{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,12,8]],"date-time":"2025-12-08T17:19:59Z","timestamp":1765214399009,"version":"3.46.0"},"reference-count":40,"publisher":"Oxford University Press (OUP)","issue":"6","license":[{"start":{"date-parts":[[2025,12,8]],"date-time":"2025-12-08T00:00:00Z","timestamp":1765152000000},"content-version":"vor","delay-in-days":37,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62272172","62266012"],"award-info":[{"award-number":["62272172","62266012"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Zhuhai Science and Technology Plan Project","award":["2320004002758"],"award-info":[{"award-number":["2320004002758"]}]},{"DOI":"10.13039\/501100012226","name":"Fundamental Research Funds for the Central Universities","doi-asserted-by":"publisher","award":["2025ZYGXZR095"],"award-info":[{"award-number":["2025ZYGXZR095"]}],"id":[{"id":"10.13039\/501100012226","id-type":"DOI","asserted-by":"publisher"}]},{"name":"High-Level Innovative Talent Project of Guizhou Province","award":["QKHPTRC-GCC2023027"],"award-info":[{"award-number":["QKHPTRC-GCC2023027"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2025,11,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Accurately identifying DNA methylation is essential for understanding complex regulatory networks and disease mechanisms. However, the dynamic nature of methylation and species differences make prediction challenging. Existing deep learning methods often overlook the potential of shared features across diverse species\u2019 methylation sequences and rely solely on token-to-token attention when modelling long-range dependencies, limiting the model\u2019s representation capabilities. To address these limitations, we propose iDNA-DAPHA, an accurate and generic two-stage deep learning framework that leverages domain-adaptive pretraining (DAP) incorporating feature alignment to learn common features across various types of methylation sequences from multiple species, followed by fine-tuning to capture task-specific features. The framework further introduces hierarchical attention (HA) to enhance its representational power. Experimental results demonstrate that iDNA-DAPHA performs better than existing state-of-the-art methods across seventeen benchmark datasets covering three representative DNA methylation types. Ablation studies validate the effectiveness and contributions of DAP and HA. Furthermore, visualization-based analyses reveal that the model can capture conserved sequence patterns and learn discriminative representations. We believe that iDNA-DAPHA will serve as a valuable framework for methylation prediction, especially in scenarios with limited training samples for specific methylation types in certain species.<\/jats:p>","DOI":"10.1093\/bib\/bbaf642","type":"journal-article","created":{"date-parts":[[2025,11,26]],"date-time":"2025-11-26T13:01:45Z","timestamp":1764162105000},"source":"Crossref","is-referenced-by-count":0,"title":["iDNA-DAPHA: a generic framework for methylation prediction via domain-adaptive pretraining and hierarchical attention"],"prefix":"10.1093","volume":"26","author":[{"given":"Wenjun","family":"Wang","sequence":"first","affiliation":[{"name":"School of Software Engineering , South China University of Technology, Guangzhou Higher Education Mega Centre, Panyu District, Guangzhou, Guangdong 510006,","place":["China"]},{"name":"School of Data Science and Information Engineering , Guizhou Minzu University, Huaxi University Town, Huaxi District, Guiyang, Guizhou 550025,","place":["China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Wen","family":"Tan","sequence":"additional","affiliation":[{"name":"School of Software Engineering , South China University of Technology, Guangzhou Higher Education Mega Centre, Panyu District, Guangzhou, Guangdong 