{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,29]],"date-time":"2026-06-29T11:47:13Z","timestamp":1782733633839,"version":"3.54.5"},"reference-count":41,"publisher":"Oxford University Press (OUP)","issue":"2","license":[{"start":{"date-parts":[[2026,5,30]],"date-time":"2026-05-30T00:00:00Z","timestamp":1780099200000},"content-version":"vor","delay-in-days":64,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"}],"funder":[{"DOI":"10.13039\/100000002","name":"National Institutes of Health","doi-asserted-by":"publisher","award":["P20GM104420"],"award-info":[{"award-number":["P20GM104420"]}],"id":[{"id":"10.13039\/100000002","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,3,27]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Integration and analysis of multi-omics data provide valuable insights for improving cancer subtype classification. However, such data are inherently heterogeneous, high-dimensional, and exhibit complex intra- and inter-modality dependencies. Graph neural networks provide a principled framework for modeling these structures, but existing approaches often rely on prior knowledge or predefined similarity networks that produce either undirected or unweighted graphs, failing to capture task-specific directionality and interaction strengths. Interpretability at both the modality and feature levels also remains limited. To address these challenges, we propose \u201cTF-DWGNet,\u201d a novel Graph Neural Network framework that combines tree-based Directed Weighted graph construction with Tensor Fusion for multiclass cancer subtype classification. TF-DWGNet introduces two key methodological innovations: (i) a supervised tree-based strategy that constructs directed weighted graphs tailored to each omics modality, and (ii) a tensor fusion mechanism that captures unimodal, bimodal, and trimodal interactions using low-rank decomposition for computational efficiency. Experiments on three real-world cancer datasets demonstrate that TF-DWGNet consistently outperforms state-of-the-art baselines across multiple metrics and statistical tests. In addition, the model provides biologically meaningful insights through modality level contribution scores and ranked feature importance. These results highlight that TF-DWGNet is an effective and interpretable solution for multi-omics integration in cancer research.<\/jats:p>","DOI":"10.1093\/nargab\/lqag054","type":"journal-article","created":{"date-parts":[[2026,5,9]],"date-time":"2026-05-09T11:41:59Z","timestamp":1778326919000},"source":"Crossref","is-referenced-by-count":1,"title":["TF-DWGNet: a directed weighted graph neural network with tensor fusion for multi-omics cancer subtype classification"],"prefix":"10.1093","volume":"8","author":[{"ORCID":"https:\/\/orcid.org\/0009-0003-3208-7999","authenticated-orcid":false,"given":"Tiantian","family":"Yang","sequence":"first","affiliation":[{"name":"Department of Mathematics and Statistical Science, University of Idaho , 875 Perimeter Drive, Moscow, ID 83844 ,","place":["United States"]},{"name":"Department of Bioinformatics and Computational Biology, University of Idaho , 875 Perimeter Drive, Moscow, ID 83844 ,","place":["United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4112-9647","authenticated-orcid":false,"given":"Zhiqian","family":"Chen","sequence":"additional","affiliation":[{"name":"Department of Computer Science and Engineering, Mississippi State University , 665 George Perry Street, Starkville, MS 39762 ,","place":["United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"286","published-online":{"date-parts":[[2026,5,30]]},"reference":[{"key":"2026053005041274600_B1","doi-asserted-by":"publisher","first-page":"1177932219899051","DOI":"10.1177\/1177932219899051","article-title":"Multi-omics data integration, interpretation, and its application","volume":"14","author":"Subramanian","year":"2020","journal-title":"Bioinform