Multivariate GWAS Identifies Genetic Architecture and Tissue-specific Risk Genes for Panvascular Diseases

We employed genomic structural equation modeling (SEM) implemented in the Genomic SEM R package (v.0.0.5c) to conduct a common factor GWAS analysis across ischemic stroke (IS), coronary artery disease (CAD), peripheral artery disease (PAD), and atherosclerosis, excluding cerebral, coronary, and PAD (ATHSCLE), with the goal of uncovering shared genetic liability underlying panvascular diseases. Genomic SEM was performed in two sequential stages. In stage 1, four univariate GWASs were prepared, and the genetic covariance matrix derived from the LDSC was used as the input for the SEM common factor model. In stage 2, the common factor model was evaluated, minimizing the hypothesized covariance matrix and the empirical covariance matrix calculated in stage 1 to identify a shared genetic signature underlying panvascular disease.