The Genetic Architecture of Brain Structure and Function: A Data-Driven Interpretation Using genomICA

Understanding the genetic basis of individual differences in human brain structure and function remains a major challenge. Traditional genome-wide association studies (GWAS) have identified numerous genetic variants associated with brain phenotypes, but their interpretation is complicated by polygenicity and pleiotropy. To address this, we applied genomICA, a data-driven multivariate method, to GWAS summary statistics of thousands of MRI-derived brain phenotypes from 33,224 UK Biobank participants. genomICA decomposes high-dimensional genetic data into independent components (ICs), capturing shared patterns of genetic influence across multiple traits. We identified 16 ICs, collectively explaining 39.2% of the variance. We describe each IC in detail here and in an online database (genomica.info). In addition to IC-brain and IC-genomic loadings, we explored the ICs biological underpinnings through gene set enrichment and trait association analyses. Our results revealed that the generated components highlighted a diversity of neurobiological processes such as stress response (IC2), inflammation (IC5, IC15), glutamatergic signaling (IC8), lipid and semaphorin pathways (IC10), and circadian rhythms (IC12). In addition, some components reflect complex behavioral/lifestyle aspects such as diet and risk taking. genomICA offers a novel, data-driven framework for dissecting the complex genetic architecture of brain phenotypes, moving beyond univariate GWAS. This exploratory study provides a valuable resource for the imaging and genetic community, with all components and associated data available at genomICA.info, serving as a starting point for hypothesis generation and downstream analyses in brain health and disease.