A Powerful Multivariate Model for Multiple-Trait Genome-Wide Association Mapping

• Activity Number: 253
• Type: Contributed
• Date/Time: Monday, July 30, 2012 : 2:00 PM to 3:50 PM
• Sponsor: Biometrics Section
• Author(s): Hung-Chih Ku*+ and Lan Zhu
• Companies: Oklahoma State University and Oklahoma State University
• Keywords: quantitative trait loci ; penalized maximum likelihood ; genome-wide association ; multivariate analysis ; false discovery rate ; multiple-trait association mapping

Abstract

Identifying quantitative trait loci (QTL, genetic regions that are associated with phenotypic traits) by association mapping is critical for understanding the genetic architecture of complex traits or diseases. Current statistical methods mainly focus on association mapping for single trait from independent individuals. However, genetic inheritable complex diseases usually affect family members and are expressed by multiple correlated traits. Appropriate statistical models that can handle multiple traits simultaneously with genome-wide data sets are in high demand. In this study, we develop a novel multivariate model for detecting the locations and estimating the effects of QTL associated with multiple correlated traits from individuals with arbitrary pedigree structure. Results from simulation studies show that the power of our proposed method ranges from moderate to excellent under several common experimental scenarios explored. Moreover, the performance of our method is quite robust to the magnitude of correlation between traits as well as the complexity of pedigree structures.