Two population genetic features of the human histocompatibility (HLA) system make it ideally suited for the study of evolutionary forces acting on a population. These are the high level of polymorphism exhibited by many of the loci of the HLA system, and the existence of significant linkage disequilibrium (non-random association) between certain pairs of antigens at different loci. The distribution of allele frequencies at the HLA loci, and the patterns of linkage disequilibria, are consistent with selective events acting in the region. This proposal involves: 1) population data analysis of a number of very large data sets, including the Caucasian and Japanese data from the 8th (1980) and 9th (1984) International Histocompatibility Workshops, Japanese data from the All Japanese HLA Workshops of 1983 and 1985, Chinese data from Singapore, Indian data from New Delhi, extension of our previous analysis of a large Danish study to include additional HLA loci, and the data that will be available from the Third Asia Oceania Histocompatibility Workshop (Sapporo, Japan, 1986); 2) simulation of two and three locus neutrality models, including migration, bottleneck and founder effects; 3) statistical analysis of the distributional properties of population wide measures of disequilibrium and application of bootstrap and jackknife techniques to determine confidence limits of the measures; 4) deterministic analysis of selection models; 5) comparison of generated population genetic parameters under neutrality and selection models with the HLA data; 6) correlation of HLA antigen disease association patterns with linkage disequilibrium patterns, including comparison of patterns in different racial groups, to study the evolution of disease predisposing genes. The nature of selective events acting in the HLA region will be delineated, and the mechanism by which disease predisposing genes become common in a population investigated. The methods to be developed have applicability to other multigene families, marker trait associations and restriction fragment length polymorphism data, in all organisms.
Showing the most recent 10 out of 65 publications
