activity
20242026
collaborators

12 papers

q-bio.BM2026

SF-Cluster: Frustration-Guided MSA Subsampling for Alternative Protein Conformation Recovery

Hanqun Cao, Zijun Gao, Chunbin Gu +3

Deep-learning structure predictors are sensitive to their multiple sequence alignment (MSA) input, making MSA subsampling a practical route to recovering alternative conformations.…

cs.CE2026

GeoCycler: Reward-Aligned 3D Diffusion for Constraint-Conditioned Cyclic Peptide Design

Jingjie Zhang, Hanqun Cao, Haosen Shi +10

Cyclic peptides are attractive therapeutic modalities because their closed-ring topology can improve stability and target specificity. However, de novo cyclic peptide design remain…

cs.CV2026

Does Engram Do Memory Retrieval in Autoregressive Image Generation?

Jinghao Wang, Qiyuan He, Chunbin Gu +1

The Engram module -- a hash-keyed, O(1) associative memory injected into Transformer layers -- was recently shown to improve large language model pretraining, with the appealing in…

q-bio.BM2026

CA-DEL: An Open Multi-Target, Multi-Modal Benchmark for Learning from DNA-Encoded Library Screens

Mutian He, Hanqun Cao, Cheng Tan +4

The success of machine learning in drug discovery hinges on learning the relationship between a chemical structure and its biological activity. While DNA-Encoded Library (DEL) tech…

q-bio.BM2026

Bi-TEAM: A Unified Cross-Scale Representation Learning Framework for Chemically Modified Biomolecules

Chunbin Gu, Zijun Gao, Mutian He +8

Representation learning for protein biochemical space faces a difficult trade-off: protein language models excel at capturing long-range biological semantics but often miss fine-gr…

q-bio.BM2025

CONFIDE: Hallucination Assessment for Reliable Biomolecular Structure Prediction and Design

Zijun Gao, Mutian He, Shijia Sun +8

Reliable evaluation of protein structure predictions remains challenging, as metrics like pLDDT capture energetic stability but often miss subtle errors such as atomic clashes or c…