计算机科学

计算机科学 ›› 2025, Vol. 52 ›› Issue (9): 71-79.doi: 10.11896/jsjkx.250100116

• 智能医学工程 • 上一篇    下一篇

基于药物子结构与蛋白质三维图信息的化合物-蛋白质相互作用预测

李亚茹1, 王倩倩1, 车超1,2, 朱德恒1   

  1. 1 大连大学先进设计与智能计算省部共建教育部重点实验室 辽宁 大连 116622
    2 大连大学软件工程学院 辽宁 大连 116622
  • 收稿日期:2025-01-17 修回日期:2025-03-30 出版日期:2025-09-15 发布日期:2025-09-11
  • 通讯作者: 朱德恒(375651307@qq.com)
  • 作者简介:(lyr1365197742@163.com)
  • 基金资助:
    国家自然科学基金(62076045);高等学校学科创新引智基地(D23006);国家外国专家项目(D20240244);大连大学学科交叉项目(DLUXK-2024-QN-010,DLUXK-2024-QN-006)

Graph-based Compound-Protein Interaction Prediction with Drug Substructures and Protein 3D Information

LI Yaru1, WANG Qianqian1, CHE Chao1,2, ZHU Deheng1   

  1. 1 Key Laboratory of Advanced Design and Intelligent Computing Ministry of Education,Dalian University,Dalian,Liaoning 116622,China
    2 College of Software Engineering,Dalian University,Dalian,Liaoning 116622,China
  • Received:2025-01-17 Revised:2025-03-30 Online:2025-09-15 Published:2025-09-11
  • About author:LI Yaru,born in 1999,postgraduate.Her main research interests include drug-target interactions and drug-drug interactions.
    ZHU Deheng,born in 1987,Ph.D.His main research interests include the study and development of robotic systems in specific scenarios,and drug repurposing.
  • Supported by:
    National Natural Science Foundation of China(62076045),the 111 Project(D23006),National Foreign Expert Project of China(D20240244) and Interdisciplinary Project of Dalian University(DLUXK-2024-QN-010,DLUXK-2024-QN-006).

PDF (PC)

11

摘要： 药物通过与蛋白质相互作用来抑制或激活特定蛋白质的功能,从而发挥治疗作用。近年来,深度学习方法在化合物蛋白质相互作用预测中取得显著进展。然而,现有的大多数研究仍然侧重于从药物和蛋白质的整体特征进行提取,对于药物和靶点的信息探索不足,忽视了蛋白质结构的三维空间信息以及药物关键子结构在化合物蛋白质相互作用预测中的作用。针对这一问题,提出了一种新的模型,其结合药物的官能团、整体结构图以及蛋白质的序列和三维空间图信息,将图神经网络和注意力机制融合,进行高效的特征学习与预测。在Human和C.elegans公开数据集上的实验结果表明,所提模型在CPI预测中表现出色,在ACC,AUROC和AUPR指标上有1%以上的提升,在非平衡数据集上表现出稳定的性能优势。

关键词: 化合物-蛋白质相互作用预测, 药物子结构, 蛋白质结构预测, 图神经网络, 深度学习

Abstract: Drugs exert therapeutic effects by interacting with proteins to inhibit or activate the functions of specific proteins.In recent years,deep learning methods have made significant progress in predicting compound-protein interactions.However,most existing studies still focus on extracting overall features from drugs and proteins,neglecting the exploration of drug target information,the three-dimensional spatial information of protein structures,and the role of drug key substructures in predicting compound-protein interactions.To address this issue,a new model is proposed,which combines the functional groups of drugs,the overall structural graphs of drugs,and the sequence and three-dimensional spatial graph information of proteins.By utilizing the fusion of graph neural networks and attention mechanisms,efficient feature learning and prediction are conducted.The experimental results on the public datasets of Human and C.elegans show that the proposed model performs excellently in CPI prediction,with an improvement of more than 1% in ACC,AUROC,and AUPR indicators,and demonstrates a stable performance advantage on imbalanced datasets.

Key words: Compound-protein interaction, Drug substructure, Protein structure prediction, Graph neural networks, Deep learning

中图分类号: 

