Background: Recently, genome-wide association studies (GWAS) for cases versus controls using single nucleotide polymorphism microarray data have shown promising findings for complex neuropsychiatric disorders, including bipolar disorder (BD).Methods: Here we describe a comprehensive genome-wide study of bipolar disorder (BD), cross-referencing analysis from a family-based study of 229 small families with association analysis from over 950 cases and 950 ethnicity-matched controls from the UK and Canada. Further, loci identified in these analyses were supported by pathways identified through pathway analysis on the samples.Results: Although no genome-wide significant markers were identified, the combined GWAS findings have pointed to several genes of interest that support GWAS findings for BD from other groups or consortia, such as atSYNE1on 6q25,PPP2R2Con 4p16.1,ZNF659on 3p24.3,CNTNAP5(2q14.3), andCDH13(16q23.3). This apparent corroboration across multiple sites gives much confidence to the likelihood of genetic involvement in BD at these loci. In particular, our two-stage strategy found association in both our combined case/control analysis and the family-based analysis on 1q21.2 (closest gene: sphingosine-1-phosphate receptor 1 gene,S1PR1) and on 1q24.1 near the geneTMCO1, and atCSMD1on 8p23.2, supporting several previous GWAS reports for BD and for schizophrenia. Pathway analysis suggests association of pathways involved in calcium signalling, neuropathic pain signalling, CREB signalling in neurons, glutamate receptor signalling and axonal guidance signalling.Conclusions: The findings presented here show support for a number of genes previously implicated genes in the etiology of BD, includingCSMD1andSYNE1, as well as evidence for previously unreported genes such as the brain-expressed genesADCY2,NCALD,WDR60,SCN7AandSPAG16.

The 5-HT2C receptor has been hypothesized to represent an important modulator in feeding behaviour. Evidence was based on the observation that knock-