Qiancheng Shen

Cross-state representation learning enables prospective discovery of dynamic protein allosteric sites

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

Accurate de novo design of peptides from programming biophysical landscape

NeoPep designs functional peptides de novo using a deep-learning framework that models biophysical principles, achieving high hit rates and structural accuracy across diverse targets.

Identification of non-covalent inhibitors for the atypical peroxiredoxin PRDX5 as a therapeutic strategy in malignant pleural mesothelioma

The paper identifies non-covalent inhibitors of PRDX5, a protein upregulated in MPM, through a machine learning guided pipeline and biochemical validation, suggesting it as a therapeutic target.

Sampling Function-Related Metastable States of Proteins With DASH

DASH samples function-related metastable states of proteins, revealing new folding states and allosteric sites, verified by experiments and crystal structures.