B1312
Title: Iterative Bayesian analysis of GLMMs for non-Gaussian GWAS data
Authors: Shuangshuang Xu - Virginia Tech (United States)
Jacob Williams - Virginia Tech (United States)
Marco Ferreira - Virginia Tech (United States) [presenting]
Abstract: A novel iterative Bayesian model selection method is proposed for generalized linear mixed models (GLMMs) specifically designed to analyze non-Gaussian Genome-wide association studies (GWAS) data. GWAS main goal is to identify single nucleotide polymorphisms (SNPs) associated with phenotypes of interest. Usually, GWAS data are analyzed with single marker analysis (SMA) methods. However, SMA methods usually suffer from a high false discovery rate (FDR). A novel iterative Bayesian method is proposed to find SNPs associated with non-Gaussian phenotypes based on generalized linear mixed models (GLMMs). Thus, the method is called iterative Bayesian GLMMs for GWAS (IBG2). IBG2 iterates two steps: a screening step that screens for candidate SNPs and a model selection step that considers all screened candidate SNPs as possible regressors. A simulation study shows that IBG2 has a favorable performance compared to GLMM-based SMA. Applications to human health and plant science illustrate the usefulness and flexibility of IBG2.
