A single-nucleotide polymorphism tagging set for human drug metabolism and transport

Kourosh R Ahmadi, Mike E Weale, Zhengyu Y Xue, Nicole Soranzo, David P Yarnall, James D Briley, Yuka Maruyama, Mikiro Kobayashi,

Nicholas W Wood, Nigel K Spurr, Daniel K Burns, Allen D Roses, Ann M Saunders & David B Goldstein

Nature Genetics  37, 84 - 89 ( January 2005)

Presented by Navdeep

A single-nucleotide polymorphism tagging set for human drug metabolism and transport

BackgroundAimsMethodsResultsConclusionsReferences

Outline

Tagging SNPs

a, SNPs. four versions of the same chromosome region in different people showing three bases where variation occurs. Each SNP has two possible alleles; the first SNP in panel a has the alleles C and T.

b, Haplotypes. A haplotype is made up of a particular combination of alleles at nearby SNPs. a. For this region, most of the chromosomes in a population survey turn out to have haplotypes 1–4.

c, Tag SNPs. Genotyping just the three tag SNPs out of the 20 SNPs is sufficient to identify these four haplotypes uniquely.

A SNP or a set of SNPs that have been selected on the basis of linkage

disequilibrium (LD) patterns to represent other SNPs

Tagging SNPs

Advantages of tagging SNPs– Fewer SNPs can be used to construct genome wide linkage

disequilibrium map.

Potential problems– How well do the selected tags represent undetected variation in

the original sample?– How well will the tags represent variation (both detected and

undetected) in a new sample from the same population? – How well do they represent different populations having different

LD patterns?

The ability of tSNP sets to tag dropped SNPs

Aims

• Identify and evaluate tSNPs for genes involved in the absorption, distribution, metabolism and excretion of drugs (ADME genes)

Materials and Methods• Selection of tSNPs

– haplotype r2 criterion

• Evaluation of tSNPs – dropped SNP−plus−resampling approach

• Comparison of SNPs with different MAF – two-tailed Wilcoxon paired-sample rank test

• Genes, SNP selection and choice of populations – a target density of 1 SNP of MAF > 10% per 2 kb of genomic DNA

• Gene clusters – no two genes from a cluster to be separated by more than 50 kb

Haplotype r2

• Haplotype r2 is the coefficient of determination (ie. The proportion of explained variation) obtained from a standard linear regression of the allelic state (coded 0/1) of a SNP in question against the haplotypes determined by the tSNP set. This regression is equivalent to a one-way analysis of variance with each tSNP-defined haplotype as a separate group.

• It allows assessment of the loss of power resulting from typing a tSNP as opposed to the causal variant with which it is associated

Long range Linkage Disequilibrium

Minor allelic frequency (MEF)

Indicates the number of occurrences of an allele seen in the total number of

chromosomes typed at the SNP site

Performance of tags selected from the full data set

Performance of tags selected from the reduced* data set

*SNPs with MAFs < 5% excluded

The effect of initial genotyping density on tag performance

Cosmopolitan tSNP set suitable for both European and Japanese populations

Performance of selected tSNPs in representing candidate functional variation

Performance of selected tSNPs in different population sample

Conclusions

• The effect of MAF on tSNP performance is heavily dependent on the size of the LD sample

• Comprehensive tagging will require a high genotyping density (one SNP of MAF ≥5% per 2.5 kb )

• Performance of population specific tSNPs sets in predicting functional variants is similar to that of random SNPs.

• Haplotype r2 based tSNP selection is highly effective even when applied to a population different from LD sample.

• Rare variants are not well tagged.

References• Goldstein, D.B., Ahmadi, K.R., Weale, M.E. & Wood, N.W. Genome scans and candidate gene

approaches in the study of common diseases and variable drug responses. Trends Genet. 19, 615−622 (2003)

• Goldstein, D.B., Tate, S.K. & Sisodiya, S.M. Pharmacogenetics goes genomic. Nat. Rev. Genet. 4, 937−947 (2003).

• Pritchard, J.P. & Przeworski, M. Linkage disequilibrium in humans: models and data. Am. J. Hum. Genet. 69, 1−14 (2001)

• Carlson, C.S. et al. Selecting a maximally informative set of single-nucleotide polymorphisms for association analyses using linkage disequilibrium. Am. J. Hum. Genet. 74, 106−120 (2004).