There are many known examples of the allelic heterogeneity and imperfect tagging phenomena. The former one is of great importance in monogenic traits but has not yet been systematically investigated and quantified in complex-trait genome-wide association studies (GWASs). We devised a multi-SNP association method that estimates the effect of loci harboring multiple association signals by using GWAS summary statistics.
Applying the method to a large anthropometric GWAS meta-analysis (from the Genetic Investigation of Anthropometric Traits consortium study), we show that for height, body mass index (BMI), and waist-to-hip ratio (WHR) substantial additional phenotypic variance can be explained compared to what obtained via single SNP association. The method also permitted an increase (by up to 50%) in the number of loci that replicate in a discovery-validation design. Specifically, we identified 74 loci at which the multi-SNP, a linear combination of SNPs, explains significantly more variance than does the best individual SNP. A detailed analysis of multi-SNPs shows that most of the additional variability explained is derived from SNPs that are not in linkage disequilibrium with the lead SNP, suggesting a major contribution of allelic heterogeneity to the missing heritability.
Our results were published in the 2012 November issue of the American Journal of Human Genetics. A Matlab package that implements the method is also available here.
Lay summary (in French) is available both at the University of Lausanne news site and that of the University Hospital (CHUV).