Changping Laboratory

Flow-matching generative model for de novo protein binder backbone design, built on a Pairformer architecture with in silico interface maturation.

Mixture-of-Experts generative model turning DNA sequence plus cell-type ATAC-seq into unified epigenomic, transcriptomic, and 3D chromatin profiles.

Protein dynamics model that samples conformational ensembles autoregressively at slow and fast timescales, generalizing zero-shot to unseen proteins.

Protein function prediction via compressed in-context learning on a sequence-structure language model, cutting 751-token demonstrations to under 16.

End-to-end protein structure alignment from 3D coordinates, matching TM-align quality at LDDT 0.56 and beating Foldseek on SCOPe40 search.

Protein language model that tokenizes sequence, backbone structure, and text into one vocabulary for function prediction, design, and fold editing.

Protein question-answering model that fuses sequence and structure into an LLM prompt as virtual tokens, answering free-form questions about function.

Protein structure tokenizer that encodes all-atom folds as artificial amino acids and decodes them back to coordinates at TM-score above 0.96.