Academic Profile : Faculty

Asst Prof Xia Kelin.jpg picture
Asst Prof Xia Kelin
Assistant Professor, School of Physical & Mathematical Sciences - Division of Mathematical Sciences
External Links
 
Dr. XIA obtained his PhD degree from the Chinese Academy of Sciences in Jan 2013. He was a visiting scholar at the Department of Mathematics, Michigan State University from Dec 2009- Dec 2012. From Jan 2013 to May 2016, he worked as a visiting assistant professor at Michigan State University. He joined Nanyang Technological University in Jun 2016.
My group focuses on Mathematical AI for Molecular Sciences. We use computational tools from PDE, differential geometry, algebraic topology and statistical learning to study the material, chemical and biomolecular structure, flexibility, dynamics, and functions. In particular, we are interested in topological data analysis (TDA) and generalized persistent models (including weighted persistent homology, persistent Ricci curvature, persistent spectral, etc), TDA-based machine learning/deep learning models, and their applications in perovskite design, catalyst design, polymer design, drug design, biomolecular interaction analysis, chromosome structure analysis, and more generally molecular data analysis from materials, chemistry, and biology.
 
  • Bipartite-Graph-Based Machine Learning Models for Biomolecular Interaction Analysis
  • Geometric And Topological Modeling Of Biomolecular Structure, Dynamics And Function
  • Hodge Laplacian based deep learning models for drug design
  • Improving Photocatalysts That Recycle Microplastics to Fuels by Artificial Intelligence
  • Incorporating Quantum Chemistry Descriptors and Topological Features to Study G-quadruplex-stabilizer Complexes by Machine Learning Models
  • Indoor air microbiomes as a gateway to exposome-guided precision medicine for respiratory disease
  • Machine Learning Models Based On Weighted Persistent-Homology for Structure-Based Drug Design
  • Topology-Based Featurization for Machine Learning Models in Materials Informatics