AI for Drug Discovery, Science for Better Lives!
Hey there! I’m currently a third-year Ph.D. student in Computer Science and Technology at Tongji University, supervised by Prof. Guang Chen, the head of the Generalist Embodied AI Lab. My research focuses on AI4Science, with specific interests in (1) drug discovery, (2) generative models, and (3) reinforcement learning.
📣 News
🎉 2025.08: Our work (SWITCH) on spatial multi-omics is accepted by Nature Computational Science (NCS)!
🎉 2025.02: Our work (RCP-Bench) on collaborative perception is accepted by CVPR 2025!
🎉 2024.12: We won the championship (1/226) in The Second Global AI Drug Development Algorithm Competition!
🎉 2024.11: Our work (AMG) on 3D structure-based molecular generation is accepted by Briefings in Bioinformatics!
🎉 2022.09: Our work on medical image segmentation is accepted by IEEE Journal of Biomedical and Health Informatics (IF: 7.7)!
🎉 2022.07: Our work (Molormer) on drug-drug interaction prediction is accepted by Briefings in Bioinformatics (IF: 13.994)!
🎉 2021.12: Our work on drug-drug interaction prediction is accepted by Briefings in Bioinformatics (IF: 13.994)!
📝 Publications
Xudong Zhang, Jing Hou, Sanqing Qu, Fan Lu, Zhixin Tian, Alois Knoll, Guang Chen*, Shaorong Gao*, Yanping Zhang*
Briefings in Bioinformatics (BIB), 2025
In this paper, we propose AMG, a framework that leverages deep reinforcementlearning as a pocket–ligand interaction agent to gradually steer fragment-based 3D molecular generation targeting protein pockets.AMG is trained using a two-stage strategy to capture interaction features and explicitly optimize the IA. The framework also introducesa pair of separate encoders for pockets and ligands, coupled with a dedicated pre-training strategy. This enables AMG to enhance itsgeneralization ability by leveraging a vast repository of undocked pockets and molecules, thus mitigating the constraints posed by the limited quantity and quality of available datasets.
Xudong Zhang, Gan Wang, Xiangyu Meng, Shuang Wang, Ying Zhang, Alfonso Rodriguez-Paton, Jianmin Wang*, Xun Wang*
Briefings in Bioinformatics (BIB), 2022
In this paper, we propose Molormer, a method based on a lightweight attention mechanism for DDIs prediction. Molormer takes the two-dimension (2D) structures of drugs as input and encodes the molecular graph with spatial information.
Tao Song*, Xudong Zhang, Mao Ding*, Alfonso Rodriguez-Paton, Shudong Wang, Gan Wang
Methods, 2022
Predicting drug-target interactions (DTIs) is essential for both drug discovery and drug repositioning. We generate global structural similarity feature based on similarity theory and generate local chemical sub-structure semantic feature using transformer for both drug and protein.
Xun Wang, Xudong Zhang, Gan Wang, Ying Zhang, Xin Shi, Huanhuan Dai, Min Liu, Zixuan Wang, Xiangyu Meng
IEEE Journal of Biomedical and Health Informatics (JBHI), 2022
We introduce TransFusionNet, which consists of a semantic feature extraction module, a local spatial feature extraction module, an edge feature extraction module, and a multi-scale feature fusion module to achieve fine-grained segmentation of liver tumors and vessels.
Shanchen Pang, Ying Zhang, Tao Song*, Xudong Zhang, Xun Wang, Alfonso Rodriguez-Patón
Briefings in Bioinformatics (BIB), 2021
The properties of the drug may be altered by the combination,which may cause unexpected drug–drug interactions (DDIs). In this work, we propose a novel attention-mechanism-based multidimensional feature encoder for DDIs prediction, namely attention-based multidimensional feature encoder.
🏆 Honors and Awards
- 2023.04: Outstanding Graduate of Shandong Province
- 2022.12: National scholarship for Postgraduates
🌏 Academic Service
Reviewer: Briefings in Bioinformatics, JBHI, Journal of Cheminformatics, BioData Mining, BMC Bioinformatics.