EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

DT-m6A: A DenseNet–Transformer Hybrid Framework for Accurate Prediction of m6A Modification Sites across Diverse Cell Lines and Tissues

Qiyu Tao, Jianhua Jia

Frontiers in Bioscience-Landmark · 2026

DOI: 10.31083/fbl48029 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Frontiers in Bioscience-Landmark, 2026. Verfasst von Qiyu Tao und 1 weiteren Autor:innen. Beteiligte Einrichtung: School of Information Engineering, Jingdezhen Ceramic University, 333403 Jingdezhen, Jiangxi, China. Hauptthemen laut MEDLINE: Adenosine, Deep Learning, RNA. Die Arbeit nennt 52 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Scientific Research Plan of the Department of Education of Jiangxi Province.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

N6-methyladenosine (m6A) RNA methylation is a crucial epigenetic modification that plays an essential role in regulating diverse biological processes. Accurate identification of m6A sites is therefore fundamental to understanding its regulatory mechanisms. In this study, we proposed DT-m6A, a novel deep learning framework that integrates DenseNet and Transformer architectures for accurate m6A site identification across diverse cell lines and tissues.

Methods

RNA sequences are first encoded using nucleotide chemical properties (NCP) for initial features extraction, after which DenseNet captures and reuses local sequence features through dense connections. The Transformer module then models long-range dependencies and extracts nonlinear representations, in which Batch Normalization replaces the conventional Layer Normalization in both sublayers to enhance training stability. Finally, a fully connected layer predicts m6A modification sites.

Results

Evaluated on 11 independent test sets spanning eight cell lines and three tissue types, DT-m6A demonstrated robust performance, achieving average accuracy (ACC) of 76.97%, Matthews correlation coefficient (MCC) of 54.27%, precision (PRE) of 75.18%, recall (REC) of 79.76%, and F1 score of 77.26%.

Conclusions

DT-m6A surpassed the state-of-the-art method MST-m6A by 0.63% in average accuracy (p = 0.0023) and 1.4% in mean MCC (p = 0.0012) across 11 independent test sets. Although its performance on the CD8T and MOLM13 cell lines was comparable to MST-m6A, DT-m6A consistently achieved superior results across all other cell lines and tissues. Overall, DT-m6A effectively captures both local patterns and global dependencies in RNA sequences, improving prediction performance across diverse biological contexts.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

AdenosineDeep LearningRNARNA MethylationHumansCell LineAnimals

Stichwörter der Autor:innen

DenseNetN6-methyladenosinedeep learningm6A site identificationtransformer

Substanzen

N-methyladenosineAdenosineRNA

Autor:innen und Institutionen

  • School of Information Engineering, Jingdezhen Ceramic University, 333403 Jingdezhen, Jiangxi, China.

Förderung

  • Scientific Research Plan of the Department of Education of Jiangxi Province

Zitieren

APA

Tao, Q., & Jia, J. (2026). DT-m6A: A DenseNet–Transformer Hybrid Framework for Accurate Prediction of m6A Modification Sites across Diverse Cell Lines and Tissues. Frontiers in Bioscience-Landmark. https://doi.org/10.31083/fbl48029

Vancouver

Tao Q, Jia J. DT-m6A: A DenseNet–Transformer Hybrid Framework for Accurate Prediction of m6A Modification Sites across Diverse Cell Lines and Tissues. Frontiers in Bioscience-Landmark. 2026. doi:10.31083/fbl48029

BibTeX
@article{Tao2026,
  author = {Tao, Qiyu and Jia, Jianhua},
  title = {DT-m6A: A DenseNet–Transformer Hybrid Framework for Accurate Prediction of m6A Modification Sites across Diverse Cell Lines and Tissues},
  journal = {Frontiers in Bioscience-Landmark},
  year = {2026},
  doi = {10.31083/fbl48029}
}
Kostenlos starten →