Structure-Aware Dual-Target Drug Design through Collaborative Learning of Pharmacophore Combination and Molecular Simulation
Dual-target drug design has gained significant attention in the treatment of complex diseases, such as cancers and autoimmune disorders. A widely employed design strategy is combining pharmacophores to incorporate the knowledge of structure-activity relationships of both targets. Unfortunately, it often struggles with long and expensive trial and error, because protein pockets of two targets impose complex structural constraints on the pharmacophore combination. In this study, we propose AIxFuse, a structure-aware dual-target drug design method that learns pharmacophore fusion patterns to satisfy the dual-target structural constraints simulated by molecular docking. We utilize two self-play reinforcement learning (RL) agents to learn pharmacophore selection and fusion by comprehensive feedback including dual-target molecular docking scores. Collaboratively, the molecular docking scores are learned by active learning (AL). Through collaborative RL and AL, AIxFuse learns to generate molecules with multiple desired properties. AIxFuse is shown to outperform state-of-the-art methods in generating dual-target drugs against glycogen synthase kinase-3 beta (GSK3{beta}) and c-Jun N-terminal kinase 3 (JNK3). When applied to another task against retinoic acid receptor-related orphan receptor{gamma} -t (ROR{gamma}t) and dihydroorotate dehydrogenase (DHODH), AIxFuse exhibits consistent performance while compared methods suffer performance drops, leading to a 5 times outperformance in success rate. Docking studies demonstrate that AIxFuse can generate molecules concurrently satisfying the binding mode required by both targets. Further free energy perturbation calculation indicates that the generated candidates have promising binding free energies against both targets. Significance StatementComplex diseases like cancers and autoimmune disorders are mostly caused by multiple genes. Designing dual-target drugs against two target proteins simultaneously can achieve synergistic effects and alleviate drug resistance. In this study, we present AIxFuse, which to our knowledge is the first structure-aware dual-target drug design method that learns pharmacophore fusion patterns to satisfy the dual-target structural constraints simulated by molecular docking. AIxFuse exhibits superior performance to previous state-of-the-art methods on comprehensive benchmarks. By generating diverse drug candidates with promising dualtarget binding free energies and other desired properties, AIxFuse holds promising prospects for accelerating the development of novel dual-target drugs for long-term therapeutic of complex diseases.