Jian Tang

Jian Tang has developed reinforcement learning and graph neural network methods for molecular generation and drug lead optimization that enable AI-guided exploration of chemical space for cancer drug design. His graph-based generative models can design novel molecular structures optimized for multiple objectives simultaneously -- potency against cancer targets, selectivity, drug-like properties, and synthetic accessibility. He has applied multi-objective reinforcement learning to the lead optimization stage of cancer drug discovery, where balancing competing molecular properties is critical. His methods enable efficient navigation of the vast chemical design space to identify optimal cancer drug candidates that satisfy complex multi-parameter pharmaceutical requirements.