{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,15]],"date-time":"2026-04-15T18:18:39Z","timestamp":1776277119565,"version":"3.50.1"},"reference-count":47,"publisher":"Oxford University Press (OUP)","issue":"11","license":[{"start":{"date-parts":[[2023,10,20]],"date-time":"2023-10-20T00:00:00Z","timestamp":1697760000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/100013015","name":"Cedars-Sinai Medical Center","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100013015","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2023,11,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:sec>\n                  <jats:title>Motivation<\/jats:title>\n                  <jats:p>The identification of cancer genes is a critical yet challenging problem in cancer genomics research. Existing computational methods, including deep graph neural networks, fail to exploit the multilayered gene\u2013gene interactions or provide limited explanations for their predictions. These methods are restricted to a single biological network, which cannot capture the full complexity of tumorigenesis. Models trained on different biological networks often yield different and even opposite cancer gene predictions, hindering their trustworthy adaptation. Here, we introduce an Explainable Multilayer Graph Neural Network (EMGNN) approach to identify cancer genes by leveraging multiple gene\u2013gene interaction networks and pan-cancer multi-omics data. Unlike conventional graph learning on a single biological network, EMGNN uses a multilayered graph neural network to learn from multiple biological networks for accurate cancer gene prediction.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Results<\/jats:title>\n                  <jats:p>Our method consistently outperforms all existing methods, with an average 7.15% improvement in area under the precision\u2013recall curve over the current state-of-the-art method. Importantly, EMGNN integrated multiple graphs to prioritize newly predicted cancer genes with conflicting predictions from single biological networks. For each prediction, EMGNN provided valuable biological insights via both model-level feature importance explanations and molecular-level gene set enrichment analysis. Overall, EMGNN offers a powerful new paradigm of graph learning through modeling the multilayered topological gene relationships and provides a valuable tool for cancer genomics research.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Availability and implementation<\/jats:title>\n                  <jats:p>Our code is publicly available at https:\/\/github.com\/zhanglab-aim\/EMGNN.<\/jats:p>\n               <\/jats:sec>","DOI":"10.1093\/bioinformatics\/btad643","type":"journal-article","created":{"date-parts":[[2023,10,20]],"date-time":"2023-10-20T17:54:56Z","timestamp":1697824496000},"source":"Crossref","is-referenced-by-count":33,"title":["Explainable Multilayer Graph Neural Network for cancer gene prediction"],"prefix":"10.1093","volume":"39","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-9905-1646","authenticated-orcid":false,"given":"Michail","family":"Chatzianastasis","sequence":"first","affiliation":[{"name":"LIX, \u00c9cole Polytechnique, IP Paris , Rte de Saclay , Palaiseau, 91120, France"}]},{"given":"Michalis","family":"Vazirgiannis","sequence":"additional","affiliation":[{"name":"LIX, \u00c9cole Polytechnique, IP Paris , Rte de Saclay , Palaiseau, 91120, France"}]},{"given":"Zijun","family":"Zhang","sequence":"additional","affiliation":[{"name":"Division of Artificial Intelligence in Medicine, Cedars-Sinai Medical Center , 116 N. Robertson Boulevard , Los Angeles, CA 90048, United States"}]}],"member":"286","published-online":{"date-parts":[[2023,10,20]]},"reference":[{"key":"2023111005484406600_btad643-B1","doi-asserted-by":"crossref","first-page":"8","DOI":"10.4238\/vol8-1gmr518","article-title":"Polymorphisms and DNA methylation of gene TP53 associated with extra-axial brain tumors","volume":"8","author":"Almeida","year":"2009","journal-title":"Genet Mol Res"},{"key":"2023111005484406600_btad643-B2","doi-asserted-by":"crossref","first-page":"745","DOI":"10.1104\/pp.104.040071","article-title":"Functional annotation of the Arabidopsis genome using controlled vocabularies","volume":"135","author":"Berardini","year":"2004","journal-title":"Plant