Biol Insights"},{"key":"2026053005041274600_B2","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/gm483","article-title":"The Human Genome Project: big science transforms biology and medicine","volume":"5","author":"Hood","year":"2013","journal-title":"Genome Med"},{"key":"2026053005041274600_B3","doi-asserted-by":"publisher","first-page":"1408843","DOI":"10.3389\/frai.2024.1408843","article-title":"Multimodal data integration for oncology in the era of deep neural networks: a review","volume":"7","author":"Waqas","year":"2024","journal-title":"Front Artif Intell"},{"key":"2026053005041274600_B4","doi-asserted-by":"publisher","first-page":"5","DOI":"10.1023\/A:1010933404324","article-title":"Random forests","volume":"45","author":"Breiman","year":"2001","journal-title":"Mach Learn"},{"key":"2026053005041274600_B5","doi-asserted-by":"crossref","first-page":"pp. 785","DOI":"10.1145\/2939672.2939785","article-title":"Xgboost: a scalable tree boosting system","volume-title":"Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining","author":"Chen","year":"2016"},{"key":"2026053005041274600_B6","doi-asserted-by":"publisher","first-page":"436","DOI":"10.1038\/nature14539","article-title":"Deep learning","volume":"521","author":"Lecun","year":"2015","journal-title":"Nature"},{"key":"2026053005041274600_B7","article-title":"Deep Learning","author":"Goodfellow","year":"2016"},{"key":"2026053005041274600_B8","doi-asserted-by":"crossref","first-page":"pp. 1","DOI":"10.1109\/MIUCC58832.2023.10278295","article-title":"A review of multi-omics and machine learning: advancements and applications in cancer research","volume-title":"2023 International Mobile, Intelligent, and Ubiquitous Computing Conference (MIUCC)","author":"Ayman","year":"2023"},{"key":"2026053005041274600_B9","doi-asserted-by":"publisher","first-page":"pp. 2208","DOI":"10.1109\/BIBM55620.2022.9995099","article-title":"A review of deep learning methods for multi-omics integration in precision medicine","volume-title":"2022 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)","author":"Zaghlool","year":"2022"},{"key":"2026053005041274600_B10","doi-asserted-by":"publisher","first-page":"18","DOI":"10.1109\/MSP.2017.2693418","article-title":"Geometric deep learning: going beyond euclidean data","volume":"34","author":"Bronstein","year":"2017","journal-title":"IEEE Signal Process Mag"},{"key":"2026053005041274600_B11","doi-asserted-by":"publisher","first-page":"249","DOI":"10.1109\/TKDE.2020.2981333","article-title":"Deep learning on graphs: a survey","volume":"34","author":"Zhang","year":"2020","journal-title":"IEEE T Knowl Data Eng"},{"key":"2026053005041274600_B12","doi-asserted-by":"publisher","first-page":"4","DOI":"10.1109\/TNNLS.2020.2978386","article-title":"A comprehensive survey on graph neural networks","volume":"32","author":"Wu","year":"2020","journal-title":"IEEE T Neu Netw Learn Syst"},{"key":"2026053005041274600_B13","doi-asserted-by":"publisher","first-page":"57","DOI":"10.1016\/j.aiopen.2021.01.001","article-title":"Graph neural networks: a review of methods and applications","volume":"1","author":"Zhou","year":"2020","journal-title":"AI Open"},{"key":"2026053005041274600_B14","article-title":"Semi-supervised classification with graph convolutional networks","volume-title":"International Conference on Learning Representations","author":"Kipf","year":"2017"},{"key":"2026053005041274600_B15","article-title":"Graph attention networks","volume-title":"International Conference on Learning Representations","author":"Petar","year":"2018"},{"key":"2026053005041274600_B16","article-title":"How powerful are graph neural networks?","volume-title":"International Conference on Learning Representations","author":"Xu","year":"2019"},{"key":"2026053005041274600_B17","first-page":"1025","article-title":"Inductive representation learning on large graphs","volume":"30","author":"Hamilton","year":"2017","journal-title":"Adv