This paper introduces DTi-pKa, a graph neural network that predicts pKa values with thermodynamic consistency and achieves high accuracy on benchmark datasets.

bioRxiv · Oct 6, 2026 · PreprintBioinformatics

Beyond RMSD: Deterministic Dlow Matching for Ultrafast and Physically Validated Transition-State Generation

hehua Li, Saisai Yu, Qiuyin Zhu, Zhenming Liu, Yongsheng Li, Shengwen Yang

This paper introduces Kairos-GFM, a model that generates transition states with high physical fidelity and low RMSD, outperforming diffusion-based methods but underperforming transport-based ones.

bioRxiv · Oct 1, 2026 · PreprintBiochemistry

Pair-complete structural gating enables Pareto optimization of de novo cyclic peptides with preserved binding geometry and improved developability

Yuanchao Hou, Yixin Zhang, Zhuoyan Liu, 彭新发, Ziye Liu, Miyesier Yusupujiang, You Xu, Jiaxin Hu, Zhenming Liu, Shengwen Yang

This paper introduces RF-PepTune, a method that improves cyclic peptide design by preserving binding geometry while optimizing multiple properties, achieving higher success rates in structural and functional criteria.

bioRxiv · Sep 29, 2026 · PreprintBioinformatics

A closed-loop reinforcement learning framework for rapid compound directed optimization

Han Wang, Dehua Lu, Weiping Lyu, Siyu Xiu, Cheng Shi, Xiaonan Zhou, Bin Xi, Wei Feng, Yang Xiao, Yanming Chen, Haixian Zhang, Qi Li, Bo Huang, Zhenming Liu

This paper introduces RCDO, a closed-loop reinforcement learning framework that accelerates compound optimization by integrating experimental feedback with generative models, achieving significant improvements in drug candidate properties.

bioRxiv · Aug 27, 2026 · PreprintBioinformatics

Assessing Conformation Validity and Rationality of Deep Learning-Generated 3D Molecules

Fan Fan, Bin Xi, Xianghu Meng, Han Wang, Bowen Zhang, Qingbo Xu, Wei Feng, Wenfeng Gao, Xiaoman Wang, Yuji Wang, Hongbo Zhang, Feng Zhou, Zhenming Liu, Wenbiao Zhou, Bo Huang

This paper introduces a two-stage method to evaluate 3D molecule conformations generated by AI, achieving high speed and quantum mechanical accuracy.

Nature Communications · Feb 7, 2026Peer-reviewedBioinformatics

A unified language model bridging de novo and fragment-based 3D molecule design delivers potent CBL-B inhibitors for cancer treatment

Han Wang, Guocheng Sun, Bowen Zhang, Yang Wang, Bin Xi, Minjian Yang, Chuanyu Liu, Yuyang Ge, Fan Fan, Wei Feng, Yanhao Zhu, Yang Xiao, Yuji Wang, Zhenming Liu, Daohua Jiang, , Bo Huang

bioRxiv · Nov 13, 2025 · PreprintBioinformatics

Pocket-based molecule generation with an SE(3)-equivariant language model leads to a potent and selective HPK1 inhibitor with in vivo efficacy

Bin Xi, Han Wang, Guocheng Sun, Bowen Zhang, Ruihan Mao, Yuyang Ge, Yang Wang, Jiangtao Zhang, Yiting Pan, Feng Zhou, Yuji Wang, Zhenming Liu, Daohua Jiang, Huting Wang, Wenbiao Zhou, Bo Huang

bioRxiv · Sep 23, 2025 · PreprintBioinformatics

f-BGM enables fungi-specific genome mining in high accuracy and interpretability

Yiran Zhou, Dong Liu, Zequan Huang, Xingsi Xie, Wenhan Lin, Liangren Zhang, Aili Fan, Zhenming Liu

f-BGM is a deep learning framework for fungal genome mining that outperforms existing methods in BGC detection and provides interpretability for enzyme identification and domain partnerships.

bioRxiv · Jul 8, 2025 · PreprintBioinformatics

Intratumoral microbiota-derived S1P sensitizes the combination therapy of Capecitabine and Nivolumab

Chenyi Dai, Sen-Tao Pan, Tian-Tian Qi, Huiling Shang, Ri‐hua Xie, Hao Liu, Zhenming Liu, Cui Yi-min, Yu-Hang Zhang

This study identifies how intratumoral microbiota-derived S1P enhances the efficacy of Capecitabine-Nivolumab in CRC by modulating PD-L1 expression and T cell dynamics.

bioRxiv · Jun 5, 2025 · PreprintMicrobiology