LAUSR

Background Brain morphometry is both heritable and correlated with psychiatric disorders. Notable structural differences include reduced volume in subcortical structures, some are observed in several disorders while others are specific to one or two disorders. The prevalence of most psychiatric disorders differs markedly by sex and sex differences are also observed in the volume of subcortical structures. Moreover, brain structure and psychiatric disorders are both highly heritable. Here we assess if genetic variants influencing subcortical brain structures differ between females and males and if the genetic overlap between brain structure and risk of illness differs by sex. Methods We present preliminary results from sex-stratified GWAS meta-analyses on seven subcortical structures derived from T1-weighted magnetic resonance imaging scans from ~6000 females and ~5400 males from 22 cohorts from the ENIGMA consortium. This sample is a subset of the larger ENIGMA consortium analyses on variation in subcortical structures among ~30,000 individuals. The subcortical structures analysed were the nucleus accumbens, amygdala, caudate nucleus, hippocampus, globus pallidus, putamen, and thalamus. The volume of these structures was calculated with validated and freely available software, and mean volume calculated from the left and right hemispheres. Genotypic data were imputed to the 1000 genome reference panel. Results Preliminary results in this discovery meta-analysis identified four genome-wide significant loci, three were novel. One intergenic locus (14q22.3), which we have previously reported as significant in a meta-analysis that combined females and males, was also associated with putamen volume in females and in males in these stratified analyses. Of the novel loci, one was an intronic locus (11q15.2) within SPON1 associated with nucleus accumbens volume in females. The second was an intronic locus (2q34) within ERBB4 associated with amygdala volume in males. The third was an intergenic locus (13q21.31) associated with putamen volume in males. In this discovery sample SNP heritability was .18 (se=.08) for the nucleus accumbens in females, was .16 (se=.08) for the putamen in females, and was .14 (se=.09) for the thalamus in males. Genetic correlations were between each of these traits and GWAS results from the psychiatric genetics consortium on schizophrenia, bipolar, major depressive disorder, anorexia, and autism were not significant. Heritability in all other structures was close to zero, thus genetic correlations were not calculated on this discovery sample. Discussion These preliminary results point to heterogeneity between females and males in genetic influence on some subcortical structures. We plan to extend these findings with data from the UK Biobank. We will present SNP heritability and genetic correlation with psychiatric and cognitive traits using the results from the second round of meta-analysis.