Integrative Genomic Association Testing via Kernel Machine Mediation Analysis

• Activity Number: 69
• Type: Contributed
• Date/Time: Sunday, July 31, 2016 : 4:00 PM to 5:50 PM
• Sponsor: Biometrics Section
• Author(s): Angela Hsiaohan Chen* and Sihai Zhao
• Companies: University of Illinois at Urbana-Champaign and University of Illinois at Urbana-Champaign
• Keywords: Kernel machine regression ; Integrative genomics ; Kernel PCA ; Mediation analysis ; Genome-wide association studies ; Genetic association testing

Abstract

An integrative approach to association testing, which combines outcome and genotype data with other types of genomic information, has shown to be a more powerful approach to detect SNPs than the standard approach. Previously, Zhao et al. (2014) proposed a regression model for integrating genotype data, gene expression, and outcome, but their method required strong modeling assumptions on the relationship between expression and phenotype. We propose a method that can relax these assumptions by using a kernel machine (KM) regression framework that can allow for complex relationships, such as non-linear or interactive effects. Simulations and methodological comparisons demonstrate the benefits of our approach.