510006,","place":["China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Lvlong","family":"Lai","sequence":"additional","affiliation":[{"name":"School of Computer Science , Guangdong Polytechnic Normal University, No. 293, West Zhongshan Avenue, Tianhe District, Guangzhou, Guangdong 510665,","place":["China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Youjun","family":"Lu","sequence":"additional","affiliation":[{"name":"School of Data Science and Information Engineering , Guizhou Minzu University, Huaxi University Town, Huaxi District, Guiyang, Guizhou 550025,","place":["China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Qingyao","family":"Wu","sequence":"additional","affiliation":[{"name":"School of Software Engineering , South China University of Technology, Guangzhou Higher Education Mega Centre, Panyu District, Guangzhou, Guangdong 510006,","place":["China"]},{"name":"Peng Cheng Laboratory , No. 6001 Shahexi Road, Nanshan District, Shenzhen, Guangdong 518055,","place":["China"]}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"286","published-online":{"date-parts":[[2025,12,8]]},"reference":[{"key":"2025120812171400200_ref1","doi-asserted-by":"publisher","first-page":"giad054","DOI":"10.1093\/gigascience\/giad054","article-title":"MuLan-methyl\u2014Multiple transformer-based language models for accurate DNA methylation prediction","volume":"12","author":"Zeng","year":"2023","journal-title":"GigaScience"},{"key":"2025120812171400200_ref2","doi-asserted-by":"publisher","first-page":"409","DOI":"10.1016\/B978-0-12-803309-8.00024-0","article-title":"Chapter 24\u2014role of DNA methylation in genome stability","volume-title":"Genome Stability","author":"Zhou","year":"2016"},{"key":"2025120812171400200_ref3","doi-asserted-by":"crossref","first-page":"4603","DOI":"10.1093\/bioinformatics\/btab677","article-title":"iDNA-ABT: advanced deep learning model for detecting DNA methylation with adaptive features and transductive information maximization","volume":"37","author":"Yingying","year":"2021","journal-title":"Bioinformatics"},{"key":"2025120812171400200_ref4","doi-asserted-by":"publisher","first-page":"100991","DOI":"10.1016\/j.isci.2020.100991","article-title":"iDNA-MS: an integrated computational tool for detecting DNA modification sites in multiple genomes","volume":"23","author":"Lv","year":"2020","journal-title":"Iscience"},{"key":"2025120812171400200_ref5","doi-asserted-by":"crossref","first-page":"8694","DOI":"10.1021\/jacs.9b02512","article-title":"Jump-seq: genome-wide capture and amplification of 5-hydroxymethylcytosine sites","volume":"141","author":"Lulu","year":"2019","journal-title":"J Am Chem Soc"},{"key":"2025120812171400200_ref6","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/s12864-019-5754-6","article-title":"Sources of artifact in measurements of 6mA and 4mC abundance in eukaryotic genomic DNA","volume":"20","author":"O\u2019Brown","year":"2019","journal-title":"BMC Genomics"},{"key":"2025120812171400200_ref7","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/s12864-020-07033-8","article-title":"Accurate prediction of DNA N4-methylcytosine sites via boost-learning various types of sequence features","volume":"21","author":"Zhao","year":"2020","journal-title":"BMC Genomics"},{"key":"2025120812171400200_ref8","doi-asserted-by":"publisher","first-page":"77","DOI":"10.2217\/epi-2016-0122","article-title":"The application of genome-wide 5-hydroxymethylcytosine studies in cancer research","volume":"9","author":"Thomson","year":"2017","journal-title":"Epigenomics"},{"key":"2025120812171400200_ref9","doi-asserted-by":"publisher","first-page":"710","DOI":"10.1016\/j.cell.2015.04.021","article-title":"An adenine code for DNA: a second life for N6-methyladenine","volume":"161","author":"Heyn","year":"2015","journal-title":"Cell"},{"key":"2025120812171400200_ref10","doi-asserted-by":"publisher","first-page":"879","DOI":"10.1016\/j.cell.2015.04.010","article-title":"N6-methyldeoxyadenosine