Extensive information on ascertainment of cases and controls is available in a related publication (Smith et al., 2009). Briefly, cases were unrelated subjects diagnosed with DSM-IV BD 1 as well as a small subset with BD spectrum disorders, using best estimate diagnosis. We used data collected using the DIGS 4.0 (Nurnberger, Jr. et al., 1994) from 321 European-American and 150 African-American individuals. Controls were drawn from a larger pool of subjects from across the U.S. who agreed to donate a blood sample for transformation into lymphoblastoid cell lines and to respond to a medical questionnaire. Only individuals with complete or near-complete psychiatric questionnaire data who did not fulfill diagnostic criteria for major depression, and did not report a history of schizophrenia, bipolar disorder or psychosis were retained – further matching for gender and ethnicity with BD cases allowed for inclusion of 1081 European-American and 680 African-American subjects.
