[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-356157-105":59,"doc-detail-356157-en":131},{"code":4,"msg":5,"data":6},0,"success",[7,13,18,23,28,33,38,43,48,51,55],{"id":8,"doc_module":4,"doc_module_name":9,"category_name":10,"show_sort_weight":11,"slug":12},1,"Document","Story & Novel",90,"story-novel",{"id":14,"doc_module":4,"doc_module_name":9,"category_name":15,"show_sort_weight":16,"slug":17},2,"Literature",80,"literature",{"id":19,"doc_module":4,"doc_module_name":9,"category_name":20,"show_sort_weight":21,"slug":22},4,"Exam",70,"exam",{"id":24,"doc_module":4,"doc_module_name":9,"category_name":25,"show_sort_weight":26,"slug":27},5,"Comic",60,"comic",{"id":29,"doc_module":4,"doc_module_name":9,"category_name":30,"show_sort_weight":31,"slug":32},6,"Technology",50,"technology",{"id":34,"doc_module":4,"doc_module_name":9,"category_name":35,"show_sort_weight":36,"slug":37},7,"Healthcare",40,"healthcare",{"id":39,"doc_module":4,"doc_module_name":9,"category_name":40,"show_sort_weight":41,"slug":42},8,"Research & Report",30,"research-report",{"id":44,"doc_module":4,"doc_module_name":9,"category_name":45,"show_sort_weight":46,"slug":47},9,"Religion & Spirituality",20,"religion-spirituality",{"id":46,"doc_module":4,"doc_module_name":9,"category_name":49,"show_sort_weight":46,"slug":50},"World Cup","world-cup",{"id":52,"doc_module":4,"doc_module_name":9,"category_name":53,"show_sort_weight":52,"slug":54},10,"Lifestyle","lifestyle",{"id":56,"doc_module":4,"doc_module_name":9,"category_name":57,"show_sort_weight":24,"slug":58},19,"General","general",{"code":4,"msg":60,"data":61},"ok",{"site_id":62,"language":63,"slug":64,"title":65,"keywords":66,"description":67,"schema_data":68,"social_meta":124,"head_meta":126,"extra_data":128,"updated_unix":130},105,"en","dnambert-a-transformer-based-model-for-non-invasive-cancer-diagnosis-using-dna-sequence-and-methylation-data","DNAmBERT - a transformer-based model for non-invasive cancer diagnosis using DNA sequence and methylation data","","DNA methylation alterations are early and stable hallmarks of cancer, enabling promising biomarkers for non-invasive detection via circulating cell-free DNA (cfDNA). Current computational methods often treat DNA sequence and methylation separately and fail to model complex read-level methylation patterns in heterogeneous, low-signal liquid biopsy data. DNAmBERT presents a transformer framework jointly learning sequence context and methylation haplotype structure using unified k-mer and haplotype token representations with masked language modelling.",{"@graph":69,"@context":123},[70,84,106],{"@type":71,"itemListElement":72},"BreadcrumbList",[73,77,79,82],{"item":74,"name":75,"@type":76,"position":8},"https://docshare.wps.com","Home","ListItem",{"item":78,"name":9,"@type":76,"position":14},"https://docshare.wps.com/document/",{"item":80,"name":40,"@type":76,"position":81},"https://docshare.wps.com/document/research-report/",3,{"item":83,"name":65,"@type":76,"position":19},"https://docshare.wps.com/document/dnambert-a-transformer-based-model-for-non-invasive-cancer-diagnosis-using-dna-sequence-and-methylation-data/356157/",{"url":83,"name":65,"@type":85,"image":86,"author":91,"headline":65,"publisher":94,"fileFormat":97,"inLanguage":63,"description":67,"dateModified":98,"datePublished":99,"encodingFormat":97,"isAccessibleForFree":100,"interactionStatistic":101},"DigitalDocument",{"url":87,"@type":88,"width":89,"height":90},"https://docshare.wps.com/thumbnails/dnambert-a-transformer-based-model-for-non-invasive-cancer-diagnosis-using-dna-sequence-and-methylation-data/356157.png","ImageObject",300,407,{"name":92,"@type":93},"anakgang17","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-11-02","2026-09-23",true,{"@type":102,"interactionType":103,"userInteractionCount":105},"InteractionCounter",{"@type":104},"ViewAction",12,{"@type":107,"mainEntity":108},"FAQPage",[109,115,119],{"name":110,"@type":111,"acceptedAnswer":112},"What