Prediction of drug-target interactions based on sequence and multi-view networks

doi:10.11772/j.issn.1001-9081.2024111664

Journal of Computer Applications ›› 2025, Vol. 45 ›› Issue (11): 3658-3665.DOI: 10.11772/j.issn.1001-9081.2024111664

• Advanced computing • Previous Articles

Prediction of drug-target interactions based on sequence and multi-view networks

Jiahao ZHANG, Qi WANG, Mingming LIU(), Xiaofeng WANG, Biao HUANG, Pan LIU, Zhi YE

College of Software，Nankai University，Tianjin 300350，China

Received:2024-11-27 Revised:2025-01-07 Accepted:2025-01-24 Online:2025-02-14 Published:2025-11-10
Contact: Mingming LIU
About author:ZHANG Jiahao， born in 1998， M. S. candidate. His research interests include machine learning， biometric identification.
WANG Qi， born in 1999， M. S. candidate. His research interests include biomedicine， link prediction.
WANG Xiaofeng， born in 1999， M. S. candidate. His research interests include cancer type prediction， deep learning.
HUANG Biao， born in 2001， M. S. candidate. His research interests include machine learning， bioinformatics.
LIU Pan， born in 2001， M. S. candidate. Her research interests include machine learning， biometric identification.
YE Zhi， born in 2000， M. S. candidate. His research interests include machine learning， biometric identification.
Supported by:
Natural Science Foundation of Tianjin(22JCYBJC01020)

基于序列和多视角网络的药物-靶标相互作用预测

张家豪, 王琪, 刘明铭(), 王晓峰, 黄彪, 刘盼, 叶至

南开大学软件学院，天津 300350

通讯作者: 刘明铭
作者简介:张家豪（1998—），男，河南商丘人，硕士研究生，主要研究方向：机器学习、生物特征识别
王琪（1999—），男，甘肃平凉人，硕士研究生，主要研究方向：生物医学、链接预测
王晓峰（1999—），男，山西临汾人，硕士研究生，主要研究方向：癌症类型预测、深度学习
黄彪（2001—），男，湖北天门人，硕士研究生，主要研究方向：机器学习、生物信息
刘盼（2001—），女，湖南邵阳人，硕士研究生，主要研究方向：机器学习、生物特征识别
叶至（2000—），男，广西贺州人，硕士研究生，主要研究方向：机器学习、生物特征识别。
基金资助:
天津自然科学基金资助项目(22JCYBJC01020)

Abstract

Abstract:

Identifying Drug-Target Interactions （DTI） is a crucial step in drug repurposing and novel drug discovery. Currently， many sequence-based computational methods have been widely used for DTI prediction. However， previous sequence-based studies typically focus solely on the sequence itself for feature extraction， neglecting heterogeneous information networks such as drug-drug interaction networks and drug-target interaction networks. Therefore， a novel method for DTI prediction based on sequence and multi-view networks was proposed， namely SMN-DTI （prediction of Drug-Target Interactions based on Sequence and Multi-view Networks）. The Variational AutoEncoder （VAE） was used to learn the embedding matrices of drug SMILES （Simplified Molecular-Input Line-Entry System） strings and target amino acid sequences in this method. Subsequently， a Heterogeneous graph Attention Network （HAN） with two-level attention mechanism was used to aggregate information from different neighbors of drugs or targets in the networks from both node and semantic perspectives， obtaining the final embeddings. Two benchmark datasets widely used for DTI prediction， Hetero-seq-A and Hetero-seq-B， were used to evaluate SMN-DTI and the baseline methods. The results show that SMN-DTI achieves the best performance in Area Under the receiver operating Characteristic curve （AUC） and the Area Under the Precision-Recall curve （AUPR） under three different positive-and-negative sample ratios. It can be seen that SMN-DTI outperforms current mainstream advanced prediction methods.