marks active transcription start sites in chlamydomonas","volume":"161","author":"Ye","year":"2015","journal-title":"Cell"},{"key":"2025120812171400200_ref11","doi-asserted-by":"publisher","first-page":"742","DOI":"10.1186\/s12864-023-09802-7","article-title":"StableDNAm: towards a stable and efficient model for predicting DNA methylation based on adaptive feature correction learning","volume":"24","author":"Zhuo","year":"2023","journal-title":"BMC Genomics"},{"key":"2025120812171400200_ref12","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1186\/s13059-022-02780-1","article-title":"iDNA-ABF: multi-scale deep biological language learning model for the interpretable prediction of DNA methylations","volume":"23","author":"Jin","year":"2022","journal-title":"Genome Biol"},{"key":"2025120812171400200_ref13","doi-asserted-by":"publisher","first-page":"5868","DOI":"10.1093\/nar\/gki901","article-title":"Reduced representation bisulfite sequencing for comparative high-resolution DNA methylation analysis","volume":"33","author":"Meissner","year":"2005","journal-title":"Nucleic Acids Res"},{"key":"2025120812171400200_ref14","doi-asserted-by":"publisher","first-page":"315","DOI":"10.1038\/nature08514","article-title":"Human DNA methylomes at base resolution show widespread epigenomic differences","volume":"462","author":"Lister","year":"2009","journal-title":"Nature"},{"key":"2025120812171400200_ref15","doi-asserted-by":"publisher","first-page":"3327","DOI":"10.1093\/bioinformatics\/btaa143","article-title":"DNA4mC-LIP: a linear integration method to identify N4-methylcytosine site in multiple species","volume":"36","author":"Tang","year":"2020","journal-title":"Bioinformatics"},{"key":"2025120812171400200_ref16","doi-asserted-by":"crossref","first-page":"bbaa124","DOI":"10.1093\/bib\/bbaa124","article-title":"DeepTorrent: a deep learning-based approach for predicting DNA N4-methylcytosine sites","volume":"22","author":"Liu","year":"2021","journal-title":"Brief Bioinform"},{"key":"2025120812171400200_ref17","doi-asserted-by":"crossref","first-page":"btae528","DOI":"10.1093\/bioinformatics\/btae528","article-title":"Deep5hmC: predicting genome-wide 5-hydroxymethylcytosine landscape via a multimodal deep learning model","volume":"40","author":"Ma","year":"2024","journal-title":"Bioinformatics"},{"key":"2025120812171400200_ref18","doi-asserted-by":"publisher","first-page":"8491","DOI":"10.1109\/ACCESS.2021.3049146","article-title":"Prediction of RNA 5-hydroxymethylcytosine modifications using deep learning","volume":"9","author":"Ali","year":"2021","journal-title":"IEEE Access"},{"key":"2025120812171400200_ref19","doi-asserted-by":"publisher","first-page":"2796","DOI":"10.1093\/bioinformatics\/btz015","article-title":"i6mA-Pred: identifying DNA N6-methyladenine sites in the rice genome","volume":"35","author":"Chen","year":"2019","journal-title":"Bioinformatics"},{"key":"2025120812171400200_ref20","doi-asserted-by":"crossref","first-page":"1071","DOI":"10.3389\/fgene.2019.01071","article-title":"SNNRice6mA: a deep learning method for predicting DNA N6-methyladenine sites in rice genome","volume":"10","author":"Haitao","year":"2019","journal-title":"Front Genet"},{"key":"2025120812171400200_ref21","doi-asserted-by":"publisher","first-page":"131","DOI":"10.1016\/j.omtn.2019.08.011","article-title":"SDM6A: a web-based integrative machine-learning framework for predicting 6mA sites in the rice genome","volume":"18","author":"Basith","year":"2019","journal-title":"Mol Therapy-Nucleic Acids"},{"key":"2025120812171400200_ref22"},{"key":"2025120812171400200_ref23","doi-asserted-by":"publisher","first-page":"1234","DOI":"10.1093\/bioinformatics\/btz682","article-title":"BioBERT: a pre-trained biomedical language representation model for biomedical text