MATERIAL AND METHODS
We studied the segregation pattern of MDI and HLA antigens combined in eight families, whose probands were diagnosed as having Major Affective Disorders, Recurrent, according to DSM-III (1), not taking into account the polarity of the disease. We investigated the first- and second-degree relatives of each proband with the Family Study Method using the same diagnostic criteria; for third- and fourth-degree relatives, as well as second-degree, when a personal interview was not possible, we used the Family History Method. All the probands and some of their first- and second-degree relatives were typed for their HLA structure, blind of their clinical diagnosis. For untyped subjects their HLA patterns were reconstructed when enough information allowed. To verify the hypothesis of a linkage between MDI and HLA, we applied the LIPED model (7). Parameters describing the MDI susceptibility locus were derived by the application of the multiple threshold models of transmission to a large sample of subjects and then experimentally tested (10). The result was a dominant SML with a reduced penetrance for the heterozygotes (q = 0.013, f(AA) = 0/0,f(Aa) = 0.923/0.926,f/(aa) = 1/1). Gene frequencies for HLA antigens were derived by a control population.