Key words: Drug-Target Interaction (DTI) prediction, Variational AutoEncoder (VAE), Heterogeneous graph Attention Network (HAN), multi-view network, attention mechanism

摘要：

识别药物-靶标相互作用（DTI）是药物再利用和创新药物发现中不可或缺的关键步骤，目前已经有许多基于序列的计算方法被广泛应用于DTI预测；然而，在以往的基于序列的研究中，特征提取通常只关注序列本身，忽视了异构信息网络，如药物-药物相互作用网络、药物-靶标相互作用网络等。因此，提出一种基于序列和多视角网络进行DTI预测的新方法SMN-DTI（prediction of Drug-Target Interactions based on Sequence and Multi-view Networks）。该方法使用变分自编码器（VAE）学习药物SMILES（Simplified Molecular-Input Line-Entry System）字符串和靶标氨基酸序列的嵌入矩阵；随后，利用具有两级注意力机制的异构图注意力网络（HAN）从节点和语义2个视角的网络中聚集来自药物或靶标的不同邻居的信息，并得到最终的嵌入。在2个广泛用于DTI预测的基准数据集Hetero-seq-A和Hetero-seq-B上对SMN-DTI和基准方法进行评估的结果表明，在3种不同正负样本比例下SMN-DTI均取得了最优的特征曲线下面积（AUC）和精确召回曲线下面积（AUPR）。可见，SMN-DTI比目前主流的先进预测方法具有更好的性能。

关键词: 药物-靶标相互作用预测, 变分自编码器, 异构图注意力网络, 多视角网络, 注意力机制

CLC Number:

TP183

Jiahao ZHANG, Qi WANG, Mingming LIU, Xiaofeng WANG, Biao HUANG, Pan LIU, Zhi YE. Prediction of drug-target interactions based on sequence and multi-view networks[J]. Journal of Computer Applications, 2025, 45(11): 3658-3665.

张家豪, 王琪, 刘明铭, 王晓峰, 黄彪, 刘盼, 叶至. 基于序列和多视角网络的药物-靶标相互作用预测[J]. 《计算机应用》唯一官方网站, 2025, 45(11): 3658-3665.

Figures/Tables 10

Fig. 1 Structure of SMN-DTI

Fig. 2 Framework of VAE module

Fig. 3 Meta-path and neighbor examples of heterogeneous information network of drug-target interactions

Tab. 1 Statistics of experimental datasets

数据集

药物数

靶标数

药物-靶标

相互作用数

药物-药物

相互作用数

药物-药物结构

相似度矩阵

靶标-靶标相互

作用数

靶标-靶标序列

相似度矩阵

Tab. 2 Comparison of AUC and AUPR of various methods on Hetero-Seq-A dataset under different positive-and-negative sample ratios

方法	正负样本比例为1∶1		正负样本比例为1∶5		正负样本比例为1∶10
方法	AUC	AUPR	AUC	AUPR	AUC	AUPR
DeepDTA	0.912 5	0.903 9	0.932 6	0.816 7	0.925 0	0.754 5
DeepConv-DTI	0.906 3	0.905 0	0.924 2	0.791 3	0.924 0	0.712 8
GraphDTA	0.906 0	0.882 9	0.920 4	0.801 8	0.914 6	0.753 1
Co-VAE	0.907 5	0.907 8	0.931 2	0.822 5	0.932 8	0.773 9
HyperAttentionDTI	0.899 0	0.899 0	0.923 8	0.786 9	0.929 0	0.728 5
MFR-DTA	0.904 7	0.901 2	0.923 0	0.819 3	0.915 7	0.745 8
IMAEN	0.899 9	0.898 4	0.926 0	0.805 6	0.927 5	0.752 3
SMN-DTI（CNN/CNN）	0.889 9	0.900 2	0.924 1	0.789 8	0.924 9	0.718 5
SMN-DTI（CNN/VAE）	0.905 9	0.924 2	0.926 7	0.826 5	0.933 1	0.785 1
SMN-DTI（VAE/CNN）	0.903 7	0.916 7	0.932 8	0.836 2	0.935 6	0.795 6
SMN-DTI	0.916 5	0.934 6	0.936 0	0.861 1	0.940 6	0.813 1

Tab. 3 Comparison of AUC and AUPR of various methods on Hetero-Seq-B dataset under different positive-and-negative sample ratios

