Na (Michael) Li (Joint work with Matthew Stephens)
Department of Biostatistics
University of Washington, Seattle
Wednesday, February 12, 2003
3:30 PM
Mayo C231 (Todd)
Minneapolis Campus
Abstract:
Linkage disequilibrium (LD) is the non-independence of alleles at different loci at the population level. LD usually arises from mutation, is maintained by linkage, and decays with recombination. Thus, variations in local recombination rate play an important role in shaping the patterns of LD. Understanding patterns of LD, and how they relate to the underlying recombination rate, has become of particular interest recently, most obviously because of its potential impact on the mapping of disease genes in humans.
Current methods for interpreting the patterns of LD are limited. They often rely only on pairwise LD measures and make poor use of the data. In addition, they do not provide a framework for likelihood-based inference. More sophisticated coalescent-based statistical methods are computationally impractical even for moderate-sized regions. Furthermore, they all assume constant recombination rate, making them poor tools for studying local recombination rates. Here we present a novel computationally-tractable statistical model for LD across multiple loci. We apply this model to the problem of inferring recombination rates from population data, and in particular identifying variation in the local recombination rate along chromosomes.