Generative AI for Biomedicine

Catalog description

Recent progress of Artificial Intelligence has been transforming the approaches of scientific research across various disciplines. Generative AI models, such as AlphaFold, have become indispensable tools in fundamental biomedical research. This course offers students an opportunity to explore the latest developments in generative AI applied to biomedicine. Topics include models and methods for the prediction of protein structure from sequences, characterization of genome functions and interactions, modeling of cellular structures and tissue organizations, single cell biology, and drug design. We will cover a variety of models, such as pre-trained biological foundation models, diffusion models, Monte Carlo methods, graph neural networks, etc. Through this course, students will gain a deep understanding of how generative AI can be leveraged to address complex challenges in biomedicine. Specifically, we have the following Learning Objectives: 1. Solid understanding generative AI models. 2. Comprehensive knowledge and interdisciplinary thinking about generative AI application to key biomedical applications, e.g., protein structure, regulatory sequence design, cellular structure and function, and drug design 3. Critical analysis of research papers on generative AI methodologies and their applications to biomedicine. 4. Project-based learning and problem solving. Prerequisites: 10-301 or 10-601 or 11-685 or 10-315 or 10-701 or 11-785