Susceptibility to chronic disease often differs between genotypes. Elucidating the role the genotype plays in the disease process contributes to understanding its etiology and developing treatments. However, the role of the genotype may be obscured by environmental factors which interact with the genotype to produce the disease. Environmental variation may produce variable expression in persons with the same genotype; cases due to the environment may be indistinguishable from cases due to the genotype; environmental factors shared within households may mimic the familiality caused by genes. Genetic models used with likelihood analysis provide an approach to reveal the effects of genes and separate them from environmental effects. This research proposed to derive and implement models used to detect the effects of genetic loci and environmental factors on disease endpoints and biochemical markers. The major gene/polygenic mixed model is used to detect genes with large effects; approximations will be derived to allow analysis of dichotomous and bivariate phenotypic data. Path analysis models are used to test for familial environmental effects; approximations will be derived for application in pedigrees and in the presence of a gene with a large effect. The models will be implemented in the computer program PAP (Pedigree Analysis Package) and tested by doing likelihood analysis on available family data. In addition, PAP will continue to be distributed to other users and be made faster and easier to use.
