Nikolay Dokholyan (Physical Seminar)
Title: "Artificial Intelligence-Driven Drug Discovery"
Nikolay Dokholyan
University of Virginia
Host: Davit Potoyan
Abstract: We introduce an integrated ecosystem of physics-based and AI-driven tools that redefine drug discovery. MedusaDock enables accurate ligand–protein binding prediction, while MedusaNetand NeuralDock scale virtual screening to billions of compounds. YuelDesign advances de novo drug design with generative AI, and DRIFT maps drug–protein interactions to predict off-target effects and extend discovery to new molecular modalities. Together, these platforms accelerate, expand, and safeguard therapeutic innovation. Looking forward, integration with quantum computing promises unprecedented modeling of molecular complexity, marking a new era where computation drives drug discovery.
Bio: Nikolay V. Dokholyan is the Weir M. and Linden C. Tucker Professor of Neurology at the University of Virginia School of Medicine. An internationally recognized leader in computational biophysics, protein engineering, and therapeutic discovery, his research integrates physics, biology, artificial intelligence, and medicine to understand complex biological systems and develop novel treatments for human disease.
Dr. Dokholyan earned his B.S. and M.S. in Physics from the Moscow Institute of Physics and Technology, a Ph.D. in Physics from Boston University, and completed postdoctoral training in Biophysics at Harvard University. He is known for pioneering advances in molecular modeling, including the development of Discrete Molecular Dynamics (DMD), Medusa force fields, and MedusaDock, computational platforms that have accelerated drug discovery and protein design.
His research has contributed to breakthroughs in neurodegenerative diseases, cystic fibrosis, protein engineering, synthetic biology, and AI-driven drug discovery. A Fellow of the American Physical Society (APS), the American Association for the Advancement of Science (AAAS), and the American Institute for Medical and Biological Engineering (AIMBE), Dr. Dokholyan has authored more than 350 peer-reviewed publications and has been cited over 26,000 times worldwide.