{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,15]],"date-time":"2026-07-15T07:06:02Z","timestamp":1784099162357,"version":"3.55.0"},"reference-count":32,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,7,15]],"date-time":"2026-07-15T00:00:00Z","timestamp":1784073600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Artif. Intell."],"abstract":"<jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Cancer recurrence and distant metastasis are major causes of cancer-related death, yet existing biomarkers and single-omics models have limited accuracy and interpretability across tumor types.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>\n                      We developed OMNIS (OMics Network Integration and Spatial representation), a convolutional deep-learning framework that embeds multi-omics profiles into a five-channel genomic image ordered by Hi-C\u2013derived chromosomal proximity. Somatic mutation, copy-number alteration, DNA methylation and gene-expression data from 1,578 TCGA tumors across 33 cancer types were used to train classifiers for recurrence risk and for primary-versus-metastatic status. Performance was assessed by 10-fold cross-validation using AUROC, AUPR and threshold-based metrics. Integrated gradients yielded per-gene attribution scores; top-ranked genes were evaluated for prognostic value in two independent non-small cell lung cancer cohorts (GSE31210,\n                      <jats:italic>n<\/jats:italic>\n                      \u202f=\u202f226; GSE135222,\n                      <jats:italic>n<\/jats:italic>\n                      \u202f=\u202f27) using survival analyses.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>OMNIS achieved high discrimination for recurrence (AUROC\/AUPR 0.970\/0.937) and metastasis (0.980\/0.883), with accuracies of 0.873\u20130.911 and negative predictive values \u22650.970 across tasks. Spatial genomic embedding accelerated convergence and outperformed non-spatial baselines. Attribution highlighted seven recurrence-associated genes (including IBA57, DNTTIP1, SLC20A2 and TMEM201) and ten metastasis-associated genes (including PLXNA1, POLR3D, TTLL4, SREBF2, TYMP and ZBTB7C). In external cohorts, expression of these genes showed independent, stage-dependent associations with progression-free and overall survival.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>OMNIS is a spatially informed multi-omics framework that couples accurate prediction with gene-level interpretability. By embedding three-dimensional genome organization into deep-learning models, OMNIS nominates biologically coherent, context-specific drivers of progression and may guide future biomarker development and personalized therapy in precision oncology.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/frai.2026.1817043","type":"journal-article","created":{"date-parts":[[2026,7,15]],"date-time":"2026-07-15T06:11:09Z","timestamp":1784095869000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["OMNIS: a spatially informed multi-omics deep-learning framework for tumor recurrence prediction and primary\u2013metastatic tumor differentiation title page"],"prefix":"10.3389","volume":"9","author":[{"given":"Junxian","family":"Li","sequence":"first","affiliation":[{"name":"Department of Blood Transfusion, Key Laboratory of Cancer Prevention and Therapy, Tianjin, Tianjin\u2019s Clinical Research Center for Cancer, Tianjin Medical University Cancer Institute & Hospital, National Clinical Research Center for Cancer, Tianjin Medical University","place":["Tianjin, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yuchen","family":"Xing","sequence":"additional","affiliation":[{"name":"Department of Cancer Epidemiology and Biostatistics, Key Laboratory of Molecular Cancer Epidemiology, Tianjin, Tianjin\u2019s Clinical Research Center for Cancer, Tianjin Medical University Cancer Institute & Hospital, National Clinical Research Center for Cancer, Tianjin Medical University","place":["Tianjin, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ximin","family":"Gao","sequence":"additional","affiliation":[{"name":"Department of Cancer Epidemiology and Biostatistics, Key Laboratory of Molecular Cancer Epidemiology, Tianjin, Tianjin\u2019s Clinical Research Center for Cancer, Tianjin Medical University Cancer Institute & Hospital, National Clinical Research Center for Cancer, Tianjin Medical University","place":["Tianjin, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Renhe","family":"Liu","sequence":"additional","affiliation":[{"name":"Department of Diagnostic and Therapeutic Ultrasonography, Key Laboratory of Cancer Prevention and Therapy, Tianjin, Tianjin\u2019s Clinical Research Center for Cancer, Tianjin Medical University Cancer Institute & Hospital, National Clinical Research Center for Cancer, Tianjin Medical University","place":["Tianjin, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,7,15]]},"reference":[{"key":"ref1","doi-asserted-by":"publisher","first-page":"248","DOI":"10.1038\/nmeth0410-248","article-title":"A method and server for predicting damaging missense mutations","volume":"7","author":"Adzhubei","year":"2010","journal-title":"Nat. Methods"},{"key":"ref2","doi-asserted-by":"publisher","first-page":"111","DOI":"10.1186\/s13059-020-02015-1","article-title":"MOFA+: a statistical framework for comprehensive integration of multi-modal single-cell data","volume":"21","author":"Argelaguet","year":"2020","journal-title":"Genome Biol."