I build deep-learning and multimodal AI frameworks to understand and design targeted therapeutics — from molecular glues and PROTACs to metabolite-mediated protein interactions, and scalable models that learn the language of chemistry.
Research Scholar, Department of Computational Biology, IIIT-Delhi · New Delhi, India
My research sits at the intersection of machine learning, molecular biology, and targeted therapeutics. I develop representation-learning and structure-aware deep-learning models for chemistry, proteins, signalling systems, and drug discovery. A central goal of my work is to convert large-scale biological and chemical data into mechanistic, testable hypotheses that can guide therapeutic design. Alongside my scientific work, I care deeply about clear scientific communication, visual explanation, and illustrations that make complex molecular systems accessible.
AI for molecular glues, PROTACs and ternary-complex design — degrading the "undruggable" with selectivity by design.
Modelling how small molecules and metabolites stabilize, weaken or rewire protein–protein interactions.
Representation learning that fuses structure, graphs, descriptors and chemical language into transferable embeddings.
Structure- and chemistry-aware deep learning that yields interpretable, mechanistic insight rather than black-box scores.
bioRxiv
doi.org/10.1101/2025.08.28.672835 →eLife
doi.org/10.7554/eLife.106397 →bioRxiv
doi.org/10.1101/2025.11.11.687860 →Chemistry — Methods, 6(4), e202500160
doi.org/10.1002/cmtd.202500160 →Zenodo
doi.org/10.5281/zenodo.4266067 →Intl. Conf. on Recent Trends in Biotechnology & Bioinformatics (ICBAB), JUIT
BuG ReMeDEE Consortium, South Dakota School of Mines & Technology
Plant biotechnology & tissue culture
Direct organogenesis for rapid in-vitro propagation of Stevia rebaudiana
A complete and current list is available on Google Scholar.