problem does DNAmBERT address in non-invasive cancer detection?","Question",{"text":113,"@type":114},"It addresses the limitation of current methods that model DNA sequence and methylation separately and cannot capture complex read-level methylation architecture in heterogeneous, low-signal cfDNA data.","Answer",{"name":116,"@type":111,"acceptedAnswer":117},"How does DNAmBERT represent and learn from DNA sequence and methylation data?",{"text":118,"@type":114},"It integrates k-mer–encoded DNA sequences with methylation haplotype tokens using a unified representation and trains with a masked language modelling objective to learn sequence–epigenetic dependencies via self-attention.",{"name":120,"@type":111,"acceptedAnswer":121},"What performance did DNAmBERT achieve across platforms and cancer types?",{"text":122,"@type":114},"In binary classification tasks, it reached high performance across RRBS, cfRRBS, and cfMethyl-seq platforms, with AUC up to 0.99–1.00, and it outperformed conventional machine learning and existing deep learning approaches.","https://schema.org",{"og:url":83,"og:type":125,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":127,"canonical":83},"index,follow",{"doc_id":129,"site_id":62},356157,1790321320,{"code":4,"msg":5,"data":132},{"doc_id":129,"user_id":133,"nickname":92,"user_avatar":134,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":135,"file_id":136,"file_url":137,"file_type":138,"file_size":139,"view_count":105,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":140,"language":141,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":142,"faqs":143,"seo_title":144,"seo_description":67,"update_tm":145,"read_time":146,"preview_page_count":19,"preview_urls":147},962090883568,"https://ap-avatar.wpscdn.com/davatar_a8503ba1806abce46bf441b54a3ca4cd","Brieﬁngs in Bioinformatics, 2026, 27, bbag455 [https://doi.org/10.1093/bib/bbag455](https://doi.org/10.1093/bib/bbag455)  \nPublished: 02 September 2026   \nProblem Solving Protocol  \nDNAmBERT: a transformer-based model for non-invasive cancer diagnosis using DNA sequence and methylation data  \nMaryam Yassi1 , 2 , Mark Ezegbogu2 , Euan J. Rodger2 , Peter Stockwell2 , Aniruddha Chatterjee2 , 3 ,‡, Matthew Parry1 , 4 , *,‡  \n1 Department of Mathematics and Statistics, University of Otago, Dunedin 9016, New Zealand  \n2 Department of Pathology, Dunedin School of Medicine, University of Otago, Dunedin 9016, New Zealand  \n3 Honorary Professor, UPES University, Dehradun 248007, India  \n4Te P¯unaha Matatini Centre of Research Excellence, University of Auckland, Auckland 1010, New Zealand  \n*Corresponding author. Department of Mathematics and Statistics, University of Otago, Dunedin 9016, [New Zealand. E-mail:](New Zealand. E-mail: matthew.parry@otago.ac.nz)[ matthew.parry@otago.ac.nz](New Zealand. E-mail: matthew.parry@otago.ac.nz)[ ](New Zealand. E-mail: matthew.parry@otago.ac.nz)‡Aniruddha Chatterjee and Matthew Parry joint senior authors.  \n\n| Abstract\u003Cbr>DNA methylation alterations are early and stable hallmarks of cancer and represent promising biomarkers for non-invasive detection using circulating cell-free DNA (cfDNA) . However, current computational approaches often model DNA sequence and methylation features separately and struggle to capture complex read-level methylation architecture in heterogeneous, low-signal liquid biopsy data. Here, we present DNAmBERT, a Transformer-based deep learning framework designed to jointly model DNA sequence context and read-level methylation haplotype structure from cfDNA methylation sequencing data. DNAmBERT integrates k-mer–encoded DNA sequences with methylation haplotype tokens using a unified representation and masked language modelling objective, enabling context-aware learning of sequence–epigenetic dependencies through self-attention. We evaluated DNAmBERT across multiple