1. Analytic methods in clinical genetic studies. a) Statistical power of current methods to detect linkage and heterogeneity, when penetrance is less than one (as appears present in the linked psychiatric disorders): Linkage can be detected with very feasible sample sizes of small families in the presence of considerable heterogeneity. When flanking markers are available, both linkage and heterogeneity are more readily detectable. b) Association methods. These are useful in attempting to identify candidate causative genes within a linkage region, through association with illness due to linkage disequilibrium. Analysis of the power of detection of disequilibrium in the presence of heterogeneity reveals that for several realistic parameter sets, only one or two linked pedigrees are needed. 2. Pedigree collection for manic-depressive illness and schizophrenia molecular mapping studies continues. In manic- depressive illness, we have 15 pedigrees with more than 200 individuals in culture. In schizophrenia, we have 7 pedigrees with more than 70 individuals in culture. 3. Family study of bulimia has been completed, supporting the co- aggregation in families of anorexia and bulimia suggested in earlier studies of anorexia, and demonstrating the co-aggregation of affective disorders with bulimia.
