Arijit Roy

TRACEDD provides a framework for explainable drug design by integrating multiple agents and tools to coordinate drug discovery tasks from target validation to synthesis planning.

This paper develops an AI-guided framework for personalized cancer immunotherapy dosing, achieving high remission rates with lower median doses compared to fixed regimens.

EnzFinder predicts promiscuous enzymes for chemical synthesis by analyzing reaction patterns, outperforming existing methods and enabling greener synthesis routes.

This paper introduces SKiD, a database linking enzyme kinetics parameters with 3D structural data of enzyme-substrate complexes, enhancing understanding of enzymatic function.

This paper introduces LMLF, a method using logical feedback to guide LLMs in generating drug leads for specific targets, showing improved binding affinities and potential novelty.

The paper combines deep neural networks and Inductive Logic Programming to generate potential inhibitors for targets with limited data, using domain knowledge to improve molecule generation.