Copy number variation in bipolar disorder

• Published in: Molecular psychiatry Vol. 21; no. 1; pp. 89 - 93
• Main Authors: Green, E K, Rees, E, Walters, J T R, Smith, K-G, Forty, L, Grozeva, D, Moran, J L, Sklar, P, Ripke, S, Chambert, K D, Genovese, G, McCarroll, S A, Jones, I, Jones, L, Owen, M J, O'Donovan, M C, Craddock, N, Kirov, G
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 01.01.2016; Nature Publishing Group
• Subjects: 45/43; 45/61; 45/77; 692/699/476/1333; Analysis; Autism; Behavioral Sciences; Biological Psychology; Bipolar disorder; Bipolar Disorder - genetics; Care and treatment; Complications and side effects; Datasets; Dentistry; DNA Copy Number Variations; Female; Genetic aspects; Genetic variation; Genetics; Genomes; Genomics; Genotyping Techniques; Humans; Intellectual disabilities; Male; Medicine; Medicine & Public Health; Middle Aged; Neurosciences; Original; original-article; Pharmacotherapy; Psychiatric research; Psychiatry; Schizophrenia; Schizophrenia - genetics; White People; United Kingdom; United Kingdom > UK; United States > US
• ISSN: 1359-4184; 1476-5578; 1476-5578
• DOI: 10.1038/mp.2014.174

Large (>100 kb), rare (<1% in the population) copy number variants (CNVs) have been shown to confer risk for schizophrenia (SZ), but the findings for bipolar disorder (BD) are less clear. In a new BD sample from the United Kingdom ( n =2591), we have examined the occurrence of CNVs and compared this with previously reported samples of 6882 SZ and 8842 control subjects. When combined with previous data, we find evidence for a contribution to BD for three SZ-associated CNV loci: duplications at 1q21.1 ( P =0.022), deletions at 3q29 ( P =0.03) and duplications at 16p11.2 ( P =2.3 × 10 −4 ). The latter survives multiple-testing correction for the number of recurrent large CNV loci in the genome. Genes in 20 regions (total of 55 genes) were enriched for rare exonic CNVs among BD cases, but none of these survives correction for multiple testing. Finally, our data provide strong support for the hypothesis of a lesser contribution of very large (>500 kb) CNVs in BD compared with SZ, most notably for deletions >1 Mb ( P =9 × 10 −4 ).