EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A token-pruning framework enables efficient representation of the human genome for RNA modification analysis

Wenjia Gao, Junlei Yu, Junru Jin, Jiajie Cai, Ke Qiu, Shun Zhang, Jianbo Qiao iD, Leyi Wei iD

Bioinformatics · 2026

DOI: 10.1093/bioinformatics/btag621 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Bioinformatics, 2026. Verfasst von Wenjia Gao und 7 weiteren Autor:innen. Beteiligte Einrichtung: School of Software, Shandong University, Jinan, China. Hauptthemen laut MEDLINE: Genome, Human, RNA, Genomics, Software. Die Arbeit nennt 48 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Motivation Modelling long genomic sequences remains challenging due to extreme sequence length, high redundancy, and the need for biological interpretability. Although Transformer-based architectures have achieved strong performance across genomic tasks, their high computational cost and reliance on fixed tokenization strategies limit their scalability and ability to focus on biologically informative regions. Results We propose ATSFormer, a token-pruning Transformer framework for efficient and biologically informed genomic sequence modelling. ATSFormer incorporates an attention-guided and parameter-free Adaptive Token Sampling (ATS) module into Transformer layers. Guided by attention-derived importance scores, ATS dynamically retains informative tokens while probabilistically discarding redundant ones, thereby reducing sequence length, FLOPs, and memory usage without introducing additional learnable parameters or extra training procedures. Importantly, the retained tokens correspond to key contributors to model predictions, enabling ATSFormer to highlight biologically meaningful sites and sequence motifs. We evaluated ATSFormer on four benchmark RNA modification datasets derived from RMVar 2.0, covering A-to-I, m1A, m5C, and m7G. Experimental results show that ATSFormer consistently outperforms existing state-of-the-art methods while achieving substantial computational savings. Furthermore, structural analysis using AlphaFold3 supports the biological relevance of the motifs identified by ATSFormer. Availability and implementation The source data and code are freely available at GitHub (https://github.com/1gao2/ATSFormer) and Zenodo (https://doi.org/10.5281/zenodo.21813541).

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Genome, HumanRNAGenomicsSoftwareSequence Analysis, RNARNA Processing, Post-TranscriptionalComputational BiologyHumansAlgorithms

Substanzen

RNA

Autor:innen und Institutionen

  • School of Software, Shandong University, Jinan, China.
  • Engineering Research Centre of Applied Technology on Machine Translation and Artificial Intelligence, Macao Polytechnic University, Macao SAR, China.
  • Joint SDU-NTU Centre for Artificial Intelligence Research (C-FAIR), Shandong University, Jinan, China.

ORCID-Profile: Jianbo Qiao, Leyi Wei

Förderung

  • National Natural Science Foundation of China

Zitieren

APA

Gao, W., Yu, J., Jin, J., Cai, J., Qiu, K., Zhang, S., Qiao, J., & Wei, L. (2026). A token-pruning framework enables efficient representation of the human genome for RNA modification analysis. Bioinformatics. https://doi.org/10.1093/bioinformatics/btag621

Vancouver

Gao W, Yu J, Jin J, Cai J, Qiu K, Zhang S, et al. A token-pruning framework enables efficient representation of the human genome for RNA modification analysis. Bioinformatics. 2026. doi:10.1093/bioinformatics/btag621

BibTeX
@article{Gao2026,
  author = {Gao, Wenjia and Yu, Junlei and Jin, Junru and Cai, Jiajie and Qiu, Ke and Zhang, Shun and Qiao, Jianbo and Wei, Leyi},
  title = {A token-pruning framework enables efficient representation of the human genome for RNA modification analysis},
  journal = {Bioinformatics},
  year = {2026},
  doi = {10.1093/bioinformatics/btag621}
}
Kostenlos starten →