We are trying to develop a functional understanding of gene sequence diversity for studied systems, notably major histocompatibility complex (MHC) loci, haplotypes, and alleles, microsatellite loci, and viral sequences. Our approach is to integrate population genetic, molecular evolution, and phylogenetic considerations with data on disease association and function. In order to do this, we are developing, improving, and applying algorithms for 1) detection and characterization of recombination and/or linkage disequilibrium among closely spaced loci; and 2) detection of genetic differentiation or identity. Currently, algorithms for detecting recombination among sequences or interpreting patterns of linkage disequilibrium are very crude and ad hoc, especially with regard to interpretation of such patterns. One of our goals is the quantitative identification of genetically alternative """"""""motifs"""""""" among alleles at MHC and other loci. Ultimately, such data will be useful for testing putative disease associations. We are also exploring the potential for microsatellite applications in population and individual identification. We have developed statistical tests for microsatellite discrimination of divergent populations, population affiliation of a given genotype, and forensic testing of a DNA sample when appropriate population databases do not exist. We have applied these algorithms to existing data sets from tiger subspecies and domestic cats. The detection of genetic differentiation among tiger subspecies despite small sample sizes is a very good paradigm for detection of genetic differences (and potential disease associations) among highly characterized patient cohorts.
