Combining Rare Variants From Families and Unrelateds

• Activity Number: 664
• Type: Invited
• Date/Time: Thursday, August 8, 2013 : 10:30 AM to 12:20 PM
• Sponsor: WNAR
• Author(s): David Fardo*+ and Iuliana Ionita-Laza
• Companies: University of Kentucky and Columbia University
• Keywords: genetic association studies ; rare variants ; GWAS ; next-generation sequencing ; family-based genetic association

Abstract

Genetic association studies can be broken into two broad categories: population studies (most often of the case-control variety) that collect unrelated individuals, and family studies that recruit related pedigrees. It is not uncommon for studies of both categories to be available for a particular disease. In this situation, it is more powerful to combine the evidence for association rather than to conduct the analyses separately. Various study design aggregation approaches have been developed, most commonly in the context of common variants. Recently developed next generation sequencing technologies have made it feasible to assess association between multiple rare variants and disease. This, in combination with motivation to target the so-called 'missing heritability' from genome-wide association studies, has made the study of rare variation particularly attractive. Methods to handle the problems due to the sparsity of rare variants generally focus on rules to collapse variation across some genetic unit, e.g., a gene. We present here a unified method to aggregate population and family data using GWAS or sequencing data and assess its performance using simulation.