While the prevalence of major depression is elevated amongst cannabis users, the role of genetics in this pattern of comorbidity is not clear. This study aimed to estimate the heritability of cannabis use and major depression, quantify the genetic overlap between these two traits, and localize regions of the genome that segregate in families with cannabis use and major depression.


Family-based univariate and bivariate genetic analysis.


San Antonio, Texas, USA PARTICIPANTS: Genetics of Brain Structure and Function study (GOBS) participants: 1,284 Mexican-Americans from 75 large multi-generation families and an additional 57 genetically unrelated spouses.


Phenotypes of lifetime history of cannabis use and major depression, measured using the semi-structured MINI-Plus interview. Genotypes measured using ~1 M single nucleotide polymorphisms (SNPs) on Illumina BeadChips. A sub-selection of these SNPs were used to build multipoint identity-by-descent matrices for linkage analysis.


Both cannabis use (h2  = 0.614, p = 1.00×10-6 , SE = 0.151) and major depression (h2  = 0.349, p = 1.06×10-5 , SE = 0.100) are heritable traits, and there is significant genetic correlation between the two (ρg  = 0.424, p = 0.0364, SE = 0.195). Genome-wide linkage scans identify a significant univariate linkage peak for major depression on chromosome 22 (LOD = 3.144 at 2 cM), with a suggestive peak for cannabis use on chromosome 21 (LOD = 2.123 at 37 cM). A significant pleiotropic linkage peak influencing both cannabis use and major depression was identified on chromosome 11, using a bivariate model (LOD = 3.229 at 112 cM). Follow-up of this pleiotropic signal identified a SNP 20 kb upstream of NCAM1 (rs7932341) that shows significant bivariate association (p = 3.10×10-5 ). However this SNP is rare (7 minor allele carriers) and does not drive the linkage signal observed.


There appears to be significant genetic overlap between cannabis use and major depression among Mexican-Americans, a pleiotropy that appears to be localized to a region on chromosome 11q23 that has been previously linked to these phenotypes. This article is protected by copyright. All rights reserved.

This article is protected by copyright. All rights reserved.