Niu Huang

PI

National Institute of Biological Sciences, Beijing

Short Biography

Dr. Niu Huang earned his undergraduate degree in Physics from Nankai University. He received his doctoral degree in computational chemistry from the University of Maryland at Baltimore, followed by postdoctoral training at the University of California at San Francisco. He is currently a principal investigator at the National Institute of Biological Sciences (NIBS), Beijing. He has authored over 60 peer-reviewed publications in leading journals in the fields of computational chemistry and medicinal chemistry, and has been granted 6 international drug patents. His lab has demonstrated that the development and application of more accurate treatments of physical interactions greatly facilitate structure-based ligand discovery. Currently, several novel preclinical drug candidates for treating unmet medical needs have been developed in his lab, and these compounds have also served as specific chemical probes to facilitate in vivo functional studies of the protein targets of interest.

Presentation Topic:  Integrating HPC and AI: A New Paradigm for Predicting Protein-ligand Binding Interactions
In the process of small molecule drug discovery, the prediction of protein-ligand interactions urgently demands enhancements in computational accuracy and efficiency, given its crucial role in identifying novel lead compounds for new targets. However, current artificial intelligence (AI) models are constrained by the scarcity of large, high-quality protein-ligand complex structures and binding data, which consequently impairs their generalization ability, limiting their effectiveness in real-world applications. We have been actively exploring the potential of physics-based high-performance computing (HPC). The remarkable computational power of HPC allows us to generate vast, top-tier datasets that are invaluable for both training and testing AI modes. When integrated with AI’s proficiency in pattern recognition and predictive modeling, this combination allows for the rapid and in-depth analysis of molecular structures, more accurate prediction of drug-target interactions. Our ongoing research and practice will highlight the profound synergy between HPC and AI in facilitating more accurate and efficient calculations of molecular interactions, illuminating viable strategies to surmount existing data limitations and improve the predictive capabilities of AI models.