Increasing Generality and Power of Rare-Variant Tests by Utilizing Extended Pedigrees.

TitleIncreasing Generality and Power of Rare-Variant Tests by Utilizing Extended Pedigrees.
Publication TypeJournal Article
Year of Publication2016
AuthorsSul, JHoon, Cade, BE, Cho, MH, Qiao, D, Silverman, EK, Redline, S, Sunyaev, S
JournalAm J Hum Genet
Date Published2016 Oct 06
KeywordsDatasets as Topic, Ethnic Groups, Family, Female, Genetic Predisposition to Disease, Genetic Variation, Genotype, Humans, Male, Models, Genetic, Pedigree, Phenotype, Research Design

Recently, multiple studies have performed whole-exome or whole-genome sequencing to identify groups of rare variants associated with complex traits and diseases. They have primarily utilized case-control study designs that often require thousands of individuals to reach acceptable statistical power. Family-based studies can be more powerful because a rare variant can be enriched in an extended pedigree and segregate with the phenotype. Although many methods have been proposed for using family data to discover rare variants involved in a disease, a majority of them focus on a specific pedigree structure and are designed to analyze either binary or continuously measured outcomes. In this article, we propose RareIBD, a general and powerful approach to identifying rare variants involved in disease susceptibility. Our method can be applied to large extended families of arbitrary structure, including pedigrees with only affected individuals. The method accommodates both binary and quantitative traits. A series of simulation experiments suggest that RareIBD is a powerful test that outperforms existing approaches. In addition, our method accounts for individuals in top generations, which are not usually genotyped in extended families. In contrast to available statistical tests, RareIBD generates accurate p values even when genetic data from these individuals are missing. We applied RareIBD, as well as other methods, to two extended family datasets generated by different genotyping technologies and representing different ethnicities. The analysis of real data confirmed that RareIBD is the only method that properly controls type I error.

Alternate JournalAm. J. Hum. Genet.
PubMed ID27666371
PubMed Central IDPMC5065687
Grant ListR01 GM078598 / GM / NIGMS NIH HHS / United States
R01 HL113264 / HL / NHLBI NIH HHS / United States
R01 HL046380 / HL / NHLBI NIH HHS / United States
R01 HL113338 / HL / NHLBI NIH HHS / United States
R35 HL135818 / HL / NHLBI NIH HHS / United States
P01 HL105339 / HL / NHLBI NIH HHS / United States
K01 HL129039 / HL / NHLBI NIH HHS / United States
UM1 HG006493 / HG / NHGRI NIH HHS / United States
R01 MH101244 / MH / NIMH NIH HHS / United States