Predicting Protein-Ligand Binding Affinity via Joint Global-Local Interaction Modeling

被引:1
作者
Zhang, Yang [1 ,2 ]
Zhou, Gengmo [1 ,2 ]
Wei, Zhewei [1 ]
Xu, Hongteng [1 ]
机构
[1] Renmin Univ China, Beijing, Peoples R China
[2] DP Technol, Beijing, Peoples R China
来源
2022 IEEE INTERNATIONAL CONFERENCE ON DATA MINING (ICDM) | 2022年
关键词
Protein-ligand binding affinity; graph neural networks; long-short interactions; drug discovery; DISCOVERY;
D O I
10.1109/ICDM54844.2022.00175
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
The prediction of protein-ligand binding affinity is of great significance for discovering lead compounds in drug research. Facing this challenging task, most existing prediction methods rely on the topological and/or spatial structure of molecules and the local interactions while ignoring the multilevel inter-molecular interactions between proteins and ligands, which often lead to sub-optimal performance. To solve this issue, we propose a novel global-local interaction (GLI) framework to predict protein-ligand binding affinity. In particular, our GLI framework considers the inter-molecular interactions between proteins and ligands, which involve not only the high-energy short-range interactions between closed atoms but also the lowenergy long-range interactions between non-bonded atoms. For each pair of protein and ligand, our GLI embeds the long-range interactions globally and aggregates local short-range interactions, respectively. Such a joint global-local interaction modeling strategy helps to improve prediction accuracy, and the whole framework is compatible with various neural network-based modules. Experiments demonstrate that our GLI framework outperforms state-of-the-art methods with simple neural network architectures and moderate computational costs.
引用
收藏
页码:1323 / 1328
页数:6
相关论文
共 30 条
[1]   A machine learning approach to predicting protein-ligand binding affinity with applications to molecular docking [J].
Ballester, Pedro J. ;
Mitchell, John B. O. .
BIOINFORMATICS, 2010, 26 (09) :1169-1175
[2]  
Chen Ming, 2020, P MACHINE LEARNING R, V119
[3]  
Cho H., 2020, INTERACTIONNET MODEL
[4]  
csardock, US
[5]  
Felner M. K., 2009, THEORY LONG RANGE IN
[6]  
Gasteiger J., 2020, INT C LEARN REPR ICL
[7]  
github, About us
[8]  
Greenspan D, 1980, LONG SHORT RANGE FOR
[9]   Performance Evaluation of 2D Fingerprint and 3D Shape Similarity Methods in Virtual Screening [J].
Hu, Guoping ;
Kuang, Guanglin ;
Xiao, Wen ;
Li, Weihua ;
Liu, Guixia ;
Tang, Yun .
JOURNAL OF CHEMICAL INFORMATION AND MODELING, 2012, 52 (05) :1103-1113
[10]   Structural bioinformatic approaches to the discovery of new antimycobacterial drugs [J].
Kantardjieff, K ;
Rupp, B .
CURRENT PHARMACEUTICAL DESIGN, 2004, 10 (26) :3195-3211