Searching high and low: a review of the genetics of bipolar disorder


Corresponding author: Dr James Potash, Johns Hopkins Hospital, Meyer 3-181, 600 N. Wolfe St., Baltimore, MD, 21287-7381, USA. Fax: +1 410 614 1530; e-mail:


Objectives : To review the methodologies and findings in the genetics of bipolar disorder (BPD), and to suggest future directions for research.

Methods: Reports of family, twin, adoption, linkage, association, cytogenetic, and animal model studies, and segregation analyses in English, were identified from multiple MEDLINE searches. Hand searches were carried out in bibliographies from review articles.

Results: Family, twin, and adoption studies have provided strong evidence for a genetic etiology in BPD. Early reports of linkage of BPD to DNA markers at several chromosomal sites have not proven robust, perhaps because of the complex nature of BPD inheritance. However, linkage findings in the 1990s, on chromosomes 18, 21q, 12q, and 4p, have provided leads that are being pursued through both genetic and physical mapping. No gene has yet been definitively implicated in BPD.

Conclusions: Strategies for increasing the power to detect BPD genes include: (1) dividing the phenotype into genetically meaningful subtypes to decrease heterogeneity; and (2) ascertaining a very large family sample – a multicenter study now in progress will collect 700 bipolar I sibling pairs. BPD may result from several genes acting in concert so that new multilocus statistical methods could enhance the capacity to detect loci involved. Family-based association studies using a very large number of newly identified single nucleotide polymorphisms (SNPs) may allow for more efficient screening of the genome. As the Human Genome Project approaches its goal of isolating all genes by 2003, the data generated is likely to speed identification of candidate BPD genes.