{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,3]],"date-time":"2026-06-03T18:28:03Z","timestamp":1780511283475,"version":"3.54.1"},"reference-count":38,"publisher":"Oxford University Press (OUP)","issue":"23","license":[{"start":{"date-parts":[[2021,6,21]],"date-time":"2021-06-21T00:00:00Z","timestamp":1624233600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/academic.oup.com\/journals\/pages\/open_access\/funder_policies\/chorus\/standard_publication_model"}],"funder":[{"DOI":"10.13039\/100000002","name":"US National Institutes of Health","doi-asserted-by":"crossref","award":["U01-CA198941"],"award-info":[{"award-number":["U01-CA198941"]}],"id":[{"id":"10.13039\/100000002","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/100000054","name":"National Cancer Institute","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100000054","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2021,12,7]]},"abstract":"<jats:title>ABSTRACT<\/jats:title>\n               <jats:sec>\n                  <jats:title>Background<\/jats:title>\n                  <jats:p>Link prediction is an important and well-studied problem in network biology. Recently, graph representation learning methods, including Graph Convolutional Network (GCN)-based node embedding have drawn increasing attention in link prediction.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Motivation<\/jats:title>\n                  <jats:p>An important component of GCN-based network embedding is the convolution matrix, which is used to propagate features across the network. Existing algorithms use the degree-normalized adjacency matrix for this purpose, as this matrix is closely related to the graph Laplacian, capturing the spectral properties of the network. In parallel, it has been shown that GCNs with a single layer can generate more robust embeddings by reducing the number of parameters. Laplacian-based convolution is not well suited to single-layered GCNs, as it limits the propagation of information to immediate neighbors of a node.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Results<\/jats:title>\n                  <jats:p>Capitalizing on the rich literature on unsupervised link prediction, we propose using node similarity-based convolution matrices in GCNs to compute node embeddings for link prediction. We consider eight representative node-similarity measures (Common Neighbors, Jaccard Index, Adamic-Adar, Resource Allocation, Hub- Depressed Index, Hub-Promoted Index, Sorenson Index and Salton Index) for this purpose. We systematically compare the performance of the resulting algorithms against GCNs that use the degree-normalized adjacency matrix for convolution, as well as other link prediction algorithms. In our experiments, we use three-link prediction tasks involving biomedical networks: drug\u2013disease association prediction, drug\u2013drug interaction prediction and protein\u2013protein interaction prediction. Our results show that node similarity-based convolution matrices significantly improve the link prediction performance of GCN-based embeddings.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Conclusion<\/jats:title>\n                  <jats:p>As sophisticated machine-learning frameworks are increasingly employed in biological applications, historically well-established methods can be useful in making a head-start.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Availability and implementation<\/jats:title>\n                  <jats:p>Our method, SiGraC, is implemented as a Python library and is freely available at https:\/\/github.com\/mustafaCoskunAgu\/SiGraC.<\/jats:p>\n               <\/jats:sec>","DOI":"10.1093\/bioinformatics\/btab464","type":"journal-article","created":{"date-parts":[[2021,6,17]],"date-time":"2021-06-17T11:12:13Z","timestamp":1623928333000},"page":"4501-4508","source":"Crossref","is-referenced-by-count":55,"title":["Node similarity-based graph convolution for link prediction in biological networks"],"prefix":"10.1093","volume":"37","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-4805-1416","authenticated-orcid":false,"given":"Mustafa","family":"Co\u015fkun","sequence":"first","affiliation":[{"name":"Department of Computer Engineering, Abdullah G\u00fcl University , Kayseri, Turkey"},{"name":"Hakkari University , Kayseri 38080, Turkey"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mehmet","family":"Koyut\u00fcrk","sequence":"additional","affiliation":[{"name":"Department of Computer and Data Sciences, Case Western Reserve University , Cleveland, OH 44106, USA"},{"name":"Center for Proteomics and Bioinformatics, Case Western Reserve University , Cleveland, OH 44106, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"286","published-online":{"date-parts":[[2021,6,21]]},"reference":[{"key":"2023061310492968400_btab464-B1","doi-asserted-by":"crossref","first-page":"211","DOI":"10.1016\/S0378-8733(03)00009-1","article-title":"Friends and neighbors on the web","volume":"25","author":"Adamic","year":"2003","journal-title":"Soc. Netw"},{"key":"2023061310492968400_btab464-B2","doi-asserted-by":"crossref","first-page":"D267","DOI":"10.1093\/nar\/gkh061","article-title":"The