Drug-Target Interactions Prediction Based on Signed Heterogeneous Graph Neural Networks

CHEN Ming; JIANG Yajian; LEI Xiujuan; PAN Yi; JI Chunyan; JIANG Wei

doi:10.23919/cje.2022.00.384

Volume 33 Issue 1

Jan. 2024

Turn off MathJax

Article Contents

Article Navigation > Chinese Journal of Electronics > 2024 > 33(1): 231-244

Ming CHEN, Yajian JIANG, Xiujuan LEI, et al., “Drug-Target Interactions Prediction Based on Signed Heterogeneous Graph Neural Networks,” Chinese Journal of Electronics, vol. 33, no. 1, pp. 231–244, 2024 doi: 10.23919/cje.2022.00.384

Citation:

Ming CHEN, Yajian JIANG, Xiujuan LEI, et al., “Drug-Target Interactions Prediction Based on Signed Heterogeneous Graph Neural Networks,” Chinese Journal of Electronics, vol. 33, no. 1, pp. 231–244, 2024 doi: 10.23919/cje.2022.00.384

Citation:

PDF( 3979 KB)

Drug-Target Interactions Prediction Based on Signed Heterogeneous Graph Neural Networks

doi: 10.23919/cje.2022.00.384

CHEN Ming^1
,,
JIANG Yajian^1
,,
LEI Xiujuan^2
,,
PAN Yi^{3
,
,},
JI Chunyan^4
,,
JIANG Wei^1
,

1.
College of Information Science and Engineering, Hunan Normal University, Changsha 410081, China
2.
School of Computer Science, Shaanxi Normal University, Xi’an 710119, China
3.
Faculty of Computer Science and Control Engineering, Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences, Shenzhen 518055, China
4.
Computer Science Department, BNU-HKBU United International College, Zhuhai 519087, China

More Information

Author Bio:
Ming CHEN is currently a Lecturer in the College of Information Science and Engineering at Hunan Normal University, Changsha, China. She received the M.S. degree in 2007 from Hunan Normal University, and the Ph.D. degree in 2012 from Wuhan University. Her current research interests mainly include graph signal processing and deep learning. (Email: chenming@hunnu.edu.cn)

Yajian JIANG is currently an M.S. student in the College of Information Science and Engineering, Hunan Normal University. His current research interests include bioinformatics, data mining, and deep learning. (Email: j_yj2020@hunnu.edu.cn)

Xiujuan LEI is currently a Professor in the School of Computer Science at Shaanxi Normal University, Xi’an, China. She received the M.S. and Ph.D. degrees from Northwestern Polytechnical University, Xi’an, China, in 2001 and 2005, respectively. Her current research interests mainly include intelligent computing and bioinformatics. (Email: xjlei@snnu.edu.cn)

Yi PAN is currently a Professor of the Faculty of Computer Science and Control Engineering, Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences. He received the B.E. and M.E. degrees in computer engineering from Tsinghua University, China, in 1982 and 1984, respectively, and the Ph.D. degree in computer science from the University of Pittsburgh, USA, in 1991. His current research interests include bioinformatics and health informatics using big data analytics, cloud computing, and machine learning technologies. (Email: yi.pan@siat.ac.cn)

Chunyan JI is currently an Assistant Professor in Department of Computer Science of BNU-HKBU United International College. She received the M.S. and Ph.D. degrees in computer science from Georgia State University. Her main research areas include deep learning, bioinformatics and sound event detection. (Email: chunyanji@uic.edu.cn)

Wei JIANG is currently an M.S. student in the College of Information Science and Engineering, Hunan Normal University. His current research interests include graph signal processing and deep learning. (Email: jw2020@smail.hunnu.edu.cn)
Corresponding author: Email: yi.pan@siat.ac.cn
Received Date: 2022-11-10
Accepted Date: 2023-03-21

Available Online: 2023-07-24

Publish Date: 2024-01-05

Abstract

Abstract

Drug-target interactions (DTIs) prediction plays an important role in the process of drug discovery. Most computational methods treat it as a binary prediction problem, determining whether there are connections between drugs and targets while ignoring relational types information. Considering the positive or negative effects of DTIs will facilitate the study on comprehensive mechanisms of multiple drugs on a common target, in this work, we model DTIs on signed heterogeneous networks, through categorizing interaction patterns of DTIs and additionally extracting interactions within drug pairs and target protein pairs. We propose signed heterogeneous graph neural networks (SHGNNs), further put forward an end-to-end framework for signed DTIs prediction, called SHGNN-DTI, which not only adapts to signed bipartite networks, but also could naturally incorporate auxiliary information from drug-drug interactions (DDIs) and protein-protein interactions (PPIs). For the framework, we solve the message passing and aggregation problem on signed DTI networks, and consider different training modes on the whole networks consisting of DTIs, DDIs and PPIs. Experiments are conducted on two datasets extracted from DrugBank and related databases, under different settings of initial inputs, embedding dimensions and training modes. The prediction results show excellent performance in terms of metric indicators, and the feasibility is further verified by the case study with two drugs on breast cancer.
- Drug-target interactions,
- Signed heterogeneous network,
- Link sign prediction,
- Graph neural networks

