Dissecting the shared genetic architecture and biological mechanisms linking sleep traits and suicide attempts: from...

Comorbidities and genetic correlations between sleep traits and suicide attempts suggest potential shared genetic architecture, but pleiotropic genetic variants and functions remain unclear. This study aimed to identify shared genetic variants, genes, and biological mechanisms linking sleep traits with suicide attempts. We conducted a genome-wide pleiotropic analysis using GWAS summary statistics for 11 sleep traits and suicide attempts. Genetic correlations and overlaps were assessed using LDSC and GPA. Pleiotropic variants were identified via PLACO, functionally annotated with FUMA, and evaluated by Bayesian colocalization. SuSiE fine-mapping and SuSiE-based colocalization further refined variant-level signals. Gene-level associations were examined using MAGMA, followed by SMR integrating blood and brain cis-eQTL data. Prioritized genes were characterized through phenotype, pathway, cell-type enrichment, and DGIdb druggability analyses. Bidirectional causality was explored via Mendelian randomization. Six sleep traits exhibited significant genetic correlations with suicide attempts, while all 11 showed significant polygenic overlap. We identified 2,224 trait-pair-specific pleiotropic variants (1,640 unique) across 50 trait-pair-specific risk loci. Self-reported sleep duration and daytime napping contributed the most variants and loci, whereas least active 5 h timing and sleep efficiency showed the fewest. Colocalization supported shared causal variants at 15 loci. Thirty-two pleiotropic genes were prioritized, showing enrichment in metabolic, developmental, synaptic, immune, and H3K27me3-related pathways. Thirteen genes (40.63%) were druggable, including five with approved drug interactions. Genetically predicted morning chronotype was associated with lower odds of suicide attempts. These findings provide new insights into the shared genetic basis linking sleep traits and suicide attempts and highlight candidate druggable genes for future translational research.