Physiol"},{"key":"2023111005484406600_btad643-B3","doi-asserted-by":"crossref","first-page":"2763","DOI":"10.3390\/ijms10062763","article-title":"Yeast two-hybrid, a powerful tool for systems biology","volume":"10","author":"Br\u00fcckner","year":"2009","journal-title":"Int J Mol Sci"},{"key":"2023111005484406600_btad643-B4","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":"2023111005484406600_btad643-B5","doi-asserted-by":"crossref","first-page":"77","DOI":"10.1038\/s41598-021-03471-w","article-title":"High BRCA1 gene expression increases the risk of early distant metastasis in ER+ breast cancers","volume":"12","author":"Chang","year":"2022","journal-title":"Sci Rep"},{"key":"2023111005484406600_btad643-B6","doi-asserted-by":"crossref","first-page":"129","DOI":"10.1186\/s13059-016-0989-x","article-title":"MUFFINN: cancer gene discovery via network analysis of somatic mutation data","volume":"17","author":"Cho","year":"2016","journal-title":"Genome Biol"},{"key":"2023111005484406600_btad643-B7","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":"2023111005484406600_btad643-B8","doi-asserted-by":"crossref","first-page":"425","DOI":"10.1126\/science.1180823","article-title":"The genetic landscape of a cell","volume":"327","author":"Costanzo","year":"2010","journal-title":"Science"},{"key":"2023111005484406600_btad643-B9","doi-asserted-by":"crossref","first-page":"348","DOI":"10.1038\/s41586-021-03922-4","article-title":"Biologically informed deep neural network for prostate cancer discovery","volume":"598","author":"Elmarakeby","year":"2021","journal-title":"Nature"},{"key":"2023111005484406600_btad643-B10","doi-asserted-by":"crossref","first-page":"btac757","DOI":"10.1093\/bioinformatics\/btac757","article-title":"GSEApy: a comprehensive package for performing gene set enrichment analysis in Python","volume":"39","author":"Fang","year":"2023","journal-title":"Bioinformatics"},{"key":"2023111005484406600_btad643-B11","doi-asserted-by":"crossref","first-page":"569","DOI":"10.1038\/ng.3259","article-title":"Understanding multicellular function and disease with human tissue-specific networks","volume":"47","author":"Greene","year":"2015","journal-title":"Nat Genet"},{"key":"2023111005484406600_btad643-B12","doi-asserted-by":"crossref","first-page":"752694","DOI":"10.3389\/fonc.2021.752694","article-title":"COL5A1 serves as a biomarker of tumor progression and poor prognosis and may be a potential therapeutic target in gliomas","volume":"11","author":"Gu","year":"2021","journal-title":"Front Oncol"},{"key":"2023111005484406600_btad643-B13","doi-asserted-by":"crossref","first-page":"83","DOI":"10.1016\/S0300-9084(01)01356-6","article-title":"TP53: a key gene in human cancer","volume":"84","author":"Guimaraes","year":"2002","journal-title":"Biochimie"},{"key":"2023111005484406600_btad643-B14","doi-asserted-by":"crossref","first-page":"222","DOI":"10.1038\/s42256-022-00454-y","article-title":"Reusability report: capturing properties of biological objects and their relationships using graph neural networks","volume":"4","author":"Hong","year":"2022","journal-title":"Nat Mach Intell"},{"key":"2023111005484406600_btad643-B15","doi-asserted-by":"crossref","first-page":"61","DOI":"10.1038\/nmeth.4514","article-title":"NetSig: network-based discovery from cancer genomes","volume":"15","author":"Horn","year":"2018","journal-title":"Nat Methods"},{"key":"2023111005484406600_btad643-B16","doi-asserted-by":"crossref","first-page":"221","DOI":"10.1016\/j.cels.2017.09.003","article-title":"Network-based coverage of mutational profiles reveals cancer genes","volume":"5","author":"Hristov","year":"2017","journal-title":"Cell Syst"},{"key":"2023111005484406600_btad643-B17","doi-asserted-by":"crossref","first-page":"484","DOI":"10.1016\/j.cels.2018.03.001","article-title":"Systematic evaluation of molecular networks for discovery of disease genes","volume":"6","author":"Huang","year":"2018","journal-title":"Cell Syst"},{"key":"2023111005484406600_btad643-B18","doi-asserted-by":"crossref","first-page":"55","DOI":"10.1186\/s13059-021-02264-8","article-title":"GeneWalk identifies relevant gene functions for a biological context using network representation learning","volume":"22","author":"Ietswaart","year":"2021","journal-title":"Genome