{"database":"biostudies-literature","file_versions":[],"scores":null,"additional":{"submitter":["Yuan D"],"funding":["Brain Tumour Charity","Ministerstvo Zdravotnictví Ceské Republiky (Ministry of Health of the Czech Republic)"],"pagination":["1283-1294"],"full_dataset_link":["https://www.ebi.ac.uk/biostudies/studies/S-EPMC12296554"],"repository":["biostudies-literature"],"omics_type":["Unknown"],"volume":["6(7)"],"pubmed_abstract":["DNA methylation-based classification of (brain) tumors has emerged as a powerful and indispensable diagnostic technique. Initial implementations used methylation microarrays for data generation, while most current classifiers rely on a fixed methylation feature space. This makes them incompatible with other platforms, especially different flavors of DNA sequencing. Here, we describe crossNN, a neural network-based machine learning framework that can accurately classify tumors using sparse methylomes obtained on different platforms and with different epigenome coverage and sequencing depth. It outperforms other deep and conventional machine learning models regarding accuracy and computational requirements while still being explainable. We use crossNN to train a pan-cancer classifier that ca"],"journal":["Nature cancer"],"pubmed_title":["crossNN is an explainable framework for cross-platform DNA methylation-based classification of tumors."],"pmcid":["PMC12296554"],"funding_grant_id":["GN-000694","NV19-03-00562"],"pubmed_authors":["Thomas C","Eils R","Reimann R","Halldorsson S","Jugas R","Mackowiak S","Pokorna P","Siewert C","Lukassen S","Harter PN","Weber KJ","Ishaque N","Zeiner PS","Appelt A","Capper D","Yuan D","Schmid S","Rechsteiner M","Sterba J","Euskirchen P","Osberg B","Schuller U","Slaby O","Albers A","Vik-Mo EO","Jabareen N"],"additional_accession":[]},"is_claimable":false,"name":"crossNN is an explainable framework for cross-platform DNA methylation-based classification of tumors.","description":"DNA methylation-based classification of (brain) tumors has emerged as a powerful and indispensable diagnostic technique. Initial implementations used methylation microarrays for data generation, while most current classifiers rely on a fixed methylation feature space. This makes them incompatible with other platforms, especially different flavors of DNA sequencing. Here, we describe crossNN, a neural network-based machine learning framework that can accurately classify tumors using sparse methylomes obtained on different platforms and with different epigenome coverage and sequencing depth. It outperforms other deep and conventional machine learning models regarding accuracy and computational requirements while still being explainable. We use crossNN to train a pan-cancer classifier that ca","dates":{"release":"2025-01-01T00:00:00Z","publication":"2025 Jul","modification":"2026-03-15T16:38:32.427Z","creation":"2025-08-13T03:04:43.558Z"},"accession":"S-EPMC12296554","cross_references":{"pubmed":["40481322"],"doi":["10.1038/s43018-025-00976-5"]}}