cfDNA methylation platforms (RRBS, cfRRBS, and cfMethyl-seq) and cancer types, including colorectal cancer, lung adenocarcinoma and hepatocellular carcinoma. In binary classification tasks, the model achieved high performance across platforms (AUC up to 0.99–1.00) and outperformed conventional machine learning and existing deep learning approaches. Aggregation of read-level predictions enabled quantitative tumour probability estimation at the sample level. Beyond binary detection, DNAmBERT supported multi-cancer and stage-aware classification, including early-stage disease, with multiclass AUC values up to 0.99. The framework further demonstrated effective cross-cancer transfer learning, maintaining robust performance under limited data availability. These results indicate that integrated sequence–haplotype representation learning provides an accurate and scalable approach for cfDNA-based multi-cancer detection.\u003Cbr>Keywords DNA methylation, circulating cell-free DNA (cfDNA), transformer models, non-invasive cancer detection, multi-cancer classification, transfer learning\u003Cbr>Introduction\u003Cbr>transcriptional silencing, genomic instability, and oncogene activa- |  |\n| --- | --- |\n| Cancer remains a leading cause of mortality worldwide, with early | tion. Importantly, these alterations arise early during tumorigenesis |\n| detection playing a crucial role in improving patient outcomes [1] . | and persist throughout disease progression, making DNA methylation |\n| Tumour development is driven by the accumulation of genetic and | an attractive biomarker for cancer detection, prognosis, and disease |\n| epigenetic alterations that disrupt normal cellular processes, including | monitoring [2, 5] . |\n| growth, differentiation, and apoptosis. While genetic mutations have | Historically, cancer methylation landscapes have been charac- |\n| long been studied, it is now well established that epigenetic dysregula- | terised using tumour","cbCaitbe3Lrbg0JO","https://ap.wps.com/l/cbCaitbe3Lrbg0JO","pdf",1950630,17,"English","# Abstract\n## Introduction\n## Problem Solving Protocol\n## Method: DNAmBERT\n## Evaluation and Results\n## Conclusion\n## Keywords","[{\"question\":\"What problem does DNAmBERT address in non-invasive cancer detection?\",\"answer\":\"It addresses the limitation of current methods that model DNA sequence and methylation separately and cannot capture complex read-level methylation architecture in heterogeneous, low-signal cfDNA data.\"},{\"question\":\"How does DNAmBERT represent and learn from DNA sequence and methylation data?\",\"answer\":\"It integrates k-mer–encoded DNA sequences with methylation haplotype tokens using a unified representation and trains with a masked language modelling objective to learn sequence–epigenetic dependencies via self-attention.\"},{\"question\":\"What performance did DNAmBERT achieve across platforms and cancer types?\",\"answer\":\"In binary classification tasks, it reached high performance across RRBS, cfRRBS, and cfMethyl-seq platforms, with AUC up to 0.99–1.00, and it outperformed conventional machine learning and existing deep learning approaches.\"}]","DNAmBERT - a transformer-based model for non-invasive cancer diagnosis using DNA sequence and methylation data | PDF",1790123698,43,[148,152,156,160],{"page_no":8,"url":149,"width":150,"height":151},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/35d4a29d3f345693dcab6e52a2a0d159.webp",983,1291,{"page_no":14,"url":153,"width":154,"height":155},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/e5a05d5b2da309c9c030fd47ee5eca80.webp",982,1290,{"page_no":81,"url":157,"width":158,"height":159},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/0b93dbd8d3eaa10fff23a4519c556466.webp",981,1289,{"page_no":19,"url":161,"width":162,"height":163},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/680e0451a378624f1e7766ec47d97e76.webp",980,1287]