Detecting Gene-Gene Interactions with Survival Times Based on Generalized Multifactor Dimensionality Reduction Method

• Activity Number: 125
• Type: Contributed
• Date/Time: Monday, August 2, 2010 : 8:30 AM to 10:20 AM
• Sponsor: Biometrics Section
• Author(s): Seungyeoun Lee*+ and Min-Seok Kwon and Jaebong Lee and Taesung Park
• Companies: Sejong University and Seoul National University and Sejong University and Seoul National University
• Keywords: gene-gene interactions ; multifactor dimensionality reduction method ; survival time ; Cox model ; score residuals

Abstract

Most statistical methods for genome-wide association studies (GWAS) have been based on identifying the genetic variants by locus-by-locus searching over all genotyped markers. However, there have been increasing empirical evidences for interaction effects between loci for complex disease in GWAS. One of approaches for identifying gene-gene interactions is the multifactor dimensionality reduction (MDR) method proposed by Ritchie et al. (2001). Recently Lou et al.(2007) proposed the generalized MDR (GMDR)), called the score-based MDR, in which the two distinct groups are defined by the score statistic. In this paper, we apply the GMDR to the survival time data in which the gene-gene interactions are considered adjusting the covariate effects such as age, sex and clinical variables. We analyze a real dataset of cancer patients and perform the simulation studies.