Identifying Carriers of CNVs from the Genotype Intensities of Related Individuals

• Activity Number: 277
• Type: Topic Contributed
• Date/Time: Tuesday, July 31, 2012 : 8:30 AM to 10:20 AM
• Sponsor: Biometrics Section
• Author(s): William C.L. Stewart*+
• Companies: Columbia University
• Keywords: Copy Number Variantion ; Deletion ; Intensity ; Elston-Stewart ; Finite Mixture Model ; Hemizygous

Abstract

Complex human traits are partially determined by genetic factors that alter the length of DNA (e.g. copy number variants (CNVs)). Typically, carriers of large (>10 kb) CNVs (e.g. deletions) can be identified easily from the genotype intensities of single nucleotide polymorphisms (SNPs) in the region. However, carrier status is difficult to impute for small deletions. We have developed a novel approach to impute carriers of any deletion (small or large, rare or common) from the genotype intensities of related individuals. We use a finite mixture model to analyze the genotype intensities of each SNP, along with a modified version of the Elston-Stewart algorithm to incorporate information from flanking SNPs. When applied to a study of 737 bipolar families, where carrier status was determined experimentally, the results show that our method imputes carriers more accurately than existing methods, and that the difference is largest when the analysis is restricted to a single SNP. Thus, our approach has the potential to (1) improve the resolution of deletion boundaries; (2) increase the power of association studies; and, (3) provide valuable insight into the mechanistic causes of disease.