学术文献库

In genetic association studies, rare variants with extremely small allele frequency play a crucial role in complex traits, and the set-based testing methods that jointly assess the effects of groups of single nucleotide polymorphisms (SNPs) were developed to improve powers for the association tests. However, the powers of these tests are still severely limited due to the extremely small allele frequency, and precise estimations for the effect sizes of individual SNPs are substantially impossible. In this article, we provide an efficient set-based inference framework that addresses the two important issues simultaneously based on a Bayesian semiparametric multilevel mixture model. We propose to use the multilevel hierarchical model that incorporate the variations in set-specific effects and variant-specific effects, and to apply the optimal discovery procedure (ODP) that achieves the largest overall power in multiple significance testing. In addition, we provide Bayesian optimal "set-based" estimator of the empirical distribution of effect sizes. Efficiency of the proposed methods is demonstrated through application to a genome-wide association study of coronary artery disease (CAD), and through simulation studies. These results suggested there could be a lot of rare variants with large effect sizes for CAD, and the number of significant sets detected by the ODP was much greater than those by existing methods.