs
attempted 4,000,000 total polymorphic SNPs
genomewide
Panel r2 gt 0.8 max r2 YRI
81 0.90 CEU 94 0.97 CHBJPT 94 0.97
24
Enabling association studiesdbSNP
International HapMap Project. Nature (2005).
25
Tagging
Reference panel HapMap data Tags SNPs chosen
for genotyping with the aim of capturing as much
information as possible Tests statistical tests
for association to disease
26
Pairwise tagging
Tags SNP 1 SNP 3 SNP 6 3 in total Test for
association SNP 1 SNP 3 SNP 6
Carlson et al. (2004) AJHG 74106
27
Testing tags for association
Genotype tags in cases and controls Each tag is
tested for association How can we better use
this information?
28
Use of haplotypes can improve genotyping
efficiency
Tags SNP 1 SNP 3 2 in total Test for
association SNP 1 captures 12 SNP 3 captures
35 AG haplotype captures SNP 46
Tags SNP 1 SNP 3 SNP 6 3 in total Test for
association SNP 1 SNP 3 SNP 6
de Bakker et al. (2005) Nat Genet 371217
29
Efficiency
de Bakker et al. (2005) Nat Genet 371217
30
Transferability among populations
CEU
CEU
CEU
Whites from Los Angeles, CA
Botnia, Finland
Utah residents with European ancestry(CEPH)
PIW de Bakker et al.
31
Genome-wide tagging coverage
Barrett and Cardon, Nat Genet (2006).
32
Population structure
Marchini, Nat Genet (2004)
33
Population structure - ?
BD 1.15
CAD 1.08
HT 1.09
CD 1.26
RA 1.06
T1D 1.07
T2D 1.10
Genomic control - ? genome-wide inflation of
median test statistic
34
Crohns collection center
Center 1 No. of samples 524
2 271
3 439
4 465
5 301
Center 3 ? 1.77 All others ? 1.09
35
IBS clustering
Compute IBS between all pairs of individuals, as
well as 270 HapMap samples Create a distance
matrix of (1-IBS) Classical multidimensional
scaling generates principal components which
capture largest fraction of variation
36
Crohns PCA
37
Genotype calling
38
Calling wrinkles gt 3 clusters
39
Plate effects
Transition to SSF site
40
Association allelic ?2
Case Control
A 70 90
T 30 10
Assumes multiplicative HW equilibrium
41
Haploview practical

• www.hapmap.org
• Find bounding hotspots for CARD15 (gt10 cM/Mb)
• Download file for this window

42
Haploview practical

• What fraction of the dataset can be captured with
  8 pairwise tags?
• How much more information can be gained by using
  multimarker tagging?

43
Haploview practical
Data in F\barrett Is our result experiment-wide
significant?