Jixiao He

This paper introduces a deep learning framework for identifying dynamic protein allosteric sites across different conformational states, achieving improved performance over existing methods.

C-PLANK improves ligand-protein interaction prediction by integrating cellular context and fragment-based chemoproteomics data, outperforming existing methods and identifying novel ligands.

MultiAlloDriver predicts and identifies cancer driver mutations by integrating protein sequence, structure, and surface information, achieving over 93% accuracy.

ED2Mol improves de novo molecular design and lead optimization by incorporating electron density information, achieving high physical reliability and success rates in generating bioactive compounds.