Abstract 9679: Atrial Fibrillation Disease Module on a Human Atrial Protein Interactome Reveals Novel Genes, Biological Pathways, and Drug Candidates

Introduction:Atrial fibrillation (AF) is a complex and progressive disease leading to the development of atriopathy, but pathobiological mechanisms beyond ion channels are poorly understood. We aimed to identify an AF disease module on a human atrial protein interactome and validate it for pathobiological relevance using large-scale genome-wide association studies (GWAS) and network analysis.Methods:A human atrium-specific protein-protein interaction (PPI) network was constructed from the GTEx RNA-seq data (n=264). The AF disease module was constructed by applying the seed connector algorithm to AF-related proteins (n=93) obtained from OMIM, GWAS, and literatures, and was validated using a meta-analysis of six European GWAS data (HUNT, deCODE, MGI, DiscovEHR, UK Biobank, and AFGen Consortium; n=60,620 AF and 970,216 controls). Network analysis and functional enrichment analysis were performed to discover novel AF-associated genes, pathways, and drugs.Results:The AF disease module (124 proteins with 172 interactions) was enriched with AF-susceptibility variants and modest GWAS p-values against the random variation (p<10-5, permutation test). Among the seed connectors, CDKN1A and NPR1 were assigned top betweenness centrality scores (p<0.001 vs. other AF module genes) and showed significant gene-level associations (q<0.05). The AF disease module was enriched with muscle contraction, cardiac conduction, and interleukin-4/13 signaling pathways (p<0.001) and significantly interacted with inflammation, fibrosis, and thrombosis endophenotype modules on the PPI network (p<10-10vs. random network). Network proximity-based drug repurposing analysis showed that drug targets of siltuximab and oseltamivir were much closer to the AF disease module than expected by chance (p<0.001).Conclusions:We used a network medicine approach to identify the AF disease module on the atrial protein interactome and discovered potential AF targets, pathways, and drug candidates that deserve further investigation.