Biol"},{"key":"2023111005484406600_btad643-B19","doi-asserted-by":"crossref","first-page":"D712","DOI":"10.1093\/nar\/gkq1156","article-title":"ConsensusPathDB: toward a more complete picture of cell biology","volume":"39","author":"Kamburov","year":"2011","journal-title":"Nucleic Acids Res"},{"key":"2023111005484406600_btad643-B20","doi-asserted-by":"crossref","first-page":"e1002886","DOI":"10.1371\/journal.pcbi.1002886","article-title":"Interpretation of genomic variants using a unified biological network approach","volume":"9","author":"Khurana","year":"2013","journal-title":"PLoS Comput Biol"},{"key":"2023111005484406600_btad643-B21","author":"Kingma","year":"2015"},{"key":"2023111005484406600_btad643-B22","author":"Kipf","year":"2017"},{"key":"2023111005484406600_btad643-B23","author":"Kokhlikyan"},{"key":"2023111005484406600_btad643-B24","doi-asserted-by":"crossref","first-page":"417","DOI":"10.1016\/j.cels.2015.12.004","article-title":"The molecular signatures database hallmark gene set collection","volume":"1","author":"Liberzon","year":"2015","journal-title":"Cell Syst"},{"key":"2023111005484406600_btad643-B25","doi-asserted-by":"crossref","first-page":"400","DOI":"10.1016\/j.cell.2018.02.052","article-title":"An integrated TCGA pan-cancer clinical data resource to drive high-quality survival outcome analytics","volume":"173","author":"Liu","year":"2018","journal-title":"Cell"},{"key":"2023111005484406600_btad643-B26","doi-asserted-by":"crossref","first-page":"486","DOI":"10.1038\/ng.3275","article-title":"Transposon mutagenesis identifies genetic drivers of Braf V600E melanoma","volume":"47","author":"Mann","year":"2015","journal-title":"Nat Genet"},{"key":"2023111005484406600_btad643-B27","doi-asserted-by":"crossref","first-page":"588","DOI":"10.1086\/514346","article-title":"Mendelian Inheritance in Man and its online version, OMIM","volume":"80","author":"McKusick","year":"2007","journal-title":"Am J Hum Genet"},{"key":"2023111005484406600_btad643-B28","doi-asserted-by":"crossref","first-page":"786","DOI":"10.1126\/science.aax4438","article-title":"Exploring genetic interaction manifolds constructed from rich single-cell phenotypes","volume":"365","author":"Norman","year":"2019","journal-title":"Science"},{"key":"2023111005484406600_btad643-B29","doi-asserted-by":"crossref","first-page":"1105","DOI":"10.1038\/gim.2017.37","article-title":"Sherloc: a comprehensive refinement of the ACMG\u2013AMP variant classification criteria","volume":"19","author":"Nykamp","year":"2017","journal-title":"Genet Med"},{"key":"2023111005484406600_btad643-B30","doi-asserted-by":"crossref","first-page":"ii120","DOI":"10.1093\/bioinformatics\/btac478","article-title":"GNN-SubNet: disease subnetwork detection with explainable graph neural networks","volume":"38","author":"Pfeifer","year":"2022","journal-title":"Bioinformatics"},{"key":"2023111005484406600_btad643-B31","doi-asserted-by":"crossref","first-page":"133","DOI":"10.1038\/s41592-020-01010-5","article-title":"Deciphering molecular interactions by proximity labeling","volume":"18","author":"Qin","year":"2021","journal-title":"Nat Methods"},{"key":"2023111005484406600_btad643-B32","doi-asserted-by":"crossref","first-page":"405","DOI":"10.1186\/1471-2105-9-405","article-title":"IRefIndex: a consolidated protein interaction database with provenance","volume":"9","author":"Razick","year":"2008","journal-title":"BMC Bioinformatics"},{"key":"2023111005484406600_btad643-B33","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1186\/s13059-018-1612-0","article-title":"The network of cancer genes (NCG): a comprehensive catalogue of known and candidate cancer genes from cancer sequencing screens","volume":"20","author":"Repana","year":"2019","journal-title":"Genome Biol"},{"key":"2023111005484406600_btad643-B34","doi-asserted-by":"crossref","first-page":"i972","DOI":"10.1093\/bioinformatics\/bty613","article-title":"Hierarchical HotNet: identifying hierarchies of altered subnetworks","volume":"34","author":"Reyna","year":"2018","journal-title":"Bioinformatics"},{"key":"2023111005484406600_btad643-B35","doi-asserted-by":"crossref","first-page":"513","DOI":"10.1038\/s42256-021-00325-y","article-title":"Integration of multiomics data with graph convolutional networks to identify new cancer genes and their associated molecular