Department of Statistics Seminar
North Carolina State University

presents

Dr. Mary Sara McPeek

University of Chicago

"Quasi-likelihood methods for case-control association testing with related individuals"

ABSTRACT

Human diseases such as asthma, diabetes, and hypertension are called "complex" diseases because they are influenced by many factors, both environmental and genetic. A fundamental problem of interest to human geneticists is to understand what the genetic risk factors are that predispose some people to get a particular complex disease. Association testing refers to testing the null hypothesis of independence of genotype and trait value, e.g., for a genetic locus or chromosomal region. We consider a study design for association testing in which cases are sampled from families with multiple affected members. Controls may be related to each other and to affecteds or unrelated. This type of design arises naturally when association testing follows linkage analysis, for which multiplex families may have been collected. The relatedness of the the individuals in the study has an impact in two ways: 1) the dependence among individuals due to their relationship can be used to weight the individuals to improve power (and in any case, must be taken into account to control type I error) and 2) a trait-predisposing allele or genotype may be enriched in individuals with multiple affected relatives, and one can use this information to improve power. We propose new, computationally efficient methods for case-control association studies of binary traits suitable for any set of related individuals, provided that their genealogy is known. We consider applications to genetic studies in which multiple small- to moderate-size pedigrees are sampled as well as a study involving a large, complex, inbred pedigree.

This is joint work with Timothy Thornton, Xiaodong Wu, Catherine Bourgain, and Carole Ober.

Friday, November, 11, 2005

3:35 - 4:35 pm

206 Cox Hall

Refreshments will be served on the second floor of Dabney Hall (left of Room 222) at 3:00 pm.