mining","volume":"36","author":"Lee","year":"2020","journal-title":"Bioinformatics"},{"key":"2025120812171400200_ref24"},{"key":"2025120812171400200_ref25"},{"key":"2025120812171400200_ref26","doi-asserted-by":"publisher","first-page":"101914","DOI":"10.1016\/j.artmed.2020.101914","article-title":"Multi-stage domain-specific pretraining for improved detection and localization of barrett\u2019s neoplasia: a comprehensive clinically validated study","volume":"107","author":"van der Putten","year":"2020","journal-title":"Artif Intell Med"},{"key":"2025120812171400200_ref27"},{"key":"2025120812171400200_ref28"},{"key":"2025120812171400200_ref29"},{"key":"2025120812171400200_ref30"},{"key":"2025120812171400200_ref31","doi-asserted-by":"crossref","first-page":"109588","DOI":"10.1016\/j.patcog.2023.109588","article-title":"MoCA: incorporating domain pretraining and cross attention for textbook question answering","volume":"140","author":"Fangzhi","year":"2023","journal-title":"Pattern Recogn"},{"key":"2025120812171400200_ref32"},{"key":"2025120812171400200_ref33","doi-asserted-by":"publisher","first-page":"7838","DOI":"10.1038\/s41467-024-52293-7","article-title":"Improving prediction performance of general protein language model by domain-adaptive pretraining on DNA-binding protein","volume":"15","author":"Zeng","year":"2024","journal-title":"Nat Commun"},{"key":"2025120812171400200_ref34","doi-asserted-by":"publisher","first-page":"4214","DOI":"10.1016\/j.csbj.2024.11.006","article-title":"iResNetDM: an interpretable deep learning approach for four types of DNA methylation modification prediction","volume":"23","author":"Yang","year":"2024","journal-title":"Comput Struct Biotechnol J"},{"key":"2025120812171400200_ref35","doi-asserted-by":"publisher","DOI":"10.1101\/2025.02.05.636730"},{"key":"2025120812171400200_ref36","doi-asserted-by":"publisher","first-page":"102103","DOI":"10.1016\/j.omtn.2023.102103","article-title":"Joint masking and self-supervised strategies for inferring small molecule-miRNA associations","volume":"35","author":"Zhou","year":"2024","journal-title":"Mol Therapy Nucleic Acids"},{"key":"2025120812171400200_ref37"},{"key":"2025120812171400200_ref38","journal-title":"Proceedings of the 31st International Conference on Neural Information Processing Systems"},{"key":"2025120812171400200_ref39","doi-asserted-by":"publisher","first-page":"2112","DOI":"10.1093\/bioinformatics\/btab083","article-title":"DNABERT: pre-trained bidirectional encoder representations from transformers model for DNA-language in genome","volume":"37","author":"Ji","year":"2021","journal-title":"Bioinformatics"},{"key":"2025120812171400200_ref40","doi-asserted-by":"publisher","first-page":"5631","DOI":"10.1109\/JBHI.2025.3546345","article-title":"EMGANet: edge-aware multi-scale group-mix attention network for breast cancer ultrasound image segmentation","volume":"29","author":"Huang","year":"2025","journal-title":"IEEE J Biomed Health Inform"}],"container-title":["Briefings in Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/bib\/article-pdf\/26\/6\/bbaf642\/65802351\/bbaf642.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bib\/article-pdf\/26\/6\/bbaf642\/65802351\/bbaf642.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,12,8]],"date-time":"2025-12-08T17:17:19Z","timestamp":1765214239000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bib\/article\/doi\/10.1093\/bib\/bbaf642\/8374030"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,11,1]]},"references-count":40,"journal-issue":{"issue":"6","published-print":{"date-parts":[[2025,11,1]]}},"URL":"https:\/\/doi.org\/10.1093\/bib\/bbaf642","relation":{},"ISSN":["1467-5463","1477-4054"],"issn-type":[{"value":"1467-5463","type":"print"},{"value":"1477-4054","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2025,11]]},"published":{"date-parts":[[2025,11,1]]},"article-number":"bbaf642"}}