方法	正负样本比例为1∶1		正负样本比例为1∶5		正负样本比例为1∶10
方法	AUC	AUPR	AUC	AUPR	AUC	AUPR
DeepDTA	0.949 2	0.940 4	0.956 7	0.861 4	0.945 2	0.802 1
DeepConv-DTI	0.938 8	0.931 6	0.951 4	0.829 3	0.949 6	0.745 2
GraphDTA	0.930 1	0.914 0	0.940 5	0.832 6	0.933 4	0.764 6
Co-VAE	0.950 6	0.941 8	0.955 6	0.858 8	0.951 3	0.799 8
HyperAttentionDTI	0.952 9	0.949 6	0.956 2	0.887 7	0.949 9	0.802 5
MFR-DTA	0.949 8	0.937 2	0.954 3	0.864 9	0.947 2	0.801 7
IMAEN	0.951 1	0.948 0	0.958 0	0.874 5	0.948 7	0.796 9
SMN-DTI（CNN/CNN）	0.947 4	0.943 9	0.960 0	0.875 4	0.954 3	0.805 3
SMN-DTI（CNN/VAE）	0.954 3	0.953 5	0.962 8	0.887 2	0.959 3	0.826 2
SMN-DTI（VAE/CNN）	0.954 2	0.953 9	0.964 1	0.892 6	0.959 2	0.826 8
SMN-DTI	0.962 4	0.963 6	0.966 9	0.901 9	0.963 1	0.844 3

Tab. 4 Experimental results of different meta-paths

药物元路径	靶标元路径	AUC	AUPR
$ρ d 1$	$ρ t 1$	0.953 0	0.953 2
$ρ d 2$	$ρ t 2$	0.927 7	0.919 3
$ρ d 3$	$ρ t 3$	0.958 8	0.960 7
$ρ d 1, ρ d 2, ρ d 3$	$ρ t 1, ρ t 2, ρ t 3$	0.962 4	0.963 6

Tab. 4 Experimental results of different meta-paths

药物元路径	靶标元路径	AUC	AUPR
$ρ d 1$	$ρ t 1$	0.953 0	0.953 2
$ρ d 2$	$ρ t 2$	0.927 7	0.919 3
$ρ d 3$	$ρ t 3$	0.958 8	0.960 7
$ρ d 1, ρ d 2, ρ d 3$	$ρ t 1, ρ t 2, ρ t 3$	0.962 4	0.963 6

