Genetic Model Selection in Two-Stage Analysis for Case-Control Association Studies

• Activity Number: 327
• Type: Topic Contributed
• Date/Time: Tuesday, July 31, 2007 : 2:00 PM to 3:50 PM
• Sponsor: Section on Nonparametric Statistics
• Author(s): Hon Keung (Tony) Ng*+ and Gang Zheng
• Companies: Southern Methodist University and National Heart, Lung, and Blood Institute
• Keywords: Efficiency robustness ; Genetic models ; Genome-wide association ; Hardy-Weinberg disequilibrium ; Two-stage analysis

Abstract

The genotype-based Cochran-Armitage trend test (CATT) has been suggested for testing association between a marker and a disease in case-control studies. When the underlying genetic model is known, the CATT with the recessive, additive, and dominant scores is used. However, the genetic model is often unknown. Therefore, robust tests, such as the CATT with the additive score, the maximum efficiency robust test (MERT) and the maximum tests (MAX) are used in practice. We propose a two-stage analysis: In the first stage, we use the difference of Hardy-Weinberg disequilibrium coefficients between the cases and controls for model selection; in the second stage, an optimal CATT, corresponding to the selected model, is used for testing association. Simulation studies show that this approach is more preferable to the CATT with the additive score and the MERT, and it is more powerful than the MAX.