Identifying therapeutic targets supported by genetic evidence has become a key strategy for improving the success rate of drug development. Existing approaches typically prioritize candidate genes at genome-wide association study (GWAS) loci using molecular phenotypes. However, these approaches are limited by the availability and tissue specificity of molecular datasets and often rely on strong locus-specific signals, restricting their ability to identify causal genes. Here, we present FINE, a gene prioritization framework that leverages genome-wide genetic architecture to identify therapeutic targets directly from GWAS summary statistics. FINE integrates network enrichment analysis with association signals across the genome to improve both statistical power and biological interpretability. Through comprehensive benchmarking using gold-standard trait–gene pairs derived from large-scale pQTL datasets, FINE provides more robust and accurate gene prioritization than state-of-the-art methods. Application of FINE to complex traits, including obesity, coronary artery disease, schizophrenia, and type 2 diabetes (T2D), successfully prioritizes causal genes at disease-associated loci and identifies potential drug-target genes. Notably, this strategy enables FINE to prioritize causal genes not only within genome-wide significant loci but also in regions lacking significant variants. Furthermore, analysis of breast cancer shows a ~4-fold enrichment for clinically validated drug targets and biomarkers. Together, we demonstrate that FINE provides a powerful framework for translating human genetics into drug discovery.

therapeutic target,fine-mapping,prioritization,GWAS