Neur Inf Proc Syst"},{"key":"2026053005041274600_B18","doi-asserted-by":"publisher","first-page":"690049","DOI":"10.3389\/fgene.2021.690049","article-title":"Graph neural networks and their current applications in bioinformatics","volume":"12","author":"Zhang","year":"2021","journal-title":"Front Genet"},{"key":"2026053005041274600_B19","doi-asserted-by":"publisher","first-page":"345","DOI":"10.1146\/annurev-biodatasci-110723-024625","article-title":"Graph artificial intelligence in medicine","volume":"7","author":"Johnson","year":"2024","journal-title":"Annu Rev Biomed Data Sci"},{"key":"2026053005041274600_B20","first-page":"1","article-title":"Deep multimodal data fusion","volume":"56","author":"Zhao","year":"2024","journal-title":"ACM Comput Surv"},{"key":"2026053005041274600_B21","doi-asserted-by":"publisher","first-page":"3445","DOI":"10.1038\/s41467-021-23774-w","article-title":"MOGONET integrates multi-omics data using graph convolutional networks allowing patient classification and biomarker identification","volume":"12","author":"Wang","year":"2021","journal-title":"Nat Commun"},{"key":"2026053005041274600_B22","doi-asserted-by":"publisher","first-page":"82","DOI":"10.1186\/s12911-023-02173-9","article-title":"MODILM: towards better complex diseases classification using a novel multi-omics data integration learning model","volume":"23","author":"Zhong","year":"2023","journal-title":"BMC Med Inform Decis Mak"},{"key":"2026053005041274600_B23","doi-asserted-by":"publisher","first-page":"lqad063","DOI":"10.1093\/nargab\/lqad063","article-title":"SUPREME: multiomics data integration using graph convolutional networks","volume":"5","author":"Kesimoglu","year":"2023","journal-title":"NAR Genom Bioinform"},{"key":"2026053005041274600_B24","doi-asserted-by":"publisher","first-page":"2788","DOI":"10.3390\/ijms25052788","article-title":"MOGAT: a multi-omics integration framework using graph attention networks for cancer subtype prediction","volume":"25","author":"Tanvir","year":"2024","journal-title":"Int J Mol Sci"},{"key":"2026053005041274600_B25","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/s12864-024-11112-5","article-title":"DeepMoIC: multi-omics data integration via deep graph convolutional networks for cancer subtype classification","volume":"25","author":"Wu","year":"2024","journal-title":"BMC Genom"},{"key":"2026053005041274600_B26","doi-asserted-by":"publisher","first-page":"806842","DOI":"10.3389\/fgene.2022.806842","article-title":"MoGCN: a multi-omics integration method based on graph convolutional network for cancer subtype analysis","volume":"13","author":"Li","year":"2022","journal-title":"Front Genet"},{"key":"2026053005041274600_B27","doi-asserted-by":"publisher","first-page":"3727","DOI":"10.1093\/bioinformatics\/bty429","article-title":"A graph-embedded deep feedforward network for disease outcome classification and feature selection using gene expression data","volume":"34","author":"Kong","year":"2018","journal-title":"Bioinformatics"},{"key":"2026053005041274600_B28","first-page":"pp. 1103","article-title":"Tensor fusion network for multimodal sentiment analysis","volume-title":"Proceedings of the 2017conference on empirical methods in natural language processing","author":"Zadeh","year":"2017"},{"key":"2026053005041274600_B29","doi-asserted-by":"publisher","first-page":"164","DOI":"10.1002\/sapm192761164","article-title":"The expression of a tensor or a polyadic as a sum of products","volume":"6","author":"Hitchcock","year":"1927","journal-title":"J Math Phys"},{"key":"2026053005041274600_B30","doi-asserted-by":"publisher","first-page":"283","DOI":"10.1007\/BF02310791","article-title":"Analysis of individual differences in multidimensional scaling via an N-way generalization of \u201cEckart-Young\u201d decomposition","volume":"35","author":"Carroll","year":"1970","journal-title":"Psychometrika"},{"key":"2026053005041274600_B31","first-page":"84","article-title":"Foundations