Graph neural networks driven acceleration in drug discovery
Rui Wang, Chunlin Zhuang
Abstract
Graph neural networks (GNNs) are revolutionizing drug design processes. Over the past five years, GNNs have emerged as transformative tools by accurately modeling molecular structures and interactions with binding targets. Breakthroughs in predicting molecular properties, drug repurposing, toxicity assessment, and interaction analysis, along with generative GNNs enhancing virtual screening and novel molecule design, have significantly sped up drug discovery. These GNN-driven innovations improve predictive accuracy, cut development costs, and reduce late-stage failures. This review focuses on the interdisciplinary integration of GNNs throughout the discovery process, including lead discovery and optimization, synthetic route design, drug–target interaction prediction, and molecular property profiling, while critically evaluating the challenges in translational medicine. GNNs are revolutionizing drug design by accurately modeling molecular structures and interactions. They enhance property prediction, drug repurposing, toxicity assessment, and novel molecule design, accelerating discovery while reducing costs and failures.