Chennai: The Indian Institute of Technology Madras has announced a breakthrough Artificial Intelligence (AI) framework that can drastically reduce the early-stage timelines of drug development — currently a billion-dollar, decade-long process.

Researchers from IIT Madras' Robert Bosch Centre for Data Science and AI and the Wadhwani School of Data Science and AI (WSAI) collaborated with Ohio State University, USA, to develop the framework.

According to the institute, the new AI system, called PURE (Policy-guided Unbiased Representations for Structure-Constrained Molecular Generation), stands apart from existing molecule-generation tools that rely on rigid scoring mechanisms or statistical optimisation.

Prof. B Ravindran, Head of WSAI, said, “What's unique about PURE is how it uses reinforcement learning — not just to optimise specific metrics, but to learn how molecules transform. By treating chemical design as a sequence of actions guided by real reaction rules, PURE moves us closer to AI systems that can reason through synthesis steps like a chemist would.”

PURE was tested on widely accepted benchmarks, including drug-likeness (QED), dopamine receptor activity (DRD2) and solubility, showing higher diversity and synthesizability in generated molecules.

Prof. Karthik Raman, WSAI, said the framework grounds its molecular search in lab feasibility, ensuring the generated compounds can be practically synthesised.

Prof. Srinivasan Parthasarathy of Ohio State University said PURE offers “game-changing” advantages for pharmaceutical research, with potential applications in drug and material discovery.

The findings have been published in the Journal of Cheminformatics, a peer-reviewed publication focusing on computational chemistry and data-driven molecular design.