Statistical Strategies for Some Variable Selection Problems in Genetic Studies

• Activity Number: 505
• Type: Topic Contributed
• Date/Time: Wednesday, August 4, 2010 : 10:30 AM to 12:20 PM
• Sponsor: Section on Statistical Learning and Data Mining
• Author(s): Wei Sun*+
• Companies: The University of North Carolina at Chapel Hill
• Keywords: variable selection ; genetics

Abstract

With the advance of experimental technique, genetic variations are often measured by densely distributed genetic markers genome-wide. One of the fundamental tasks in genetic studies is to select multiple genetic markers that are related with the trait of interest. There are several challenges in this type of variable selection problem. First, it is often a high-dimension and low-sample-size problem. Secondly, there may be interaction effects between genetic markers. Thirdly, several traits are often measured to characterize one disease, and it is desirable to jointly study these traits. We will discuss some statistical strategies to tackle these problems.