Two-stage Designs for Gene-disease Association Studies

• Activity Number: 326
• Type: Topic Contributed
• Date/Time: Wednesday, August 11, 2004 : 10:30 AM to 12:20 PM
• Sponsor: ENAR
• Author(s): Varghese George*+
• Companies: University of Alabama, Birmingham
• Keywords: association ; candidate genes ; complex diseases ; disequilibrium-mapping ; linkage ; two-stage design

Abstract

Linkage and disequilibrium mapping methods to identify causative loci are important tools for understanding the genetic etiology of complex diseases and traits. Refinements of these methods will enhance ongoing efforts to effectively dissect the architecture of complex human diseases. We will discuss a two-stage TDT design in which association and linkage disequilibrium are tested sequentially. A major advantage of this design is that if a significant association observed during the first stage is spurious due to population admixture/stratification, it is likely to disappear in the second stage (unless due to Type I errors). If no significant association is found in the first stage, there is no need to proceed with the second-stage analysis for fine mapping in that region. On the other hand, if we find significance at both stages, that may be concluded as confirmatory finding of a candidate gene. The two-stage approach should be more powerful because fewer tests are performed overall and hence the multiple-testing issue is mitigated. We will compare population-based case-control design, classical TDT, and the two-stage approach in terms of power and efficiency.