Structure-Based Drug Discovery: From the Pre-AI Era to the AI Era

In this lecture, Mykola Protopopov provided a structured overview of the evolution of Structure-Based Drug Discovery (SBDD), tracing its development from classical crystallography-driven ligand design to contemporary AI- and machine learning–assisted approaches. He highlighted key methodological milestones that shaped the field and discussed how advances in structural biology and computational chemistry enabled systematic exploration of chemical space.

Particular emphasis was placed on the exponential growth of chemical space and the role of modern AI/ML-driven tools in accelerating virtual screening, target engagement analysis, and lead optimization. Drawing on practical examples, the lecture demonstrated how

• homology modeling,
• virtual screening,
• and in silico prediction of biological activity

have been successfully applied to the development of bioactive compounds targeting protein kinases and microorganisms, illustrating the transformative impact of AI on modern drug discovery workflows.