Integrative ensemble modeling and simulation of proteins, driven by artificial intelligence (AI) and physics, represents a frontier in contemporary structural biology. This talk will demonstrate how machine learning and generative AI apply to biomolecular MD simulations, including: 1) developing reaction coordinate deep learning methods using long-timescale MD data and flow-model generative AI; 2) building fully automated ML-enhanced sampling algorithms by fusing MD adaptive sampling, dynamic feature extraction, and enhanced sampling; 3) constructing conditional diffusion models (based on membrane protein physical constraints and conformation predictors) to generate functional-state conformations of membrane proteins (e.g., P-type ATPases).
Dr. Yong Wang is a PI at Zhejiang University, leading interdisciplinary research on multiscale modeling and AI-driven advanced sampling. Prior to this, he was an Assistant Professor at the University of Copenhagen. He holds a BSc in Chemistry (minor in Computer Science) from Jilin University, an MSc from the University of the Chinese Academy of Sciences, and a PhD in Biochemistry from the University of Copenhagen (supervised by Prof. Kresten Lindorff-Larsen). In 2016, he received an EMBO Short-Term Fellowship to research at Prof. Michele Parrinello’s lab (ETH Zurich). His work integrates atomistic simulations, coarse-grained modeling, and machine learning to elucidate dynamic mechanisms of biological nanomachines (e.g., insect odorant receptors, bacterial flagellar motors). He also collaborates with experimental teams to link computational predictions with experimental measurements via integrative structural biology.