Complex traits and the HapMap Project

1
Complex traits and the HapMap Project
2
Identification of genes
Map
Clone
Identify the Phenotype
3
Trait / Phenotype

• Mendelian (single gene traits)
• establish diagnostic criteria
• determine mode of inheritance, gene frequency and
  penetrance.
• linkage mapping

4
Linkage
1
2
I
1
2
4
5
II
1
3
5
7
2
4
6
8
III
5
Mapping and sequencing
10000 Kb
Markers
100 Kb
DNA clones
6
Trait / Phenotype

• Mendelian (single gene traits)
• establish diagnostic criteria
• determine mode of inheritance, gene frequency and
  penetrance.
• linkage mapping
• Non-Mendelian (complex traits)
• difficult to layout diagnostic criteria (e.g.
  psychiatric disorders).
• difficult to determine mode of inheritance, gene
  penetrance and frequency.
• So, what do we do?

7
Association Studies

• Population based studies that test if markers are
  enriched in one population compared to a second
  population

8
SNPs (snips)

• A SNP is a site in the DNA where different
  chromosomes differ in the base they have.

9
SNPs
Paternal allele CCCGCCTTCTTGGCTTTACA Maternal
allele CCCGCCTTCTCGGCTTTACA
Paternal allele CCCGCCTTCTTGGCTTTACA Maternal
allele CCCGCCTTCTTGGCTTTACA
10
Case-Control Association Studies

• Genetic factors are compared within large case
  (affected) and control populations to identify
  correlations with a defined phenotype

11
Advantages Over Linkage Analysis

• No need for multi-generation family pedigrees
  Individuals are unrelated
• More powerful at detecting small and moderate
  genetic effects i.e. The phenotype can be mild or
  demonstrate incomplete penetrance

12
Case-Control Association Studies

• Family studies population studies

C/C
C/T
C/C
C/C
C/C
C/C
C/T
40 T 60 C
15 T 85 C
Case
Control
C/C
C/T
C/T
Polygenic Complex Inheritance majority of common
disease
Single gene disorders Mendelian Inheritance lt10
common disease
13
Case-Control Association Studies

• Single nucleotide polymorphisms (SNPs) are the
  markers of choice
• Approx 1 SNP every kb
• Low mutation rate
• Easily genotyped (i.e. High-throughput)

14
Identification of genes underlying human
Mendelian traits and genetically complex traits
in humans and other species
Science, Volume 298, Number 5602, Issue of 20 Dec
2002, pp. 2345-2349
15
The HapMap project

• 100 million - three years - to catalogue common
  haplotype blocks.
• Francis Collins The Hap/Map will provide the
  missing link between the DNA sequence of the
  genome and the way in which the genome influences
  the risk of disease.

16
How many SNPs?

• gt9.8 million SNPs have been identified (17 August
  2004).
• 93 of genes contain a SNP.
• 98 of genes are within 5000 bp of a SNP.
• 10 million common SNPS in human genome

17

• HapMap project allows a more systematic approach
  for searching for genetic factors associated with
  common diseases.
• it allows rapid scanning of the genome for
  disease gene by using a number of SNPS and
  characterising patterns of linkage disequilibrium
  (LD).

18
Linkage Disequilibrium (LD)

• Non-random association between alleles at
  different loci

Locus 1
Locus 2
Locus 1
Locus 2
A
C
C 50
A 50
Maternal
G
T
Paternal
T 50
G 50
Haplotype
Hardy-Weinberg
LD
C
A
0.5 X 0.5 0.25
0.40
C
G
0.5 X 0.5 0.25
0.10
T
A
0.5 X 0.5 0.25
0.10
T
G
0.5 X 0.5 0.25
0.40
19
Haplotype blocks

• Genome is divided into blocks?!
• High LD
• Low diversity
• separated by recombination hotspots

20
Haplotype blocks
21
Gabriel et al., The structure of haplotype blocks
in the human genome. Science 296 (2002), pp.
22252229.

