Design of the first generation of Genome Wide Association Studies
Genome-wide association studies (GWAS) involving hundreds of thousands of SNPs in thousands of cases and controls are now possible and ongoing. The first of many analytical challenges in these studies involves the choice of SNPs to genotype. Despite active research in methods for haplotype `tagging' and intense debate about the optimal type of SNPs to genotype, it is not realistic or practical to construct a different panel of tag-SNPs to suit the interests and needs of each investigator. The Wellcome Trust Case Control Consortium (WTCCC) is one of the largest studies in the first generation of GWAS, the initial stage of which involves the genotyping of 675,000 SNPs in 3000 controls and 1000 cases from each of 8 common diseases. We present analyses of the many factors which contributed to the selection of these SNPs, which are a combined approach using the Affymetrix 500K chip along with 175K carefully chosen fill-in SNPs. The final set of SNPs covers > 90% of all common variants, includes all known, polymorphic nonsynonymous coding SNPs, captures all known variation in the extended MHC and contains built-in redundancy for robustness in large LD bins. This panel is planned for use in a number of additional pending studies using this technology platform. We therefore discuss the overall performance as well as implications for study analysis.