unified medical language system (UMLS): integrating biomedical terminology","volume":"32","author":"Bodenreider","year":"2004","journal-title":"Nucleic Acids Res"},{"key":"2023061310492968400_btab464-B3","doi-asserted-by":"crossref","first-page":"i219","DOI":"10.1093\/bioinformatics\/btu263","article-title":"New directions for diffusion-based network prediction of protein function: incorporating pathways with confidence","volume":"30","author":"Cao","year":"2014","journal-title":"Bioinformatics"},{"key":"2023061310492968400_btab464-B4","doi-asserted-by":"crossref","first-page":"540","DOI":"10.1016\/j.cels.2016.10.017","article-title":"Compact integration of multi-network topology for functional analysis of genes","volume":"3","author":"Cho","year":"2016","journal-title":"Cell Syst"},{"key":"2023061310492968400_btab464-B5","doi-asserted-by":"crossref","first-page":"485","DOI":"10.1109\/ICDMW.2015.195","volume-title":"2015 IEEE International Conference on Data Mining Workshop (ICDMW)","author":"Co\u015fkun","year":"2015"},{"key":"2023061310492968400_btab464-B6","doi-asserted-by":"crossref","first-page":"551","DOI":"10.1038\/nrg.2017.38","article-title":"Network propagation: a universal amplifier of genetic associations","volume":"18","author":"Cowen","year":"2017","journal-title":"Nat. Rev. Genet"},{"key":"2023061310492968400_btab464-B7","doi-asserted-by":"crossref","first-page":"D948","DOI":"10.1093\/nar\/gky868","article-title":"The comparative toxicogenomics database: update 2019","volume":"47","author":"Davis","year":"2019","journal-title":"Nucleic Acids Res"},{"key":"2023061310492968400_btab464-B8","doi-asserted-by":"crossref","first-page":"i464","DOI":"10.1093\/bioinformatics\/btaa459","article-title":"GLIDE: combining local methods and diffusion state embeddings to predict missing interactions in biological networks","volume":"36","author":"Devkota","year":"2020","journal-title":"Bioinformatics"},{"key":"2023061310492968400_btab464-B9","doi-asserted-by":"crossref","first-page":"19","DOI":"10.1186\/1756-0381-4-19","article-title":"DADA: degree-aware algorithms for network-based disease gene prioritization","volume":"4","author":"Erten","year":"2011","journal-title":"BioData Min"},{"key":"2023061310492968400_btab464-B10","doi-asserted-by":"crossref","first-page":"1561","DOI":"10.1089\/cmb.2011.0154","article-title":"Vavien: an algorithm for prioritizing candidate disease genes based on topological similarity of proteins in interaction networks","volume":"18","author":"Erten","year":"2011","journal-title":"J. Comput. Biol"},{"key":"2023061310492968400_btab464-B11","doi-asserted-by":"crossref","first-page":"D808","DOI":"10.1093\/nar\/gks1094","article-title":"STRING v9. 1: protein-protein interaction networks, with increased coverage and integration","volume":"41","author":"Franceschini","year":"2013","journal-title":"Nucleic Acids Res"},{"key":"2023061310492968400_btab464-B12","first-page":"1263","volume-title":"Proceedings of the 34th International Conference on Machine Learning","author":"Gilmer","year":"2017"},{"key":"2023061310492968400_btab464-B13","doi-asserted-by":"crossref","first-page":"496","DOI":"10.1038\/msb.2011.26","article-title":"PREDICT: a method for inferring novel drug indications with application to personalized medicine","volume":"7","author":"Gottlieb","year":"2011","journal-title":"Mol. Syst. Biol"},{"key":"2023061310492968400_btab464-B14","doi-asserted-by":"crossref","first-page":"855","DOI":"10.1145\/2939672.2939754","volume-title":"Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining","author":"Grover","year":"2016"},{"key":"2023061310492968400_btab464-B15","article-title":"Representation learning on graphs: methods and applications (2017)","author":"Hamilton","year":"2019","journal-title":"IEEE Data Engineering Bulletin"},{"key":"2023061310492968400_btab464-B16","article-title":"Semi-supervised classification with graph convolutional networks","author":"Kipf","year":"2016"},{"key":"2023061310492968400_btab464-B17","article-title":"Variational graph auto-encoders","author":"Kipf","year":"2016"},{"key":"2023061310492968400_btab464-B18","doi-asserted-by":"crossref","first-page":"355","DOI":"10.1093\/bioinformatics\/bts688","article-title":"A novel link prediction algorithm for reconstructing protein\u2013protein interaction networks by topological similarity","volume":"29","author":"Lei","year":"2013","journal-title":"Bioinformatics"},{"key":"2023061310492968400_btab464-B19","author":"Li","year":"2018"},{"key":"2023061310492968400_btab464-B20","doi-asserted-by":"crossref","first-page":"1187","DOI":"10.1093\/bioinformatics\/btw770","article-title":"LRSSL: predict and interpret drug\u2013disease associations based on data integration using sparse subspace learning","volume":"33","author":"Liang","year":"2017","journal-title":"Bioinformatics"},{"key":"2023061310492968400_btab464-B21","doi-asserted-by":"crossref","first-page":"1019","DOI":"10.1002\/asi.20591","article-title":"The link-prediction problem for social