The role of machine learning in molecular modeling and drug discovery is rapidly growing. This course introduces fundamental concepts and methods of chemoinformatics, structural biology, and machine learning. Topics covered include molecular representations, physics based protein modeling, ligand docking, chemical conformation generation, virtual screening, bioactivity data modeling, chemical data curation, drug discovery pipeline, machine learning bioactivity prediction, deep-learning architectures for molecular systems. We will discuss recent breakthroughs in applying scientific foundation models and generalizability of predictions for drug discovery. Emphasis of the classes is in understanding fundamental concepts in bioinformatics and practical application of data science tools and methods to problems in medicinal chemistry and their own research. By the end of the course students will be able to use ML models to tackle challenges in molecular modeling and drug discovery, including defining objectives as quantifiable tasks, curating chemical, structural, and bioactivity data, and appling state of the art machine learning methods.
In the Winter of 2025, Matthew O'Meara and Terra Sztain co-taught this course as BIONF595 at the University of Michigan. Lectures, labs, and final projects are in the BIONF595w25 directory
In the Winter of 2026, Matthew O'Meara taughtthis course as BIONF595 at the University of Michigan. Lectures, labs, and final projects are in the BIONF595w26 directory