King's College London

Research portal

Phenotypic Association Analyses with Copy Number Variation in Recurrent Depressive Disorder

Research output: Contribution to journalArticlepeer-review

James J H Rucker, Katherine E. Tansey, Margarita Rivera, Dalila Pinto, Sarah Cohen-Woods, Rudolf Uher, Katherine J. Aitchison, Nick Craddock, Michael J. Owen, Lisa Jones, Ian Jones, Ania Korszun, Michael R. Barnes, Martin Preisig, Ole Mors, Wolfgang Maier, John Rice, Marcella Rietschel, Florian Holsboer, Anne E. Farmer & 4 more Ian W. Craig, Stephen W. Scherer, Peter McGuffin, Gerome Breen

Original languageEnglish
Pages (from-to)329–336
JournalBiological psychiatry
Issue number4
Early online date25 Feb 2015
Accepted/In press13 Feb 2015
E-pub ahead of print25 Feb 2015
Published25 Feb 2015


  • 1_s2.0_S000632231500150X_main

    1_s2.0_S000632231500150X_main.pdf, 181 KB, application/pdf

    Uploaded date:03 Feb 2016

    Version:Final published version

    Licence:CC BY

King's Authors



Defining the molecular genomic basis of the likelihood of developing depressive disorder is a considerable challenge. We previously associated rare, exonic deletion copy number variants (CNV) with recurrent depressive disorder (RDD). Sex chromosome abnormalities also have been observed to co-occur with RDD. 


In this reanalysis of our RDD dataset(N = 3106 cases; 459 screened control samples and 2699 population control samples), we further investigated the role of larger CNVs and chromosomal abnormalities in RDD and performed association analyses with clinical data derived from this dataset. 


We found an enrichment of Turner's syndrome among cases of depression compared with the frequency observed in a large population sample(N = 34,910) of live-born infants collected in Denmark (two-sided p = .023, odds ratio = 7.76 [95% confidence interval = 1.79-33.6]), a case of diploid/triploid mosaicism, and several cases of uniparental isodisomy. In contrast to our previous analysis, large deletion CNVs were no more frequent in cases than control samples, although deletion CNVs in cases contained more genes than control samples (two-sided p = .0002). 


After statistical correction for multiple comparisons, our data do not support a substantial role for CNVs in RDD, although (as has been observed in similar samples) occasional cases may harbor large variants with etiological significance. Genetic pleiotropy and sample heterogeneity suggest that very large sample sizes are required to study conclusively the role of genetic variation in mood disorders.

Download statistics

No data available

View graph of relations

© 2020 King's College London | Strand | London WC2R 2LS | England | United Kingdom | Tel +44 (0)20 7836 5454