The Genetic Architecture of Amygdala Nuclei

Abstract Background Whereas a number of genetic variants influencing total amygdala volume have been identified in previous research, genetic architecture of its distinct nuclei have yet to be thoroughly explored. We aimed to investigate whether increased phenotypic specificity through segmentation of the nuclei aids genetic discoverability and sheds light on the extent of shared genetic architecture and biological pathways between the nuclei and disorders associated with the amygdala. Methods T1-weighted brain MRI scans (n=36,352, mean age= 64.26 years, 52% female) of trans-ancestry individuals from the UK Biobank were segmented into nine amygdala nuclei with FreeSurfer v6.1, and genome-wide association analyses were performed on the full sample and a European-only subset (n=31,690). We estimated heritability using Genome-wide Complex Trait Analysis, derived estimates of polygenicity, discoverability and power using MiXeR, and determined genetic correlations and shared loci between the nuclei using Linkage Disequilibrium Score Regression, followed by functional annotation using FUMA. Results The SNP-based heritability of the nuclei ranged between 0.17-0.33, and the central nucleus had the greatest statistical power for discovery. Across the whole amygdala and the nuclei volumes, 38 novel significant (p < 5×10 −9 ) loci were identified, with most loci mapped to the central nucleus. The mapped genes and associated pathways revealed both unique and shared effects across the nuclei, and immune-related pathways were particularly enriched across several nuclei. Conclusions These findings indicate that the amygdala nuclei volumes have significant genetic heritability, increased power for discovery compared to whole amygdala volume, may have unique and shared genetic architectures, and a significant immune component to their aetiology.