Comprehending Gene-Based Association Signals: A Penalized Regression Approach for Haplotype-Based Analysis with Application in Pharmacogenetic Studies and Individualized Medicine

• Activity Number: 337
• Type: Topic Contributed
• Date/Time: Tuesday, August 2, 2011 : 10:30 AM to 12:20 PM
• Sponsor: Section on Statistics in Epidemiology
• Author(s): Megan Lee Koehler*+ and Jung-Ying Tzeng and Howard D. Bondell
• Companies: North Carolina State University and North Carolina State University and North Carolina State University
• Keywords: haplotype ; penalized regression ; pharmacogenetics ; individualized medicine

Abstract

Gene-based/marker-set association analysis has become a promising tool for detecting association signals. It has proven to be a powerful approach for detecting global associations between a set of markers and a phenotype. However, not many tools are available for comprehending the overall signals identified. In this work, we propose a penalized-likelihood method that is able to identify the sources of the overall association signals by studying multi-marker genetic and gene-environment interaction effects. The proposed method also has significant utility in pharmacogenetic research, for which the focus is to evaluate genetic-drug interactions and to obtain information useful in the creation of individualized treatment regimes. Simulation studies show that the proposed penalized method has comparable or more power than the standard approach and maintains control on Type I error rates. In addition, a pharmacogenetic application is provided and highlights the utility of the proposed method.