Our group mainly focuses on machine learning, computational biology, large-scale data analysis and their intersections. We develop and apply machine learning algorithms to analyze high-throughput biological data (e.g., genomic data and cryo-EM images) and gain important insights into complex molecular mechanisms and cellular functions. Below are some of our current research projects:

Artificial intelligence / machine learning for drug discovery and drug repositioning

We develop effective artificial intelligence (machine learning) algorithms to predict new drug-target interactions, which are then validated experimentally to discover new drugs or find the new uses of old drugs.

Three-dimensional genome structure and gene regulation

We develop methods to analyze the genome-wide chromosome conformation capture data (e.g., Hi-C) and model the three-dimensional genome structure. By integrating the modeled high-order genome architecture with large-scale epigenomic data, we are able to better understand the underlying mechanisms of gene regulation.

Computational method development in Cryo-EM

We develop efficient computational methods to analyze large-scale cryo-EM image data, reconstruct the three-dimensional map and determine the molecular complex structures.

Applications of Artificial Intelligence/Machine Learning in computational genomics

We are interested in developing and applying machine learning tools to analyze large-scale genomic data, e.g., analysis of genome-wide genomic sequence data to study RNA post-transcriptional regulation and modifications.