Fig. 4 Heat maps of real and predicted drug-target interaction networks

Fig. 5 Performance comparison under different numbers of attention heads

Fig. 6 Performance comparison of different final embedding dimensions

References 31

[1]	KAPETANOVIC I M. Computer-Aided Drug Discovery and Development （CADDD）： in silico-chemico-biological approach［J］. Chemico-Biological Interactions， 2008， 171（2）： 165-176.
[2]	HE Z， ZHANG J， SHI X H， et al. Predicting drug-target interaction networks based on functional groups and biological features［J］. PLoS ONE， 2010， 5（3）： No.e9603.
[3]	COBANOGLU M C， LIU C， HU F Z， et al. Predicting drug-target interactions using probabilistic matrix factorization［J］. Journal of Chemical Information and Modeling， 2013， 53（12）： 3399-3409.
[4]	ÖZTÜRK H， ÖZGÜR A， OZKIRIMLI E. DeepDTA： deep drug-target binding affinity prediction［J］. Bioinformatics， 2018， 34（17）： i821-i829.
[5]	TSUBAKI M， TOMII K， SESE J. Compound-protein interaction prediction with end-to-end learning of neural networks for graphs and sequences［J］. Bioinformatics， 2019， 35（2）： 309-318.
[6]	ZHAO Q， XIAO F， YANG M， et al. AttentionDTA： prediction of drug-target binding affinity using attention model［C］// Proceedings of the 2019 IEEE International Conference on Bioinformatics and Biomedicine. Piscataway： IEEE， 2019： 64-69.
[7]	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.
[8]	LI T， ZHAO X M， LI L. Co-VAE： drug-target binding affinity prediction by co-regularized variational autoencoders［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence， 2022， 44（12）： 8861-8873.
[9]	KINGMA D P， WELLING M. Auto-encoding variational Bayes［EB/OL］. ［2023-11-01］..
[10]	ZHANG P， WEI Z， CHE C， et al. DeepMGT-DTI： transformer network incorporating multilayer graph information for Drug-Target interaction prediction［J］. Computers in Biology and Medicine， 2022， 142： No.105214.
[11]	WANG X， JI H， SHI C， et al. Heterogeneous graph attention network［C］// Proceedings of the 2019 World Wide Web Conference. New York： ACM， 2019： 2022-2032.
[12]	LI J， WANG J R， LV H， et al. IMCHGAN： inductive matrix completion with heterogeneous graph attention networks for drug-target interactions prediction［J］. IEEE/ACM Transactions on Computational Biology and Bioinformatics， 2022， 19（2）： 655-665.
[13]	VELIČKOVIĆ P， CUCURULL G， CASANOVA A， et al. Graph attention networks［EB/OL］. ［2024-07-01］..
[14]	VASWANI A， SHAZEER N， PARMAR N， et al. Attention is all you need［C］// Proceedings of the 31st International Conference on Neural Information Processing Systems. Red Hook： Curran Associates Inc.， 2017： 6000-6010.
[15]	ZHAO Q， ZHAO H， ZHENG K， et al. HyperAttentionDTI： improving drug-protein interaction prediction by sequence-based deep learning with attention mechanism［J］. Bioinformatics， 2022， 38（3）： 655-662.
[16]	ZHANG Y， HU Y， LI H， et al. Drug-protein interaction prediction via variational autoencoders and attention mechanisms［J］. Frontiers in Genetics， 2022， 13： No.1032779.
[17]	HIGGINS I， MATTHEY L， PAL A， et al. β-VAE： learning basic visual concepts with a constrained variational framework［EB/OL］. ［2023-11-01］..
[18]	DAUPHIN Y N， FAN A， AULI M， et al. Language modeling with gated convolutional networks［C］// Proceedings of the 34th International Conference on Machine Learning. New York： JMLR.org， 2017： 933-941.
[19]	LUO Y， ZHAO X， ZHOU J， et al. A network integration approach for drug-target interaction prediction and computational drug repositioning from heterogeneous information［J］. Nature Communications， 2017， 8： No.573.
[20]	SUN Y， HAN J， YAN X， et al. PathSim： meta path-based top-k similarity search in heterogeneous information networks［J］. Proceedings of the VLDB Endowment， 2011， 4（11）： 992-1003.
[21]	LI M， CAI X， XU S， et al. Metapath-aggregated heterogeneous graph neural network for drug-target interaction prediction［J］. Briefings in Bioinformatics， 2023， 24（1）： No.bbac578.
[22]	MAAS A L， HANNUN A Y， NG A Y. Rectifier nonlinearities improve neural network acoustic models［EB/OL］. ［2023-12-03］..
[23]	CLEVERT D A， UNTERTHINER T， HOCHREITER S. Fast and accurate deep network learning by Exponential Linear Units （ELUs）［EB/OL］. ［2023-12-03］..
[24]	SHAO K H， ZHANG Y， WEN Y， et al. DTI-HETA： prediction of drug-target interactions based on GCN and GAT on heterogeneous graph［J］. Briefings in Bioinformatics， 2022， 23（3）： No.bbac109.
[25]	ZHENG Y， PENG H， ZHANG X， et al. Predicting drug targets from heterogeneous spaces using anchor graph hashing and ensemble learning［C］// Proceedings of the 2018 International Joint Conference on Neural Networks. Piscataway： IEEE， 2018： 1-7.
[26]	LEE I， KEUM J， NAM H. DeepConv-DTI： prediction of drug-target interactions via deep learning with convolution on protein sequences［J］. PLoS Computational Biology， 2019， 15（6）： No.e1007129.
[27]	KIPF T N， WELLING M. Semi-supervised classification with graph convolutional networks［EB/OL］. ［2023-12-03］..
[28]	HUA Y， SONG X， FENG Z， et al. MFR-DTA： a multi-functional and robust model for predicting drug-target binding affinity and region［J］. Bioinformatics， 2023， 39（2）： No.btad056.
[29]	ZHANG J， LIU Z， PAN Y， et al. IMAEN： an interpretable molecular augmentation model for drug-target interaction prediction［J］. Expert Systems with Applications， 2024， 238（Pt C）： No.121882.
[30]	PASZKE A， GROSS S， MASSA F， et al. PyTorch： an imperative style， high-performance deep learning library［C］// Proceedings of the 33rd International Conference on Neural Information Processing Systems. Red Hook： Curran Associates Inc.， 2019： 8026-8037.
[31]	WANG M， ZHENG D， YE Z， et al. Deep graph library： a graph-centric， highly-performant package for graph neural networks［EB/OL］. ［2023-11-15］..

