Bayesian Hierarchical Variable Selection for Genome-Wide Association Studies

• Activity Number: 26
• Type: Topic Contributed
• Date/Time: Sunday, August 9, 2015 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistics in Imaging
• Author(s): Yize Zhao* and Hongtu Zhu and Zhaohua Lu and Rebecca C. Knickmeyer and Fei Zou
• Companies: SAMSI and The University of North Carolina at Chapel Hill and Penn State and The University of North Carolina and The University of North Carolina
• Keywords: Biological information ; Genome-wide association studies ; Grouping e ect ; Hierarchical variable selection ; Markov chain Monte Carlo

Abstract

It becomes increasingly important in the GWAS to select important genetic information in relation to a dichotomous variable or a quantitative trait. Currently, the discovery of biological association among SNPs motivates the strategy to construct the SNP-sets along the genome, which motivates the strategy to incorporate the set information into the selection. To this end, we proposed a unified Bayesian framework which allows the hierarchical variable selection while simultaneously encouraging grouping effect among SNPs. To accommodate the ultra high-dimensionality, we overcome the limitation of existing approaches and propose a novel sampling scheme. By constructing an auxiliary variable selection model under SNP-set level, we utilizes the posterior samples of the auxiliary model to guide the posterior inference for the SNP-level selection model. We apply the proposed method to a variety of simulation studies and show that our method is computational efficient and achieve substantially better performance than completing approaches. Applying the method to the ADNI data, we identify meaningful genetic information that are highly associated with different neuroimaging phenotypes.