  • TP391
[1]YANG Z,ZHONG W,ZHAO L,et al.MGraphDTA:deep multiscale graph neural network for explainable drug-target binding affinity prediction[J].Chemical science,2022,13(3):816-833.
[2]DU L,GENG C,ZENG Q,et al.Dockey:a modern integrated tool for large-scale molecular docking and virtual screening[J].Briefings in Bioinformatics,2023,24(2):47.
[3]KRIZHEVSKY A,SUTSKEVER I,HINTON G E.Imagenetclassification with deep convolutional neural networks[J].Communications of the ACM,2017,60(6):84-90.
[4]FENG B,ZHANG Y,ZHENG H,et al.Defusedti:Interpretable drug target interaction prediction model with dual-branch encoder and multiview fusion[J].Future Generation Computer Systems,2024,161:239-247.
[5]WU L,HUANG Y,TAN C,et al.Psc-cpi:Multi-scale protein sequence-structure contrasting for efficient and generalizable compound-protein interaction prediction[C]//Proceedings of the AAAI Conference on Artificial Intelligence.2024:310-319.
[6]HUA Y,FENG Z,SONG X,et al.MMDG-DTI:Drug-target interaction prediction via multimodal feature fusion and domain generalization[J].Pattern Recognition,2025,157:110887.
[7]CHEN L,TAN X,WANG D,et al.TransformerCPI:improving compound-protein interaction prediction by sequence-based deep learning with self-attention mechanism and label reversal experiments[J].Bioinformatics,2020,36(16):4406-4414.
[8]DING Y J,GUO F,TIWARI P,et al.Identification of drug-side effect association via multi-view semi-supervised sparse model[J].IEEE Transactions on Artificial Intelligence,2024,5(5):2151-2162.
[9]JIANG M,LI Z,ZHANG S,et al.Drug-target affinity predictionusing graph neural network and contact maps[J].RSCAdvan-ces,2020,10(35):20701-20712.
[10]ZHANG Z,LIU Q,WANG H,et al.Motif-based graph self-supervised learning for molecular property prediction[J].Advances in Neural Information Processing Systems,2021,34:15870-15882.
[11]DEGEN J,WEGSCHEID-GERLACH C,ZALIANI A,et al.On the art of compiling and using ‘drug-like’ chemical fragment spaces[J].ChemMedChem,2008,3(10):1503.
[12]VELIČKOVIĆ P,CUCURULL G,CASANOVA A,et al.Graph attention networks[C]//ICLR 2018 Conference Blind Submission.2018.
[13]KIPF T N,WELLING M.Semi-supervised classification withgraph convolutional networks[C]//Proceedings of the International Conference on Learning Representations.2017:1-14.
[14]XU K,HU W,LESKOVEC J,et al.How Powerful are Graph Neural Networks?[C]//Proceedings of the 7th International Conference on Learning Representations.2019.
[15]LEE J,LEE Y,KIM J,et al.Set transformer:A framework for attention-based permutation-invariant neural networks[C]//Proceedings of the International Conference on Machine Lear-ning.PMLR,2019:3744-3753.
[16]MAK K,WONG Y,PICHIKA M.Artificial intelligence in drug discovery and development[J].Drug Discovery and Evaluation:Safety and Pharmacokinetic Assays,2024(10):1461-98.
[17]ÖZTÜRK H,ÖZGÜR A,OZKIRIMLI E.DeepDTA:deep drug-target binding affinity prediction[J].Bioinformatics,2018,34(17):821-829.
[18]SHIN B,PARK S,KANG K,et al.Self-attention based molecule representation for predicting drug-target interaction[C]//Proceedings of the Machine Learning for Healthcare Conference.PMLR,2023:230-248.
[19]VASWANI A,SHAZEER N,PARMAR N,et al.Attention isall you need[C]//Proceedings of the 31st International Confe-rence on Neural Information Processing Systems.2017:6000-6010.
[20]QIU Z,JIAO Q,WANG Y,et al.RzMLP-DTA:gMLP network with ReZero for sequence-based drug-target affinity prediction[C]//Proceedings of the 2021 IEEE International Conference on Bioinformatics and Biomedicine(BIBM).2021:308-313.
[21]KAO P,KAO S,HUANG N,et al.Toward drug-target interaction prediction via ensemble modeling and transfer learning[C]//Proceedings of the 2021 IEEE International Conference on Bioinformatics and Biomedicine.2021:2384-2391.
[22]YUAN W,CHEN G,CHEN C.FusionDTA:attention-basedfeature polymerizer and knowledge distillation for drug-target binding affinity prediction[J].Briefings in Bioinformatics,2022,23(1):506.
[23]SIAMI-NAMINI S,TAVAKOLI N,NAMIN A S.The perfor-mance of LSTM and BiLSTM in forecasting time series[C]//Proceedings of the 2019 IEEE International Conference on Big Data.2019:3285-3292.
[24]NGUYEN T,LE H,QUINN T P,et al.GraphDTA:predicting drug-target binding affinity with graph neural networks[J].Bioinformatics,2021,37(8):1140-1147.
[25]LI F,ZHANG Z,GUAN J,et al.Effective drug-target interaction prediction with mutual interaction neural network[J].Bioinformatics,2022,38(14):3582-3589.
[26]WANG P,ZHENG S,JIANG Y,et al.Structure-aware multimodal deep learning for drug-protein interaction prediction[J].Journal of chemical information and modeling,2022,62(5):1308-1317.
[27]BENTO A,HERSEY A,FÉLIX E,et al.An open source chemical structure curation pipeline using RDKit[J].Journal of Cheminformatics,2020,12(1):1-16.
[28]VARADI M,ANYANGO S,DESHPANDE M,et al.AlphaFold Protein Structure Database:massively expanding the structural coverage of protein-sequence space with high-accuracy models[J].Nucleic Acids Research,2022,50(D1):439-444.
[29]LIU H,SUN J,GUAN J,et al.Improving compound-protein interaction prediction by building up highly credible negative samples[J].Bioinformatics,2015,31(12):221-229.
[30]SUTHAHARAN S.Support vector machine[J].MachineLearning Models and Algorithms for Big Data Classification:Thinking with Examples for Effective Learning,2016(11):207-235.
[31]RIGATTI S.Random forest[J].Journal of Insurance Medicine,2017,47(1):31-39.
[32]HUANG K,XIAO C,GLASS L M,et al.MolTrans:molecular interaction transformer for drug-target interaction prediction[J].Bioinformatics,2021,37(6):830-836.
[33]LI M,LU Z,WU Y,et al.BACPI:a bi-directional attention neural network for compound-protein interaction and binding affinity prediction[J].Bioinformatics,2022,38(7):1995-2002.
[34]PENG L,LIU X,YANG L,et al.BINDTI:a bi-directional intention network for drug-target interaction identification based on attention mechanisms[J].IEEE Journal of Biomedical and Health Informatics,2024,28:1-11.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市渝北区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发