Prediction of drug-target interactions based on sequence and multi-view networks

基于序列和多视角网络的药物-靶标相互作用预测

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 10

References 31

Related Articles 15

Recommended Articles

Metrics

[1]	Jinggang LYU, Shaorui PENG, Shuo GAO, Jin ZHOU. Speech enhancement network driven by complex frequency attention and multi-scale frequency enhancement [J]. Journal of Computer Applications, 2025, 45(9): 2957-2965.
[2]	Weigang LI, Jiale SHAO, Zhiqiang TIAN. Point cloud classification and segmentation network based on dual attention mechanism and multi-scale fusion [J]. Journal of Computer Applications, 2025, 45(9): 3003-3010.
[3]	Xiang WANG, Zhixiang CHEN, Guojun MAO. Multivariate time series prediction method combining local and global correlation [J]. Journal of Computer Applications, 2025, 45(9): 2806-2816.
[4]	Chao JING, Yutao QUAN, Yan CHEN. Improved multi-layer perceptron and attention model-based power consumption prediction algorithm [J]. Journal of Computer Applications, 2025, 45(8): 2646-2655.
[5]	Jinhao LIN, Chuan LUO, Tianrui LI, Hongmei CHEN. Thoracic disease classification method based on cross-scale attention network [J]. Journal of Computer Applications, 2025, 45(8): 2712-2719.
[6]	Haifeng WU, Liqing TAO, Yusheng CHENG. Partial label regression algorithm integrating feature attention and residual connection [J]. Journal of Computer Applications, 2025, 45(8): 2530-2536.
[7]	Jin ZHOU, Yuzhi LI, Xu ZHANG, Shuo GAO, Li ZHANG, Jiachuan SHENG. Modulation recognition network for complex electromagnetic environments [J]. Journal of Computer Applications, 2025, 45(8): 2672-2682.
[8]	Yihan WANG, Chong LU, Zhongyuan CHEN. Multimodal sentiment analysis model with cross-modal text information enhancement [J]. Journal of Computer Applications, 2025, 45(7): 2237-2244.
[9]	Haoyu LIU, Pengwei KONG, Yaoli WANG, Qing CHANG. Pedestrian detection algorithm based on multi-view information [J]. Journal of Computer Applications, 2025, 45(7): 2325-2332.
[10]	Xiaoqiang ZHAO, Yongyong LIU, Yongyong HUI, Kai LIU. Batch process quality prediction model using improved time-domain convolutional network with multi-head self-attention mechanism [J]. Journal of Computer Applications, 2025, 45(7): 2245-2252.
[11]	Huibin WANG, Zhan’ao HU, Jie HU, Yuanwei XU, Bo WEN. Time series forecasting model based on segmented attention mechanism [J]. Journal of Computer Applications, 2025, 45(7): 2262-2268.
[12]	Chen LIANG, Yisen WANG, Qiang WEI, Jiang DU. Source code vulnerability detection method based on Transformer-GCN [J]. Journal of Computer Applications, 2025, 45(7): 2296-2303.
[13]	Sheping ZHAI, Yan HUANG, Qing YANG, Rui YANG. Multi-view entity alignment combining triples and text attributes [J]. Journal of Computer Applications, 2025, 45(6): 1793-1800.
[14]	Xiang WANG, Qianqian CUI, Xiaoming ZHANG, Jianchao WANG, Zhenzhou WANG, Jialin SONG. Wireless capsule endoscopy image classification model based on improved ConvNeXt [J]. Journal of Computer Applications, 2025, 45(6): 2016-2024.
[15]	Weigang LI, Xinyi LI, Yongqiang WANG, Yuntao ZHAO. Point cloud classification and segmentation method based on adaptive dynamic graph convolution and parameter-free attention [J]. Journal of Computer Applications, 2025, 45(6): 1980-1986.