1. Ascertainment and collection of pedigrees for linkage and association studies. In a collaboration we helped originate, under the aegis of the NIMH, we and several extramural centers have joined together to assemble national pedigree samples for linkage studies in Bipolar manic-depressive illness, schizophrenia, and Alzheimer;s disease. Our Branch is part of the manic-depressive effort in this collaboration. We have screened 90 families, of whom 12 were found eligible for study and are currently partially or fully collected. Since 1990, we have screened 208 families with panic, and accepted 20, and 16 families with schizophrenia, of whom we accepted 5 into our study. 2. Analysis of clinical strategy issues in linkage studies. A 6.7 cM interval genomic map gives acceptable exclusion power, in our 21 family Bipolar pedigree series, to cover the entire genome, under a genetic model of illness with 50% of pedigrees linked, dominant inheritance, and 50% penetrance. This analytic model detects linkage in data simulated under numerous oligogenic inheritance models. Replication of linkage in follow-up of previously reported linked pedigrees: Generally, simulations are required to estimate the significance of the increment or decrement in the lod scores, and the statistical power of the new data to reject or accept linkage. For some pedigree structures, a conditional lod score for follow-up can be computed as Lod (A and B, O) - Lod (A,O), where A is the distribution in he original pedigree of phenotypes and markers, and B is the distribution in the follow-up data. The new total lod score, Lod (A and B, O), is biased toward erroneously accepting linkage. 3. A longitudinal study of well adolescent offspring (age 15-25, n-30) of bipolar parents, compared to adolescent children (n-56) of normal parents, is ongoing to determine if biological or psychological variables can predict who may become ill.