• 51 autosomal regions
• average size 250,000 bp
• 13 megabases of the human genome (0.4 of
  human genome)
• Samples Africa, Europe, and Asia
• haplotype blocks can be reliably identified by
  genotyping a sample of common markers within
  their span
• 1/2 of the human genome exists in blocks of
• 22 kb or larger in African and African-Americans
• 44 kb or larger in European and Asian samples
• Within each block, only three to five haplotypes
  capture 90 of all chromosomes in each
  population.
• Both the boundaries of blocks and the specific
  haplotypes observed are shared to a remarkable
  extent across populations.
• 300,000 to 1,000,000 SNPs are adequate in an
  association study to survey for common haplotypes
  
• but see Goldstien et al. 150,000 - 300,000

22
Issues

• Block definition
• hotspots defined directly for only one region
  (HLA)
• over representation of common alleles
• could miss rare variants involved in disease,
• but what if Common disease - Common variant
  hypothesis is true?
• What is the optimal map density (higher SNP
  density --gt shorter blocks)

23
SNP density and haplotype blocks
24
Do we need to genotype all markers?

• redundant SNPs
• haplotype tagging SNPs

25
(No Transcript)
26
(No Transcript)
27
(No Transcript)
28
Genome-wide or candidate gene scans?

• Genome-wide
• costly
• 3 orders of magnitude more SNPs needed
• chance of finding a causal SNP is 10 fold less
• So why not start with candidate genes and regions?

29
Cloning DPP10 gene involved in asthma

• Linkage
• interval between 2q14 and 2q32
• mouse bronchial hyper-responsiveness linked to
  the region homologous to 2q14
• M. Allen et al. (2003), Positional cloning of a
  novel gene influencing asthma from chromosome
  2q14, Nature Genet., 35258-263.

30

• Samples
• 244 families including 239 children with asthma
• used the total serum IgE concentration (LnIgE) as
  a quantitative measure of atopy.

31

• Found association between asthma and the
  microsatellite D2S308 (near IL1 locus).
• Allele 3 of D2S308 (D2S3083) positively
  associated with asthma.

32
TDT transmission disequilibrium test
33

• Constructed BAC/PAC contig surrounding D2S308
  marker.
• Used SNPs to measure LD

LD tendency of a set of alleles at one locus to
segregate with a particular group of alleles on a
second closely linked locus
34

• Found strongest association with WTC122P SNP, 1
  kb proximal to D2S308
• genotyped WTC91P, WTC122P and D2S308 in 1,047
  schoolchildren aged 911 years

35

• measured the frequency of the WTC122P1 D2S3083
  haplotype in
• 178 severe steroid-dependent asthmatics
• 92 mild asthmatics
• 304 unrelated controls
• WTC122P1 D2S3083 haplotype was significantly
  more frequent among the severe asthmatics but not
  among the mild asthmatics

36

• Identified a novel gene called DPP10
• no coding polymorphisms in DPP10
• strongest association with WTC122P
• WTC122P alleles showed differential protein
  binding with nuclear extracts from the T-cell
  line C8166

37
Exon structure of DPP10

• Effects on asthma susceptibility may result in
  part from the presence or absence of the CdxA
  promoter element before exon Ib, leading to
  alternative splicing between membrane-bound and
  other forms of the protein.

38

• A haplotype-based molecular analysis of CFTR
  mutations associated with respiratory and
  pancreatic diseases
• JH Lee et al. (2003), Human Molecular Genetics,
  2003, Vol. 122321-2332.

39

• Mutations in cystic fibrosis transmembrane
  conductance regulator (CFTR)
• cystic fibrosis (F508del, G532X, N130K)
• severe and high penetrance
• pancreatic and other respiratory disorders
• congenital bilateral aplasia of the vas deferens
  (CBAVD), which leads to male infertility, may
  occur in isolation or as a manifestation of
  cystic fibrosis

40

• Samples
• South Asian
• healthy controls
• bronchiectasis patients
• chronic pancreatitis patients

41
Identified mutations
42
(No Transcript)
43
97 of controls V470-2694T or M470-2694G
44
Of the two IVS8 T5 containing haplotypes only
haplotype 5, which also has the low activity
variation of V470, showed an association with the
disease phenotype
45

• Haplotype 12
• frequency in normal Belgians 7.3
• gives rise to gt95 of the three most common CF
  cases, F508del, 542X, N1303
• but in Koreans frequency is 0.9 (CF is rare in
  South Asia)!