Comparison of Different Haplotype-Based Haplotype-Based Association Methods for Gene-Environment (G×E) Interactions in Case-Control Studies when Haplotype-Phase Is Ambiguous

Objective: We compared four haplotype-based approaches for the analysis of gene-environment interactions when haplotype-phase is ambiguous. The methods employ different versions of the expectation maximization algorithm and differ in the choice of the reference group and in the way the risk of disease is modeled (retrospective versus prospective). Furthermore, the methods are based on distinct assumptions (such as Hardy Weinberg equilibrium). The haplotypebased methods were also compared to single-marker logistic regression (LR). Methods: We simulated case-control scenarios where the risk variant was directly genotyped (direct scenario) or in linkage disequilibrium with the genotyped markers (indirect scenario). Results: The retrospective methods tended to be more powerful for detecting interactions than the prospective methods. In the indirect scenarios, the power of all methods was decreased. However, the power of the retrospectives methods was high in some scenarios and the interactions may only be detectable when Received: October 16, 2008 Accepted after revision: April 3, 2009 Published online: July 22, 2009 Jenny Chang-Claude German Cancer Research Center, DKFZ Division of Cancer Epidemiology, Unit of Genetic Epidemiology Im Neuenheimer Feld 280, DE–69120 Heidelberg (Germany) Tel. +49 6221 422 373, Fax +49 6221 422 203, E-Mail [email protected] © 2009 S. Karger AG, Basel 0001–5652/09/0684–0252$26.00/0 Accessible online at: www.karger.com/hhe D ow nl oa de d by : 54 .1 91 .4 0. 80 9 /1 6/ 20 17 1 1: 09 :5 4 A M