https://github.com/DSE-MSU/signed-bipartite-networks

https://github.com/huangjunjie-cs/SBGNN

https://github.com/pyg-team/pytorch_geometric

https://github.com/dmlc/dgl/tree/master/examples/pytorch/han

FullText(HTML)

References(45)

References

[1]	X. Q. Ru, X. C. Ye, T. Sakurai, et al., “Current status and future prospects of drug-target interaction prediction,” Briefings in Functional Genomics, vol. 20, no. 5, pp. 312–322, 2021. doi: 10.1093/bfgp/elab031
[2]	X. Chen, C. C. Yan, X. T. Zhang, et al., “Drug-target interaction prediction: Databases, web servers and computational models,” Briefings in Bioinformatics, vol. 17, no. 4, pp. 696–712, 2016. doi: 10.1093/bib/bbv066
[3]	C. C. Wang, Y. Zhao, and X. Chen, “Drug-pathway association prediction: From experimental results to computational models,” Briefings in Bioinformatics, vol. 22, no. 3, article no. bbaa061, 2021. doi: 10.1093/bib/bbaa061
[4]	W. Zhang, W. R. Lin, D. Zhang, et al., “Recent advances in the machine learning-based drug-target interaction prediction,” Current Drug Metabolism, vol. 20, no. 3, pp. 194–202, 2019. doi: 10.2174/1389200219666180821094047
[5]	Y. Pan, X. J. Lei, and Y. C. Zhang, “Association predictions of genomics, proteinomics, transcriptomics, microbiome, metabolomics, pathomics, radiomics, drug, symptoms, environment factor, and disease networks: A comprehensive approach,” Medicinal Research Reviews, vol. 42, no. 1, pp. 441–461, 2022. doi: 10.1002/med.21847
[6]	X. H. Wu, J. W. Duan, Y. Pan, et al., “Medical knowledge graph: Data sources, construction, reasoning, and applications,” Big Data Mining and Analytics, vol. 6, no. 2, pp. 201–217, 2023. doi: 10.26599/BDMA.2022.9020021
[7]	R. Li, X. Yuan, M. Radfar, et al., “Graph signal processing, graph neural network and graph learning on biological data: A systematic review,” IEEE Reviews in Biomedical Engineering, vol. 16, pp. 109–135, 2023. doi: 10.1109/RBME.2021.3122522
[8]	X. Chen, N. N. Guan, Y. Z. Sun, et al., “MicroRNA-small molecule association identification: From experimental results to computational models,” Briefings in Bioinformatics, vol. 21, no. 1, pp. 47–61, 2020. doi: 10.1093/bib/bby098
[9]	G. Z. Zhang, M. L. Li, H. Deng, et al., “SGNNMD: Signed graph neural network for predicting deregulation types of miRNA-disease associations,” Briefings in Bioinformatics, vol. 23, no. 1, article no. bbab464, 2022. doi: 10.1093/bib/bbab464
[10]	L. Guo, X. J. Lei, M. Chen, et al., “MSResG: Using GAE and residual GCN to predict drug–drug interactions based on multi-source drug features,” Interdisciplinary Sciences: Computational Life Sciences, vol. 15, no. 2, pp. 171–188, 2023. doi: 10.1007/s12539-023-00550-6
[11]	Y. C. Zhang, X. J. Lei, Y. Pan, et al., “Drug repositioning with GraphSAGE and clustering constraints based on drug and disease networks,” Frontiers in Pharmacology, vol. 13, article no. 872785, 2022. doi: 10.3389/fphar.2022.872785
[12]	G. Zhao, Q. G. Wang, F. Yao, et al., “Survey on large-scale graph neural network systems,” Journal of Software, vol. 33, no. 1, pp. 150–170, 2022. (in Chinese) doi: 10.13328/j.cnki.jos.006311
[13]	L. Zhang, C. C. Wang, and X. Chen, “Predicting drug-target binding affinity through molecule representation block based on multi-head attention and skip connection,” Briefings in Bioinformatics, vol. 23, no. 6, article no. bbac468, 2022. doi: 10.1093/bib/bbac468
[14]	H. Z. Wang, F. Huang, Z. K. Xiong, et al., “A heterogeneous network-based method with attentive meta-path extraction for predicting drug-target interactions,” Briefings in Bioinformatics, vol. 23, no. 4, article no. bbac184, 2022. doi: 10.1093/bib/bbac184