of the PARAFAC procedure: models and conditions for an \u201cexplanatory\u201d multi-modal factor analysis","volume":"16","author":"Harshman","year":"1970","journal-title":"UCLA Work Pap Phonetics"},{"key":"2026053005041274600_B32","doi-asserted-by":"publisher","first-page":"455","DOI":"10.1137\/07070111X","article-title":"Tensor decompositions and applications","volume":"51","author":"Kolda","year":"2009","journal-title":"SIAM Rev"},{"key":"2026053005041274600_B33","first-page":"pp. 7","article-title":"Rank, decomposition, and uniqueness for 3-way and N-way arrays","volume-title":"Multiway Data Analysis","author":"Kruskal","year":"1989"},{"key":"2026053005041274600_B34","doi-asserted-by":"publisher","first-page":"pp. 451","DOI":"10.1007\/BFb0035776","article-title":"Tensor rank is NP-complete","volume-title":"International Colloquium on Automata, Languages, and Programming","author":"H\u00e5stad","year":"1989"},{"key":"2026053005041274600_B35","doi-asserted-by":"publisher","first-page":"193","DOI":"10.3390\/e22020193","article-title":"Deep residual learning for nonlinear regression","volume":"22","author":"Chen","year":"2020","journal-title":"Entropy"},{"key":"2026053005041274600_B36","doi-asserted-by":"publisher","first-page":"135","DOI":"10.1016\/S0304-3800(02)00064-9","article-title":"Illuminating the \u201cblack box\u201d: a randomization approach for understanding variable contributions in artificial neural networks","volume":"154","author":"Olden","year":"2002","journal-title":"Ecol Modell"},{"key":"2026053005041274600_B37","doi-asserted-by":"publisher","first-page":"3507","DOI":"10.1093\/bioinformatics\/btaa164","article-title":"forgeNet: a graph deep neural network model using tree-based ensemble classifiers for feature graph construction","volume":"36","author":"Kong","year":"2020","journal-title":"Bioinformatics"},{"key":"2026053005041274600_B38","doi-asserted-by":"publisher","article-title":"Falcon: High Performance Supercomputer","author":"Idaho C3+3 Collaboration","DOI":"10.7923\/falcon.id"},{"key":"2026053005041274600_B39","doi-asserted-by":"publisher","first-page":"bbad378","DOI":"10.1093\/bib\/bbad378","article-title":"Multiomics dynamic learning enables personalized diagnosis and prognosis for pancancer and cancer subtypes","volume":"24","author":"Lu","year":"2023","journal-title":"Brief Bioinform"},{"key":"2026053005041274600_B40","doi-asserted-by":"publisher","first-page":"357","DOI":"10.1038\/s41419-022-04813-w","article-title":"Cancer cells adapt FAM134B\/BiP mediated ER-phagy to survive hypoxic stress","volume":"13","author":"Chipurupalli","year":"2022","journal-title":"Cell Death Dis"},{"key":"2026053005041274600_B41","doi-asserted-by":"publisher","first-page":"75729","DOI":"10.18632\/oncotarget.12370","article-title":"FOXC1 identifies basal-like breast cancer in a hereditary breast cancer cohort","volume":"7","author":"Johnson","year":"2016","journal-title":"Oncotarget"}],"container-title":["NAR Genomics and Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/nargab\/article-pdf\/8\/2\/lqag054\/68435205\/lqag054.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/nargab\/article-pdf\/8\/2\/lqag054\/68435205\/lqag054.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,30]],"date-time":"2026-05-30T09:04:18Z","timestamp":1780131858000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/nargab\/article\/doi\/10.1093\/nargab\/lqag054\/8698434"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,3,27]]},"references-count":41,"journal-issue":{"issue":"2","published-print":{"date-parts":[[2026,3,27]]}},"URL":"https:\/\/doi.org\/10.1093\/nargab\/lqag054","relation":{},"ISSN":["2631-9268"],"issn-type":[{"value":"2631-9268","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2026,6]]},"published":{"date-parts":[[2026,3,27]]},"article-number":"lqag054"}}