|
Activity Number:
|
210
|
|
Type:
|
Invited
|
|
Date/Time:
|
Tuesday, August 5, 2008 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
International Indian Statistical Association
|
| Abstract - #300402 |
|
Title:
|
Efficient Statistical Analysis of Case-Control, Genome-Wide Association Study Data
|
|
Author(s):
|
Michael Boehnke*+
|
|
Companies:
|
The University of Michigan
|
|
Address:
|
Department of Biostatistics, Ann Arbor, MI, 48109,
|
|
Keywords:
|
genetics ; association ; stratification ; winner's curse ; type 2 diabetes
|
|
Abstract:
|
With the availability of a catalog of common human variants and the rapid drop in genotyping costs, genome-wide association studies now provide a powerful means to localize genetic variants that predispose to human diseases. Many such studies are now in progress. In this talk, I will focus on solutions to two statistical problems posed by genome-wide association studies: avoiding spurious results owing to population stratification by matching based on genome-wide marker data and correcting for overestimation of genetic effect size owing to the "winner's curse" by conditioning on the observation of a significant result. I will illustrate the value of these methods with data from the Finland-United States Investigation of NIDDM Genetics (FUSION) study.
|
