IIT Madras, Ohio State University develop AI framework to aid drug discovery
The model, called PURE (policy-driven unbiased proxies for structure-constrained molecular manufacturing), aims to reduce the early-stage drug discovery process, which often costs billions of dollars and can take a decade or more.
It may be particularly useful in combating drug resistance in cancer and infectious diseases.
Srinivasan Parthasarathy of Ohio State University said the model could accelerate the search for alternative drug candidates, especially in cases of resistance or toxicity, and support discoveries in new materials research.
Unlike existing AI tools that rely on predefined scoring or optimization metrics, PURE uses reinforcement learning to simulate how molecules transform through real chemical reactions. This allows it to generate new, diverse, and synthetically applicable molecules without being explicitly trained on these evaluation parameters.
It also identified plausible synthetic routes for produced molecules, making it a general-purpose molecular discovery engine. WSAI president B Ravindran said the framework treats “chemical design as a set of actions guided by actual reaction rules” and enables AI systems to reason through synthesis, just like a chemist. Karthik Raman, also from WSAI, added that PURE’s reaction rule-based approach “grounds molecule production in synthesizability” and addresses a key challenge in computational drug design.
PURE’s approach, which blends self-supervised and reinforcement learning, helps overcome a persistent limitation in AI-driven drug design where many virtual molecules cannot be synthesized in the laboratory.
By linking digital discovery to actual chemical synthesis, the model could help shorten development timelines and increase the success rate of early-stage drug candidates, researchers say.
