MethylBERT enables read-level DNA methylation pattern identification and tumour deconvolution using a Transformer-based model
Abstract DNA methylation (DNAm) is a key epigenetic mark that shows profound alterations in cancer. Read-level methylomes enable more in-depth analyses, due to their broad genomic coverage and preservation of rare cell-type signals, compared to summarized data such as 450K/EPIC microarrays. Here, we...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-01-01
|
| Series: | Nature Communications |
| Online Access: | https://doi.org/10.1038/s41467-025-55920-z |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832594652825387008 |
|---|---|
| author | Yunhee Jeong Clarissa Gerhäuser Guido Sauter Thorsten Schlomm Karl Rohr Pavlo Lutsik |
| author_facet | Yunhee Jeong Clarissa Gerhäuser Guido Sauter Thorsten Schlomm Karl Rohr Pavlo Lutsik |
| author_sort | Yunhee Jeong |
| collection | DOAJ |
| description | Abstract DNA methylation (DNAm) is a key epigenetic mark that shows profound alterations in cancer. Read-level methylomes enable more in-depth analyses, due to their broad genomic coverage and preservation of rare cell-type signals, compared to summarized data such as 450K/EPIC microarrays. Here, we propose MethylBERT, a Transformer-based model for read-level methylation pattern classification. MethylBERT identifies tumour-derived sequence reads based on their methylation patterns and local genomic sequence, and estimates tumour cell fractions within bulk samples. In our evaluation, MethylBERT outperforms existing deconvolution methods and demonstrates high accuracy regardless of methylation pattern complexity, read length and read coverage. Moreover, we show its applicability to cell-type deconvolution as well as non-invasive early cancer diagnostics using liquid biopsy samples. MethylBERT represents a significant advancement in read-level methylome analysis and enables accurate tumour purity estimation. The broad applicability of MethylBERT will enhance studies on both tumour and non-cancerous bulk methylomes. |
| format | Article |
| id | doaj-art-cfa8ddeed7fa44d38f1b8984d80e3675 |
| institution | Kabale University |
| issn | 2041-1723 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Nature Communications |
| spelling | doaj-art-cfa8ddeed7fa44d38f1b8984d80e36752025-01-19T12:29:54ZengNature PortfolioNature Communications2041-17232025-01-0116111410.1038/s41467-025-55920-zMethylBERT enables read-level DNA methylation pattern identification and tumour deconvolution using a Transformer-based modelYunhee Jeong0Clarissa Gerhäuser1Guido Sauter2Thorsten Schlomm3Karl Rohr4Pavlo Lutsik5Division of Cancer Epigenomics, German Cancer Research Center (DKFZ)Division of Cancer Epigenomics, German Cancer Research Center (DKFZ)Institute for Pathology, University Medical Center Hamburg-EppendorfDepartment of Urology, Charité – Universitätsmedizin BerlinBiomedical Computer Vision Group, BioQuant, IPMB, Heidelberg UniversityDivision of Cancer Epigenomics, German Cancer Research Center (DKFZ)Abstract DNA methylation (DNAm) is a key epigenetic mark that shows profound alterations in cancer. Read-level methylomes enable more in-depth analyses, due to their broad genomic coverage and preservation of rare cell-type signals, compared to summarized data such as 450K/EPIC microarrays. Here, we propose MethylBERT, a Transformer-based model for read-level methylation pattern classification. MethylBERT identifies tumour-derived sequence reads based on their methylation patterns and local genomic sequence, and estimates tumour cell fractions within bulk samples. In our evaluation, MethylBERT outperforms existing deconvolution methods and demonstrates high accuracy regardless of methylation pattern complexity, read length and read coverage. Moreover, we show its applicability to cell-type deconvolution as well as non-invasive early cancer diagnostics using liquid biopsy samples. MethylBERT represents a significant advancement in read-level methylome analysis and enables accurate tumour purity estimation. The broad applicability of MethylBERT will enhance studies on both tumour and non-cancerous bulk methylomes.https://doi.org/10.1038/s41467-025-55920-z |
| spellingShingle | Yunhee Jeong Clarissa Gerhäuser Guido Sauter Thorsten Schlomm Karl Rohr Pavlo Lutsik MethylBERT enables read-level DNA methylation pattern identification and tumour deconvolution using a Transformer-based model Nature Communications |
| title | MethylBERT enables read-level DNA methylation pattern identification and tumour deconvolution using a Transformer-based model |
| title_full | MethylBERT enables read-level DNA methylation pattern identification and tumour deconvolution using a Transformer-based model |
| title_fullStr | MethylBERT enables read-level DNA methylation pattern identification and tumour deconvolution using a Transformer-based model |
| title_full_unstemmed | MethylBERT enables read-level DNA methylation pattern identification and tumour deconvolution using a Transformer-based model |
| title_short | MethylBERT enables read-level DNA methylation pattern identification and tumour deconvolution using a Transformer-based model |
| title_sort | methylbert enables read level dna methylation pattern identification and tumour deconvolution using a transformer based model |
| url | https://doi.org/10.1038/s41467-025-55920-z |
| work_keys_str_mv | AT yunheejeong methylbertenablesreadleveldnamethylationpatternidentificationandtumourdeconvolutionusingatransformerbasedmodel AT clarissagerhauser methylbertenablesreadleveldnamethylationpatternidentificationandtumourdeconvolutionusingatransformerbasedmodel AT guidosauter methylbertenablesreadleveldnamethylationpatternidentificationandtumourdeconvolutionusingatransformerbasedmodel AT thorstenschlomm methylbertenablesreadleveldnamethylationpatternidentificationandtumourdeconvolutionusingatransformerbasedmodel AT karlrohr methylbertenablesreadleveldnamethylationpatternidentificationandtumourdeconvolutionusingatransformerbasedmodel AT pavlolutsik methylbertenablesreadleveldnamethylationpatternidentificationandtumourdeconvolutionusingatransformerbasedmodel |