Chijian Xiang

StructureSAFE: A structure-aware chemical language model for unified hit identification and lead optimization

StructureSAFE is a chemical language model that combines protein structure and evolutionary information with molecular representations to improve hit identification and lead optimization, achieving state-of-the-art performance in chemical plausibility and binding affinity prediction.

3D Structure-based Generative Small Molecule Drug Design: Are We There Yet?

This paper introduces two metrics to evaluate the chemical plausibility of molecules generated by structure-based generative algorithms and compares them to virtual screening hits for three protein targets.

SLOGEN: A Structure-based Lead Optimization Model Unifying Fragment Generation and Screening

Slogen unifies fragment generation and screening for lead optimization, improving chemical diversity and synthetic tractability while matching state-of-the-art performance.

Fusobacterium nucleatum produces previously unappreciated AHR-activating metabolites and promotes CRC cell proliferation via the AHR-TERT axis

This study identifies Fusobacterium nucleatum-produced indole metabolites that activate the AHR-TERT axis, promoting CRC cell proliferation and revealing FTIN as a potential biomarker.

Evolutionary Tree in Chemical Space of Natural Products

The paper evaluates if evolutionary proximity correlates with chemical similarity in natural products across multiple taxa using deep learning embeddings.

Molecular Basis of PAC1R Allosteric Modulation with Lipids in Membranes

This study identifies lipid binding sites on PAC1R using simulations, showing cholesterol and GM3 interact with the receptor in multiple domains, potentially influencing its activation.