[15]	N. B. Torres and C. Altafini, “Drug combinatorics and side effect estimation on the signed human drug-target network,” BMC Systems Biology, vol. 10, no. 1, article no. 74, 2016. doi: 10.1186/s12918-016-0326-8
[16]	B. F. Hu, H. Wang, and Z. M. Yu, “Drug side-effect prediction via random walk on the signed heterogeneous drug network,” Molecules, vol. 24, no. 20, article no. 3668, 2019. doi: 10.3390/molecules24203668
[17]	Y. F. Shang, L. Gao, Q. Zou, et al., “Prediction of drug-target interactions based on multi-layer network representation learning,” Neurocomputing, vol. 434, pp. 80–89, 2021. doi: 10.1016/j.neucom.2020.12.068
[18]	Y. X. Gong, B. Liao, P. Wang, et al., “DrugHybrid_BS: Using hybrid feature combined with bagging-SVM to predict potentially druggable proteins,” Frontiers in Pharmacology, vol. 12, article no. 771808, 2021. doi: 10.3389/fphar.2021.771808
[19]	Y. Y. Chu, A. C. Kaushik, X. G. Wang, et al., “DTI-CDF: A cascade deep forest model towards the prediction of drug-target interactions based on hybrid features,” Briefings in Bioinformatics, vol. 22, no. 1, pp. 451–462, 2021. doi: 10.1093/bib/bbz152
[20]	Y. J. Ding, J. J. Tang, F. Guo, et al., “Identification of drug-target interactions via multiple kernel-based triple collaborative matrix factorization,” Briefings in Bioinformatics, vol. 23, no. 2, article no. bbab582, 2022. doi: 10.1093/bib/bbab582
[21]	X. Chen, M. X. Liu, and G. Y. Yan, “Drug-target interaction prediction by random walk on the heterogeneous network,” Molecular BioSystems, vol. 8, no. 7, pp. 1970–1978, 2012. doi: 10.1039/c2mb00002d
[22]	Y. C. Zhang, X. J. Lei, Z. Q. Fang, et al., “CircRNA-disease associations prediction based on metapath2vec++ and matrix factorization,” Big Data Mining and Analytics, vol. 3, no. 4, pp. 280–291, 2020. doi: 10.26599/BDMA.2020.9020025
[23]	X. Liu and M. Yang, “Research on conversational machine reading comprehension based on dynamic graph neural network,” Journal of Integration Technology, vol. 11, no. 2, pp. 67–78, 2022. (in Chinese) doi: 10.12146/j.issn.2095-3135.20211122001
[24]	Y. Y. Wang, X. J. Lei, Y. Pan, “Predicting microbe-disease association based on heterogeneous network and global graph feature learning,” Chinese Journal of Electronics, vol. 31, no. 2, pp. 345–353, 2022. doi: 10.1049/cje.2020.00.212
[25]	H. T. Fu, F. Huang, X. Liu, et al., “MVGCN: Data integration through multi-view graph convolutional network for predicting links in biomedical bipartite networks,” Bioinformatics, vol. 38, no. 2, pp. 426–434, 2022. doi: 10.1093/bioinformatics/btab651
[26]	D. S. Wishart, Y. D. Feunang, A. C. Guo, et al., “DrugBank 5.0: A major update to the drugbank database for 2018,” Nucleic Acids Research, vol. 46, no. D1, pp. D1074–D1082, 2018. doi: 10.1093/nar/gkx1037
[27]	T. Lee and Y. Yoon, “Drug repositioning using drug-disease vectors based on an integrated network,” BMC Bioinformatics, vol. 19, no. 1, article no. 446, 2018. doi: 10.1186/s12859-018-2490-x
[28]	M. Chen, Y. Pan, and C. Y. Ji, “Predicting drug drug interactions by signed graph filtering-based convolutional networks,” in Proceedings of the 17th International Symposium on Bioinformatics Research and Applications, Shenzhen, China, pp. 375–387, 2021.
[29]	M. Chen, W. Jiang, Y. Pan, et al., “SGFNNs: Signed graph filtering-based neural networks for predicting drug–drug interactions,” Journal of Computational Biology, vol. 29, no. 10, pp. 1104–1116, 2022. doi: 10.1089/cmb.2022.0113
[30]	T. Derr, Y. Ma, and J. L. Tang, “Signed graph convolutional networks,” in Proceedings of 2018 IEEE International Conference on Data Mining, Singapore, pp. 929–934, 2018.