},{"key":"ref3","doi-asserted-by":"publisher","first-page":"205","DOI":"10.1186\/s13046-020-01712-w","article-title":"Tubulin tyrosine ligase like 4 (TTLL4) overexpression in breast cancer cells is associated with brain metastasis and alters exosome biogenesis","volume":"39","author":"Arnold","year":"2020","journal-title":"J. Exp. Clin. Cancer Res."},{"key":"ref4","doi-asserted-by":"publisher","first-page":"118","DOI":"10.1038\/s41746-020-00324-0","article-title":"The state of artificial intelligence-based FDA-approved medical devices and algorithms: an online database","volume":"3","author":"Benjamens","year":"2020","journal-title":"NPJ Digital Medicine."},{"key":"ref5","doi-asserted-by":"publisher","first-page":"289","DOI":"10.1111\/j.2517-6161.1995.tb02031.x","article-title":"Controlling the false discovery rate: a practical and powerful approach to multiple testing","volume":"57","author":"Benjamini","year":"1995","journal-title":"J. Royal Stat. Soc. Series B Stat. Methodol."},{"key":"ref6","doi-asserted-by":"publisher","first-page":"2711","DOI":"10.3389\/fimmu.2018.02711","article-title":"Exosomes and miRNA-loaded biomimetic Nanovehicles, a focus on their potentials preventing type-2 diabetes linked to metabolic syndrome","volume":"9","author":"Beuzelin","year":"2018","journal-title":"Front. Immunol."},{"key":"ref7","doi-asserted-by":"publisher","first-page":"10285","DOI":"10.3390\/ijms241210285","article-title":"Characterization of the clinical significance and immunological landscapes of a novel TMEMs signature in hepatocellular carcinoma and the contribution of TMEM201 to Hepatocarcinogenesis","volume":"24","author":"Chen","year":"2023","journal-title":"Int. J. Mol. Sci."},{"key":"ref8","author":"Despraz","year":"2017"},{"key":"ref9","doi-asserted-by":"publisher","first-page":"104100","DOI":"10.1016\/j.ebiom.2022.104100","article-title":"DNTTIP1 promotes nasopharyngeal carcinoma metastasis via recruiting HDAC1 to DUSP2 promoter and activating ERK signaling pathway","volume":"81","author":"Ding","year":"2022","journal-title":"EBioMedicine"},{"key":"ref10","doi-asserted-by":"publisher","first-page":"265","DOI":"10.1038\/nature19800","article-title":"Formation of new chromatin domains determines pathogenicity of genomic duplications","volume":"538","author":"Franke","year":"2016","journal-title":"Nature"},{"key":"ref11","doi-asserted-by":"publisher","first-page":"1664","DOI":"10.1038\/s41588-019-0538-0","article-title":"Activity-by-contact model of enhancer\u2013promoter regulation from thousands of CRISPR perturbations","volume":"51","author":"Fulco","year":"2019","journal-title":"Nat. Genet."},{"key":"ref12","doi-asserted-by":"publisher","first-page":"101047","DOI":"10.1016\/j.neo.2024.101047","article-title":"PLXNA1 confers enzalutamide resistance in prostate cancer via AKT signaling pathway","volume":"57","author":"Hu","year":"2024","journal-title":"Neoplasia"},{"key":"ref13","doi-asserted-by":"publisher","first-page":"1137","DOI":"10.1158\/0008-5472.CAN-11-2433","article-title":"KR-POK interacts with p53 and represses its ability to activate transcription of p21WAF1\/CDKN1A","volume":"72","author":"Jeon","year":"2012","journal-title":"Cancer Res."},{"key":"ref14","doi-asserted-by":"publisher","first-page":"1932948","DOI":"10.1155\/2020\/1932948","article-title":"The SLC family are candidate diagnostic and prognostic biomarkers in clear cell renal cell carcinoma","volume":"2020","author":"Kang","year":"2020","journal-title":"Biomed. Res. Int."},{"key":"ref15","doi-asserted-by":"publisher","first-page":"110493","DOI":"10.1016\/j.celrep.2022.110493","article-title":"CRISPR-Cas9 screen identifies oxidative phosphorylation as essential for cancer cell survival at low extracellular pH","volume":"38","author":"Michl","year":"2022","journal-title":"Cell Rep."},{"key":"ref16","doi-asserted-by":"publisher","first-page":"15","DOI":"10.1186\/s13148-020-0807-x","article-title":"A DNA methylation signature to improve survival prediction of gastric cancer","volume":"12","author":"Peng","year":"2020","journal-title":"Clin. Epigenetics"},{"key":"ref17","doi-asserted-by":"publisher","first-page":"3735","DOI":"10.1016\/j.csbj.2021.06.030","article-title":"Integration strategies of multi-omics data for machine learning analysis","volume":"19","author":"Picard","year":"2021","journal-title":"Comput. Struct. Biotechnol. J."