crossNN is an explainable framework for cross-platform DNA methylation-based classification of tumors

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 can discriminate more than 170 tumor types across all organ sites. Validation in more than 5,000 tumors profiled on different platforms, including nanopore and targeted bisulfite sequencing, demonstrates its robustness and scalability with 99.1% and 97.8% precision for the brain tumor and pan-cancer models, respectively.

Authors

Institutions

Publication Details

Journal
Nature Cancer
Published
2025-06-06
DOI
https://doi.org/10.1038/s43018-025-00976-5
Citations
9
Primary Topic
Epigenetics and DNA Methylation
Type
article
Field-Weighted Citation Impact
7.25

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

crossNN is an explainable framework for cross-platform DNA methylation-based classification of tumors

Skarphéðinn Halldórsson, Patrick N. Harter, Soeren Lukassen, Philipp Euskirchen et al.
9 citations
Nature Cancer
Epigenetics and DNA Methylation
7.25
article

crossNN is an explainable framework for cross-platform DNA methylation-based classification of tumors

Skarphéðinn Halldórsson, Patrick N. Harter, Soeren Lukassen, Philipp Euskirchen, Christian Thomas, Einar Osland Vik-Mo, Pia S. Zeiner, Markus Rechsteiner, David Capper, Christin Siewert, Roland Eils, Naveed Ishaque, Simone Schmid, Sebastian D. Mackowiak, Ulrich Schüller, Brendan Osberg, Anne Albers, Regina Reimann, Katharina Johanna Weber, Nabil Jabareen, Dongsheng Yuan, Robin Jugas, Petra Pokorná, Jaroslav Štěrba, Anton Appelt
article en
9 citations

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 can discriminate more than 170 tumor types across all organ sites. Validation in more than 5,000 tumors profiled on different platforms, including nanopore and targeted bisulfite sequencing, demonstrates its robustness and scalability with 99.1% and 97.8% precision for the brain tumor and pan-cancer models, respectively.

Nature CancerVol. 6(7)
Goethe University Frankfurt (DE), Oslo University Hospital (NO), Central European Institute of Technology (CZ), Universität Hamburg (DE), German Cancer Research Center (DE), University of Oslo (NO), University of Zurich (CH), Masaryk University (CZ), Heidelberg University (DE), University of Münster (DE), LMU Klinikum (DE), Humboldt-Universität zu Berlin (DE), Central European Institute of Technology – Masaryk University (CZ), University Hospital of Zurich (CH), University Hospital Brno (CZ), University Medical Center Hamburg-Eppendorf (DE), Kinderkrebs-Zentrum Hamburg (DE), University Hospital Frankfurt (DE), Berlin Institute of Health at Charité - Universitätsmedizin Berlin (DE), Frankfurt Cancer Institute (DE), Freie Universität Berlin (DE), Charité - Universitätsmedizin Berlin (DE), Ludwig-Maximilians-Universität München (DE)
Deutsches Krebsforschungszentrum, European Commission, Deutsche Forschungsgemeinschaft, Ministerstvo Zdravotnictví Ceské Republiky, Westfälische Wilhelms-Universität Münster, Deutsche Krebshilfe, Humboldt-Universität zu Berlin, Universität Zürich, Deutsche Kinderkrebsstiftung, Freie Universität Berlin, Deutschen Konsortium für Translationale Krebsforschung, Berlin Institute of Health, Universitätsklinikum Hamburg-Eppendorf
Reduced inequalities
Openalex Percentile: Top 3%
Epigenetics and DNA Methylation
7.25
DOIOpen Access PDFOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.