Genome-wide genetic overlap analysis of the genetic contribution shared between stroke and trouble falling asleep, with cross-ethnic validation

Abstract

Stroke is among the leading causes of death and disability. Trouble falling asleep is a common sleep-related problem. Epidemiological studies show a correlation, but the genetic mechanisms remain unclear. We combined large-scale GWAS summary statistics with clinical cohort data from an Asian population for validation, and used the MiXeR model to assess the polygenic overlap between the two traits. CondFDR/conjFDR analyses were used to identify novel loci, and then these were validated by multivariable logistic regression and TWAS. The results indicated that there was substantial overlap in polygenic signals, with approximately 393 shared genetic variants identified. Clinical data showed that 46.7% of stroke patients experienced trouble falling asleep. Male sex, hemorrhagic stroke, and a history of hypertension were all independent risk factors. TWAS using the S-PrediXcan method identified four genes (ALDH2, TWIST1, ARL17A, and LRRC37A2) that may contribute to susceptibility to both stroke and trouble falling asleep. These findings suggest substantial shared genetic architecture between stroke and difficulty initiating sleep, highlighting potential pleiotropic loci that may inform future mechanistic studies and risk stratification.