中国科学院机构知识库网格
Chinese Academy of Sciences Institutional Repositories Grid
首页 机构 成果 学者
  
  1. CAS IR Grid
  2. 上海药物研究所
  3. 中国科学院上海药物研究所
DMPE: A Dual-Branch Molecular Property Encapsulation Framework With Kolmogorov-Arnold Networks

文献类型：期刊论文

作者Duo, Lihui1; Tang, Bencan1; Hirst, Jonathan D.2; Xie, Hua3; Ren, Jianfeng4
刊名JOURNAL OF COMPUTATIONAL CHEMISTRY
出版日期2025-11-15
卷号46期号:30页码:16
关键词cancer drug discovery graph attention networks molecular property prediction molecular representations
ISSN号0192-8651
DOI10.1002/jcc.70273
通讯作者Tang, Bencan(bencan.tang@nottingham.edu.cn) ; Hirst, Jonathan D.(jonathan.hirst@nottingham.ac.uk) ; Ren, Jianfeng(jianfeng.ren@nottingham.edu.cn)
英文摘要Accurate molecular representation learning is critical for accelerating property prediction in drug discovery. Here, we propose the Dual-branch Molecular Property Encapsulation (DMPE) framework. Its core component, the Refined Interactive Graph Attention Framework (RIGAF), captures both intramolecular topology and intermolecular structural similarity within the broader chemical space. Subsequently, a Kolmogorov-Arnold Network (KAN)-based Embedding and Fusion (KAEF) module uses spline-based transformations to integrate these graph features with complementary molecular fingerprints, enhancing model interpretability while maintaining strong generalizability. DMPE performs competitively, with superior ROC-AUC scores on scaffold-based BBBP (0.927) and multi-label SIDER (0.691) benchmarks, alongside promising accuracy across nine of 14 breast cancer cell lines. Key components of our architecture are validated through ablation studies. Model reliability is bolstered by Monte Carlo dropout for uncertainty estimation, which also serves as an ensemble strategy to enhance accuracy. The framework was applied to the identification of potential hematopoietic progenitor kinase 1 (HPK1) inhibitors for hepatocellular carcinoma.
资助项目Ningbo Science and Technology Bureau[2020Z092] ; Ningbo Natural Science Foundation Programm[2022J171] ; Zhejiang Provincial Department of Science and Technology[2020E10018] ; Department of Science, Innovation and Technology (DSIT) and the Royal Academy of Engineering[CiET2021_17] ; National Natural Science Foundation of China[22171153] ; National Key R&D Program of Intergovernmental Key Projects[2018YFE0101700] ; Ningbo Municipal Key Laboratory on Clean Energy Conversion Technologies[2014A22010]
WOS研究方向Chemistry
语种英语
WOS记录号WOS:001617858800002
出版者WILEY
源URL[http://119.78.100.183/handle/2S10ELR8/322001]  
专题中国科学院上海药物研究所
通讯作者Tang, Bencan; Hirst, Jonathan D.; Ren, Jianfeng
作者单位1.Univ Nottingham Ningbo China, Dept Chem & Environm Engn, Ningbo, Peoples R China
2.Univ Nottingham, Sch Chem, Nottingham, England
3.Shanghai Inst Mat Med, Shanghai, Peoples R China
4.Univ Nottingham Ningbo China, Sch Comp Sci, Ningbo, Peoples R China
推荐引用方式
GB/T 7714		 Duo, Lihui,Tang, Bencan,Hirst, Jonathan D.,et al. DMPE: A Dual-Branch Molecular Property Encapsulation Framework With Kolmogorov-Arnold Networks[J]. JOURNAL OF COMPUTATIONAL CHEMISTRY,2025,46(30):16.
APA Duo, Lihui,Tang, Bencan,Hirst, Jonathan D.,Xie, Hua,&Ren, Jianfeng.(2025).DMPE: A Dual-Branch Molecular Property Encapsulation Framework With Kolmogorov-Arnold Networks.JOURNAL OF COMPUTATIONAL CHEMISTRY,46(30),16.
MLA Duo, Lihui,et al."DMPE: A Dual-Branch Molecular Property Encapsulation Framework With Kolmogorov-Arnold Networks".JOURNAL OF COMPUTATIONAL CHEMISTRY 46.30(2025):16.

入库方式： OAI收割

来源：上海药物研究所

浏览0
下载0
收藏0
其他版本

除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。