},{"key":"ref18","doi-asserted-by":"publisher","first-page":"1591","DOI":"10.1002\/cpbi.17","article-title":"AscatNgs: identifying somatically acquired copy-number alterations from whole-genome sequencing data","volume":"56","author":"Raine","year":"2016","journal-title":"Curr. Protoc. Bioinform."},{"key":"ref19","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-030-28954-6","volume-title":"Explainable AI: Interpreting, Explaining and Visualizing Deep Learning","author":"Samek GM","year":"2019"},{"key":"ref20","doi-asserted-by":"publisher","first-page":"54","DOI":"10.1007\/s41109-021-00396-1","article-title":"Hi\u2013C interaction graph analysis reveals the impact of histone modifications in chromatin shape","volume":"6","author":"Sefer","year":"2021","journal-title":"Appl. Netw. Sci."},{"key":"ref21","doi-asserted-by":"publisher","first-page":"287","DOI":"10.1186\/s12864-022-08498-5","article-title":"ProbC: joint modeling of epigenome and transcriptome effects in 3D genome","volume":"23","author":"Sefer","year":"2022","journal-title":"BMC Genomics"},{"key":"ref22","doi-asserted-by":"publisher","first-page":"425","DOI":"10.1089\/cmb.2015.0210","article-title":"Deconvolution of ensemble chromatin interaction data reveals the latent mixing structures in cell subpopulations","volume":"23","author":"Sefer","year":"2016","journal-title":"J. Comput. Biol."},{"key":"ref23","doi-asserted-by":"publisher","first-page":"2906","DOI":"10.1093\/bioinformatics\/btp543","article-title":"Integrative clustering of multiple genomic data types using a joint latent variable model with application to breast and lung cancer subtype analysis","volume":"25","author":"Shen","year":"2009","journal-title":"Bioinformatics"},{"key":"ref24","author":"Shrikumar","year":"2017"},{"key":"ref25","doi-asserted-by":"publisher","first-page":"17","DOI":"10.3322\/caac.21763","article-title":"Cancer statistics, 2023","volume":"73","author":"Siegel","year":"2023","journal-title":"CA Cancer J. Clin."},{"key":"ref26","author":"Sundararajan","year":"2017"},{"key":"ref27","doi-asserted-by":"publisher","first-page":"314","DOI":"10.1038\/nbt.3772","article-title":"Toil enables reproducible, open source, big biomedical data analyses","volume":"35","author":"Vivian","year":"2017","journal-title":"Nat. Biotechnol."},{"key":"ref28","doi-asserted-by":"publisher","first-page":"e2305212","DOI":"10.1002\/advs.202305212","article-title":"Nanoparticle targeting in chemo-resistant ovarian cancer reveals dual axis of therapeutic vulnerability involving cholesterol uptake and cell redox balance","volume":"11","author":"Wang","year":"2024","journal-title":"Adv. Sci."},{"key":"ref29","doi-asserted-by":"publisher","first-page":"421","DOI":"10.1038\/s41420-023-01714-5","article-title":"USP35 promotes HCC development by stabilizing ABHD17C and activating the PI3K\/AKT signaling pathway","volume":"9","author":"Wang","year":"2023","journal-title":"Cell Death Discov."},{"key":"ref30","doi-asserted-by":"publisher","first-page":"29","DOI":"10.1016\/j.inffus.2021.07.016","article-title":"Unbox the black-box for the medical explainable AI via multi-modal and multi-Centre data fusion: a mini-review, two showcases and beyond","volume":"77","author":"Yang","year":"2022","journal-title":"Inf. Fusion"},{"key":"ref31","doi-asserted-by":"publisher","first-page":"107966","DOI":"10.1016\/j.cmpb.2023.107966","article-title":"SurvIAE: survival prediction with interpretable autoencoders from diffuse large B-cells lymphoma gene expression data","volume":"244","author":"Zaccaria","year":"2024","journal-title":"Comput. Methods Prog. Biomed."},{"key":"ref32","doi-asserted-by":"publisher","first-page":"10005","DOI":"10.1038\/s41467-024-54352-5","article-title":"Proteomic profiling of gliomas unveils immune and metabolism-driven subtypes with implications for anti-nucleotide metabolism therapy","volume":"15","author":"Zhang","year":"2024","journal-title":"Nat. Commun."}],"container-title":["Frontiers in Artificial Intelligence"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2026.1817043\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,15]],"date-time":"2026-07-15T06:11:10Z","timestamp":1784095870000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2026.1817043\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,15]]},"references-count":32,"alternative-id":["10.3389\/frai.2026.1817043"],"URL":"https:\/\/doi.org\/10.3389\/frai.2026.1817043","relation":{},"ISSN":["2624-8212"],"issn-type":[{"value":"2624-8212","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,15]]},"article-number":"1817043"}}