Identification of therapeutic targets for insomnia through genetic and plasma proteomic approaches

Insomnia is a prevalent sleep disorder, and for which effective therapeutic targets remain lacking. In this study, we conducted a GWAS meta-analysis on 3 datasets: ukb-b-3957, FinnGen R10, and ebi-a-GCST004695, using METAL. Building on this, we further investigated the causality of plasma proteins using a protein Mendelian randomization (MR) approach. The MR analysis utilized protein datasets from deCODE and UKB, incorporating co-localization. Reverse MR was employed to examine the potential presence of reverse causality. MAGMA was employed to investigate key proteins associated with insomnia, along with tissue enrichment analysis. The insomnia GWAS meta-analysis revealed that 250 genes exhibited statistically significant signals after FDR correction (P < .05). A total of 81 risk loci were identified (P < 5e-8), and the risk genes were predominantly enriched in tissues such as the cerebellum, cerebellar hemisphere, and prefrontal cortex. Our PW-MR analysis identified 36 proteins with replicated causal associations with insomnia. After excluding one protein (HEXIM1) due to evidence of reverse causality, we established a final list of 35 high-confidence candidate proteins. Further colocalization analysis supported a shared causal variant for several candidates, including DNPH1 and PTK7. This study identifies a robust list of plasma proteins with genetically-predicted causal links to insomnia risk. These findings offer a valuable resource for subsequent functional validation studies aimed at elucidating the molecular mechanisms underlying the disorder.