Statistical Challenges with Genetic Next-Generation Sequence Data

• Activity Number: 59
• Type: Topic Contributed
• Date/Time: Sunday, July 31, 2011 : 4:00 PM to 5:50 PM
• Sponsor: Section on Statistical Computing
• Author(s): Paul W. Livermore Auer*+
• Companies: Fred Hutchinson Cancer Research Center
• Keywords: rare variants ; single nucleotide polymorphisms ; next generation sequencing

Abstract

Next generation sequencing technologies are being used in a variety of ways to generate data for genetic association studies. In this talk I will focus on one particular application, namely exome-sequencing. Population based exome-sequencing studies typically focus on finding rare genetic variants associated with a complex trait of interest. To enhance power, these studies often employ an outcome dependent sampling (ODS) mechanism. Although there are many recently proposed statistical methods for testing rare-variant associations with complex traits, no current approach explicitly accounts for the ODS mechanism. I will discuss the intuition behind some of the more popular rare-variant tests along with ways to incorporate ODS. Several different approaches will be compared through simulations and applications on real sequencing data.