Nonlinear Sufficient Dimension Reduction for Association Testing of Complex Traits

• Activity Number: 241
• Type: Contributed
• Date/Time: Monday, July 30, 2012 : 2:00 PM to 3:50 PM
• Sponsor: Biometrics Section
• Author(s): Hongjie Zhu*+ and Lexin Li and Hua Zhou
• Companies: Duke University and North Carolina State University and North Carolina State University
• Keywords: Association Study ; Kernel Methods ; Dimention Reduction ; Rare Variants ; Interaction

Abstract

Association tests based on next-generation sequencing data are often under-powered due to presence of rare variants and large amount of neutral or protective variants. A successful strategy is to aggregate genetic information within meaningful SNP-sets, e.g., genes or pathways, and test association on SNP-sets. Many existing methods for group-wise tests require specific assumptions about the direction of individual SNP effects and/or perform poorly in the presence of interactions. To meet these challenges, we propose a joint association test strategy based on kernel sufficient dimension reduction methods. Accompanying this strategy, we also propose a class of new kernels specially designed for genotype data.The strategy coupled with the new kernels shows superior performance in identifying causal genes over existing methods across various disease models simulated from sequence data of real genes. The class of new kernels can potentially boost the power of various kernel-based methods that analyze genotype data.