[31]	J. J. Huang, H. W. Shen, Q. Cao, et al., “Signed bipartite graph neural networks,” in Proceedings of the 30th ACM International Conference on Information & Knowledge Management, Queensland, Australia, pp. 740–749, 2021.
[32]	S. Kim, J. Chen, T. J. Cheng, et al., “PubChem in 2021: New data content and improved web interfaces,” Nucleic Acids Research, vol. 49, no. D1, pp. D1388–D1395, 2021. doi: 10.1093/nar/gkaa971
[33]	D. Szklarczyk, A. L. Gable, K. C. Nastou, et al., “The string database in 2021: Customizable protein–protein networks, and functional characterization of user-uploaded gene/measurement sets,” Nucleic Acids Research, vol. 49, no. D1, pp. D605–D612, 2021. doi: 10.1093/nar/gkaa1074
[34]	T. Derr, C. Johnson, Y. Chang, et al., “Balance in signed bipartite networks,” in Proceedings of the 28th ACM International Conference on Information and Knowledge Management, Beijing, China, pp. 1221–1230, 2019.
[35]	T. N. Kipf and M. Welling, “Semi-supervised classification with graph convolutional networks,” in Proceedings of the 5th International Conference on Learning Representations, Toulon, France, 2017.
[36]	W. L. Hamilton, R. Ying, and J. Leskovec, “Inductive representation learning on large graphs,” in Proceedings of the 31st International Conference on Neural Information Processing Systems, Long Beach, CA, USA, pp. 1025–1035, 2017.
[37]	X. Wang, H. Y. Ji, C. Shi, et al., “Heterogeneous graph attention network,” in Proceedings of the World Wide Web Conference, San Francisco, CA, USA, pp. 2022–2032, 2019.
[38]	A. T. Jacobs, D. M. Castaneda-Cruz, M. M. Rose, et al., “Targeted therapy for breast cancer: An overview of drug classes and outcomes,” Biochemical Pharmacology, vol. 204, article no. 115209, 2022. doi: 10.1016/j.bcp.2022.115209
[39]	Y. S. Lu, K. S. Lee, T. Y. Chao, et al., “A phase IB study of alpelisib or buparlisib combined with tamoxifen plus goserelin in premenopausal women with HR-positive HER2-negative advanced breast cancer,” Clinical Cancer Research, vol. 27, no. 2, pp. 408–417, 2021. doi: 10.1158/1078-0432.CCR-20-1008
[40]	Y. B. Baytur, K. Ozbilgin, S. Cilaker, et al., “A comparative study of the effect of raloxifene and gosereline on uterine leiomyoma volume changes and estrogen receptor, progesterone receptor, bcl-2 and p53 expression immunohistochemically in premenopausal women,” European Journal of Obstetrics & Gynecology and Reproductive Biology, vol. 135, no. 1, pp. 94–103, 2007. doi: 10.1016/j.ejogrb.2006.07.042
[41]	E. A. Mostaghel, P. S. Nelson, P. Lange, et al., “Targeted androgen pathway suppression in localized prostate cancer: A pilot study,” Journal of Clinical Oncology, vol. 32, no. 3, pp. 229–237, 2014. doi: 10.1200/JCO.2012.48.6431
[42]	Y. L. Lo and W. Wang, “Formononetin potentiates epirubicin-induced apoptosis via ROS production in HeLa cells in vitro,” Chemico-Biological Interactions, vol. 205, no. 3, pp. 188–197, 2013. doi: 10.1016/j.cbi.2013.07.003
[43]	Y. L. Lo and W. C. Tu, “Co-encapsulation of chrysophsin-1 and epirubicin in PEGylated liposomes circumvents multidrug resistance in HeLa cells,” Chemico-Biological Interactions, vol. 242, pp. 13–23, 2015. doi: 10.1016/j.cbi.2015.08.023
[44]	F. Fan, R. Hu, A. Munzli, et al., “Utilization of human nuclear receptors as an early counter screen for off-target activity: A case study with a compendium of 615 known drugs,” Toxicological Sciences, vol. 145, no. 2, pp. 283–295, 2015. doi: 10.1093/toxsci/kfv052
[45]	H. Hua, H. Y. Zhang, Q. B. Kong, et al., “Mechanisms for estrogen receptor expression in human cancer,” Experimental Hematology & Oncology, vol. 7, article no. 24, 2018. doi: 10.1186/s40164-018-0116-7