AI-Empowered Virtual Screening-Driven Innovation and Practice of Virtual–Real Integrated Classroom Teaching in Medicinal Chemistry

Abstract

In order to solve the pain points of the traditional “Medicinal Chemistry” classroom such as focusing only on theory without practice, lack of applications of AI and virtual screening tools, and disconnecting teaching content from regional industry needs, we, under support of the "Emerging Engineering Education", "Artificial Intelligence Plus" action plans, proposed an integrated virtual–real innovative teaching model based on AI-based druggability prediction, virtual screening and teaching-session design; then conducted teaching with the practical case of molecular docking of hexahydrocurcumin with heat shock protein 90α (HSP90α), which is included in the chapter about natural antitumor drugs. The model realized two-way empowerment between the virtual and the physical – the average time students spend using AutoDock Vina was decreased by 32%, 92.4% of students reported that their learning interest has been significantly increased, the experimental group significantly outperformed the control group in connecting molecular docking with experimental design, and provides a replicable paradigm for cultivating pharmaceutical talents at local application-oriented undergraduate institutions.