networks","volume":"58","author":"Liben-Nowell","year":"2007","journal-title":"J. Am. Soc. Inf. Sci. Technol"},{"key":"2023061310492968400_btab464-B22","doi-asserted-by":"crossref","first-page":"105","DOI":"10.1109\/2.36","article-title":"Self-organization in a perceptual network","volume":"21","author":"Linsker","year":"1988","journal-title":"Computer"},{"key":"2023061310492968400_btab464-B23","doi-asserted-by":"crossref","first-page":"1150","DOI":"10.1016\/j.physa.2010.11.027","article-title":"Link prediction in complex networks: a survey","volume":"390","author":"L\u00fc","year":"2011","journal-title":"Physica A"},{"key":"2023061310492968400_btab464-B24","doi-asserted-by":"crossref","first-page":"i28","DOI":"10.1093\/bioinformatics\/btn296","article-title":"Functional coherence in domain interaction networks","volume":"24","author":"Pandey","year":"2008","journal-title":"Bioinformatics"},{"key":"2023061310492968400_btab464-B25","first-page":"701","author":"Perozzi","year":"2014"},{"key":"2023061310492968400_btab464-B26","first-page":"385","author":"Ribeiro","year":"2017"},{"key":"2023061310492968400_btab464-B27","doi-asserted-by":"crossref","first-page":"40321","DOI":"10.1038\/srep40321","article-title":"Drug response prediction as a link prediction problem","volume":"7","author":"Stanfield","year":"2017","journal-title":"Sci. Rep"},{"key":"2023061310492968400_btab464-B28","doi-asserted-by":"crossref","first-page":"D447","DOI":"10.1093\/nar\/gku1003","article-title":"STRING v10: protein\u2013protein interaction networks, integrated over the tree of life","volume":"43","author":"Szklarczyk","year":"2015","journal-title":"Nucleic Acids Res"},{"key":"2023061310492968400_btab464-B29","first-page":"1067","author":"Tang","year":"2015"},{"key":"2023061310492968400_btab464-B30","doi-asserted-by":"crossref","first-page":"497","DOI":"10.1093\/bioinformatics\/bty637","article-title":"Random walk with restart on multiplex and heterogeneous biological networks","volume":"35","author":"Valdeolivas","year":"2019","journal-title":"Bioinformatics"},{"key":"2023061310492968400_btab464-B31","author":"Veli\u010dkovi\u0107","year":"2019"},{"key":"2023061310492968400_btab464-B32","doi-asserted-by":"crossref","first-page":"1336","DOI":"10.1039\/C7MB00188F","article-title":"Predicting protein\u2013protein interactions from protein sequences by a stacked sparse autoencoder deep neural network","volume":"13","author":"Wang","year":"2017","journal-title":"Mol. Biosyst"},{"key":"2023061310492968400_btab464-B33","doi-asserted-by":"crossref","first-page":"D1074","DOI":"10.1093\/nar\/gkx1037","article-title":"DrugBank 5.0: a major update to the DrugBank database for 2018","volume":"46","author":"Wishart","year":"2018","journal-title":"Nucleic Acids Res"},{"key":"2023061310492968400_btab464-B34","author":"Wu","year":"2019"},{"key":"2023061310492968400_btab464-B35","doi-asserted-by":"crossref","first-page":"1056","DOI":"10.1109\/TCBB.2015.2495170","article-title":"Improving identification of key players in aging via network de-noising and core inference","volume":"14","author":"Yoo","year":"2017","journal-title":"IEEE\/ACM Trans. Comput. Biol. Bioinform"},{"key":"2023061310492968400_btab464-B36","doi-asserted-by":"crossref","first-page":"1241","DOI":"10.1093\/bioinformatics\/btz718","article-title":"Graph embedding on biomedical networks: methods, applications and evaluations","volume":"36","author":"Yue","year":"2020","journal-title":"Bioinformatics"},{"key":"2023061310492968400_btab464-B37","doi-asserted-by":"crossref","first-page":"90","DOI":"10.1016\/j.jbi.2018.11.005","article-title":"Manifold regularized matrix factorization for drug-drug interaction prediction","volume":"88","author":"Zhang","year":"2018","journal-title":"J. Biomed. Inform"},{"key":"2023061310492968400_btab464-B38","doi-asserted-by":"crossref","first-page":"623","DOI":"10.1140\/epjb\/e2009-00335-8","article-title":"Predicting missing links via local information","volume":"71","author":"Zhou","year":"2009","journal-title":"Eur. Phys. J. B"}],"container-title":["Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"http:\/\/academic.oup.com\/bioinformatics\/advance-article-pdf\/doi\/10.1093\/bioinformatics\/btab464\/39300687\/btab464.pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/37\/23\/4501\/50579656\/btab464.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/37\/23\/4501\/50579656\/btab464.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,6,13]],"date-time":"2023-06-13T10:51:32Z","timestamp":1686653492000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article\/37\/23\/4501\/6307262"}},"subtitle":[],"editor":[{"given":"Jonathan","family":"Wren","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2021,6,21]]},"references-count":38,"journal-issue":{"issue":"23","published-print":{"date-parts":[[2021,12,7]]}},"URL":"https:\/\/doi.org\/10.1093\/bioinformatics\/btab464","relation":{},"ISSN":["1367-4803","1367-4811"],"issn-type":[{"value":"1367-4803","type":"print"},{"value":"1367-4811","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2021,12,1]]},"published":{"date-parts":[[2021,6,21]]}}}