Improve the Accuracy of Calling Short CNVs in Genomewide Association Studies

• Activity Number: 144
• Type: Invited
• Date/Time: Monday, August 2, 2010 : 10:30 AM to 12:20 PM
• Sponsor: International Indian Statistical Association
• Author(s): Jianxin Shi*+
• Companies: National Cancer Institute
• Keywords: genome-wide association studies ; copy number variants ; linkage disequilibrium

Abstract

Germline copy number variants (CNVs) were reported to be associated with many complex diseases including cancers. However, the accuracy of calling short CNVs based on GWAS genotyping platforms is not satisfying, particularly when the scientific goal is to identify rare and highly penetrant CNVs associated with complex diseases. In this talk, we will discuss two strategies that may potentially improve the accuracy of calling short CNVs. The first approach aims to account for the linkage disequilibrium when modeling B-allele frequencies. This strategy is particularly useful for calling short deletions. The second approach aims to model the relatedness among subjects and it can be applied in both family-based genetic studies or population-based genetic studies. The application to GWAS of cancers will be be presented.