%0 Journal Article %A David M. Howard %A Toni-Kim Clarke %A Mark J. Adams %A Jonathan D. Hafferty %A Eleanor M. Wigmore %A Yanni Zeng %A Lynsey S. Hall %A Jude Gibson %A Thibaud S. Boutin %A Caroline Hayward %A Pippa A. Thomson %A David J. Porteous %A Blair H. Smith %A Alison D. Murray %A Major Depressive Disorder Working Group of the Psychiatric GWAS Consortium %A Chris S. Haley %A Ian J. Deary %A Heather C. Whalley %A Andrew M. McIntosh %T The stratification of major depressive disorder into genetic subgroups %D 2017 %R 10.1101/134601 %J bioRxiv %P 134601 %X Major depressive disorder (MDD) is a heritable condition (h2 = 37%)1 and a leading cause of disability worldwide2. MDD is clinically heterogeneous and comorbid with a variety of conditions and it has been hypothesised that this causal heterogeneity may have confounded previous attempts to elucidate its genetic architecture3-5. We applied a relatively new technique, Buhmbox6, to identify the presence of heterogeneous sub-groups within MDD using summary data from genome-wide association studies. We analysed two independent cohorts (ntotal = 31,981) and identified significant evidence (Pcorrected < 0.05) for 10 sub-groups across both cohorts, including subgroups with a liability for migraine, alcohol consumption and eczema. The most notable subgroups (Pcorrected ≤ 2.57 × 10−8 in both cohorts) were for blood levels of cholesterol and triglycerides, and blood pressure, indicating subgroups within MDD cases of individuals with a genetic predisposition for anomalous levels of these metabolic traits. Our findings provide strong evidence for novel causal heterogeneity of MDD and identify avenues for both stratification and treatment. %U https://www.biorxiv.org/content/biorxiv/early/2017/05/05/134601.full.pdf