This proposal has three major goals. First, we evaluate existing statistical techniques and propose new tests for detecting nonallelic heterogeneity in human genetic disease through linkage studies. We will analyze family data on Mendelian disorders and expect to discover greater heterogeneity than possible before. Since part of this heterogeneity may be due to population variability in recombination, we will evaluate this by analyzing linkage data on DNA polymorphisms, protein and blood group loci. DNA polymorphisms in human Beta-globin and growth hormone clusters will also be used to estimate recombination at the molecular level and study its variation with physical distance. Using computer simulations, we shall critically examine the assumptions made in these analyses. These studies will elucidate the degree and nature of linkage heterogeneity in general and nonallelic heterogeneity in particular. Second, we will determine the degree of allelic heterogeneity (number of different mutant alleles) that can be maintained under different selection schemes, migration, recurrent mutation, inter-allelic gene conversion, and population growth and bottlenecks. By analyzing data on Beta-globin mutations and linked DNA polymorphisms we wish to delineate heterogeneity and determine probable mechanisms of origin of specific mutations. We hope to test the hypothesis that the high frequency of certain Beta-globin mutations in human populations is due to multiple origins. Third, we will study the utility and efficiency of closely linked multiple marker genes for the prenatal diagnosis of a heterogeneous genetic disease. Specifically, we will determine the effect of non-random associatins between markers on the proportion of informative families and the efficacy of a sequential marker testing scheme. These studies are aimed at discovering a general strategy of using the fewest markers for detecting the majority of affecteds and gene carriers.
Showing the most recent 10 out of 40 publications