mechanisms","volume":"3","author":"Schulte-Sasse","year":"2021","journal-title":"Nat Mach Intell"},{"key":"2023111005484406600_btad643-B36","doi-asserted-by":"crossref","first-page":"717","DOI":"10.1038\/s41578-021-00339-3","article-title":"Machine learning methods to model multicellular complexity and tissue specificity","volume":"6","author":"Sealfon","year":"2021","journal-title":"Nat Rev Mater"},{"key":"2023111005484406600_btad643-B37","doi-asserted-by":"crossref","first-page":"27","DOI":"10.1093\/carcin\/bgp220","article-title":"Epigenetics in cancer","volume":"31","author":"Sharma","year":"2010","journal-title":"Carcinogenesis"},{"key":"2023111005484406600_btad643-B38","doi-asserted-by":"crossref","first-page":"1634","DOI":"10.1038\/s41587-022-01353-8","article-title":"Genome-wide mapping of somatic mutation rates uncovers drivers of cancer","volume":"40","author":"Sherman","year":"2022","journal-title":"Nat Biotechnol"},{"key":"2023111005484406600_btad643-B39","doi-asserted-by":"crossref","first-page":"696","DOI":"10.1038\/s41568-018-0060-1","article-title":"The cosmic cancer gene census: describing genetic dysfunction across all human cancers","volume":"18","author":"Sondka","year":"2018","journal-title":"Nat Rev Cancer"},{"key":"2023111005484406600_btad643-B40","doi-asserted-by":"crossref","first-page":"D607","DOI":"10.1093\/nar\/gky1131","article-title":"STRING v11: protein\u2013protein association networks with increased coverage, supporting functional discovery in genome-wide experimental datasets","volume":"47","author":"Szklarczyk","year":"2019","journal-title":"Nucleic Acids Res"},{"key":"2023111005484406600_btad643-B41","author":"Velikovi","year":"2018"},{"key":"2023111005484406600_btad643-B42","doi-asserted-by":"crossref","first-page":"180061","DOI":"10.1038\/sdata.2018.61","article-title":"Unifying cancer and normal RNA sequencing data from different sources","volume":"5","author":"Wang","year":"2018","journal-title":"Sci Data"},{"key":"2023111005484406600_btad643-B43","doi-asserted-by":"crossref","first-page":"1113","DOI":"10.1038\/ng.2764","article-title":"The cancer genome atlas pan-cancer analysis project","volume":"45","author":"Weinstein","year":"2013","journal-title":"Nat Genet"},{"key":"2023111005484406600_btad643-B44","author":"Xu","year":"2019"},{"key":"2023111005484406600_btad643-B45","first-page":"6419695","article-title":"The hypoxia-related gene COL5A1 is a prognostic and immunological biomarker for multiple human tumors","volume":"2022","author":"Zhu","year":"2022","journal-title":"Oxid Med Cell Longev"},{"key":"2023111005484406600_btad643-B46","doi-asserted-by":"crossref","first-page":"i190","DOI":"10.1093\/bioinformatics\/btx252","article-title":"Predicting multicellular function through multi-layer tissue networks","volume":"33","author":"Zitnik","year":"2017","journal-title":"Bioinformatics"},{"key":"2023111005484406600_btad643-B47","doi-asserted-by":"crossref","first-page":"71","DOI":"10.1016\/j.inffus.2018.09.012","article-title":"Machine learning for integrating data in biology and medicine: principles, practice, and opportunities","volume":"50","author":"Zitnik","year":"2019","journal-title":"Inf Fusion"}],"container-title":["Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/bioinformatics\/advance-article-pdf\/doi\/10.1093\/bioinformatics\/btad643\/52306228\/btad643.pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/39\/11\/btad643\/53006775\/btad643.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/39\/11\/btad643\/53006775\/btad643.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,11,10]],"date-time":"2023-11-10T06:14:21Z","timestamp":1699596861000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article\/doi\/10.1093\/bioinformatics\/btad643\/7325352"}},"subtitle":[],"editor":[{"given":"Pier Luigi","family":"Martelli","sequence":"additional","affiliation":[]}],"short-title":[],"issued":{"date-parts":[[2023,10,20]]},"references-count":47,"journal-issue":{"issue":"11","published-print":{"date-parts":[[2023,11,1]]}},"URL":"https:\/\/doi.org\/10.1093\/bioinformatics\/btad643","relation":{},"ISSN":["1367-4803","1367-4811"],"issn-type":[{"value":"1367-4803","type":"print"},{"value":"1367-4811","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2023,11,1]]},"published":{"date-parts":[[2023,10,20]]},"article-number":"btad643"}}