主持人：西湖大学理学院 邓力教授

时间：2025年5月28日（周三）下午15:00-17:30

地点：E10-201, 云谷校区

报告语言：英文

主题1：The Wonders of Electron Density: From Fractional Electrons, Excited States to Optical Spectroscopy and Noncovalent Interactions

主讲人：

杨伟涛教授，美国杜克大学

个人简介/Bio:

Professor Weitao Yang is the Philip Handler Professor of Chemistry and Physics at Duke University.  He received his BS in 1982 from Peking University and his PhD in 1986 from the University of North Carolina. 

Professor Yang's major contributions have been in the development of theoretical and computational methods in electronic structure theory. Yang pioneered the field of the linear scaling approach with the divide-and-conquer method for electronic structure calculations of large systems.  Professor Yang has contributed to the development of density-functionals that go beyond the local density-functional approximation. The Becke-Lee-Yang-Parr (B3LYP) density-functional, the combination of the Lee-Yang-Parr correlation functional with the Becke exchange functional, is the most widely used approximation in practical electronic structure calculations.  His recent studies have revealed the origins of failure of common density functional approximations as the delocalization and static correlation error, through the perspectives of fractional charges and fractional spins. This further leads to the development of much improved approximations, and the formulation of theory for excited states. Professor Yang has also developed multiscale approaches combining the methods of quantum chemistry and statistical mechanics to address the reaction mechanisms of solution and enzymatic catalysis. His contributions in developing innovative chemical concepts of fukui functions, noncovalent interactions index and frontier molecular orbitalets have been impactful.

Professor Yang has received numerous honors, including the ACS Theory Award, ACS Award for Computers in Chemical and Pharmaceutical Research, Annual Medal of the International Academy of Quantum Molecular Science, Humboldt Research Award for Senior U.S. Scientists , elected member of the International Academy of Quantum Molecular Science, Fellow of the American Association of the Advancement of Science, Fellow of the American Physical Society and Fellow of Chinses Chemical Society.

摘要/Abstract:

Interactions between electrons determine the structure and properties of matter from molecules to solids. To describe interacting electrons, the simple three-dimensional electron density can be used as the basic variable within density functional theory, negating the need in many cases for the massively complex many-electron wave function.  This lecture will review the concept of electron density and many of its remarkable features, including latest developments.  Electron density determines the energetics of chemical and physical changes, such as in drug design and catalytic processes. Electron density, as a classical variable, leads to the concepts of fractional charges and fractional spins, which are key to understanding and advancing density functional theory for broad applications, particularly in optical spectroscopy. It can also be used to reveal non-covalent interactions such as hydrogen bonds, van der Waals attractions and steric repulsions, and covalent interactions in chemical processes in complex systems. Most recently, the theory has been established for excited states, as rigorously as for ground states.

主题2：New opportunity for molecular simulations, where AI meets physics

主讲人：

高毅勤教授，北京大学

个人简介/Bio:

Yi Qin Gao received his bachelor’s degree from Chemistry Department of Sichuan University in 1993, a master’s degree from Institute of Chemistry, Chinese Academy of Sciences in 1996, and a PhD degree from California Institute of Technology in 2001. Between 2001 and 2004, he was a postdoc at Caltech and then Harvard. In 2004, he became an assistant professor in Chemistry Department, Texas A&M University. Since 2010, he has been a Professor in College of Chemistry & Molecular Engineering, Peking University. He joined BIOPIC as a PI in 2013. He is a New Cornerstone Fellow, the associate dean of School of Science of Peking University, and an associate editor of the ACS journal JCTC.

摘要/Abstract:

In this talk, we will discuss on recent progress made on the development of AI assisted molecular dynamics simulation package, SPONGE, and its application to various molecular systems including studies on chemical reactions and protein related modeling tasks. In protein and protein complex structure prediction, we introduce GRASP, which efficiently and flexibly incorporates the different forms of information obtained through relatively high throughput experiment, such as XL, CL, CSP, and DMS. GRASP significantly improves AI model’s capability of structure prediction. We also developed a probabilistic tokenization model for metastable protein structures, ProTokens. By unifying 1-dimensional and 3-dimensional representations of protein structures, ProTokens also enable sequence and all-atom protein structure design via various generative models. We demonstrate that generative pretraining over ProToken vocabulary allows scalable foundation models to perceive, process and explore the microscopic and dynamical structures as well as functions of biomolecules through this sequence-strucure-function-leterature multi-mode model. We finally show how these various simulation methods and packages can be conveniently executed through the help of an AI agent.  

联系人/Contact:

理学院，朱老师，zhuyi48@westlake.edu.cn