EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

K-attention: a biologically informed attention operator for data-efficient sequence-based omics modeling

Tao Liu, Jing-Yi Li, Ziyu Chen, Chan Gu, Tong Zhou iD, Shi-Qi Yang, Yang Ding iD, Ge Gao iD

Briefings in Bioinformatics · 03.07.2026 · 27 (4)

DOI: 10.1093/bib/bbag351 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Briefings in Bioinformatics, online erschienen am 07.07.2026. Verfasst von Tao Liu und 7 weiteren Autor:innen. Beteiligte Einrichtung: College of Life Sciences, Beijing Normal University, 19 Xinjiekouwai Avenue, Beijing 100875, China. Hauptthemen laut MEDLINE: Genomics, Deep Learning, Computational Biology. Die Arbeit nennt 54 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: National Science and Technology Major Project, State Key Laboratory of Gene Function and Modulation Research, Beijing Advanced Innovation Center for Genomics at Peking University.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Deep learning-based modeling of omics data often suffers from insufficiency and heterogeneity of the data itself. As a step towards addressing these issues, we present K-attention, a biologically informed operator that models interactions between sequence fragments effectively and efficiently. Across both biologically informed simulated datasets and two real-world omics tasks, K-attention-based networks consistently outperform canonical convolutional neural network (CNN)- and Transformer-based models, with the largest gains observed in low-data regimes. Collectively, these results indicate that K-attention enables data-efficient and biologically grounded modeling under real-world constraints.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

GenomicsDeep LearningComputational BiologyConvolutional Neural NetworksAlgorithmsHumans

Stichwörter der Autor:innen

computational genomicsdata-efficiencysequence pattern recognition

Autor:innen und Institutionen

  • College of Life Sciences, Beijing Normal University, 19 Xinjiekouwai Avenue, Beijing 100875, China.
  • Changping Laboratory, No. 28 Kexueyuan Road, Changping District, Beijing 102206, China.
  • State Key Laboratory of Gene Function and Modulation Research, School of Life Sciences, Biomedical Pioneering Innovative Center (BIOPIC) and Beijing Advanced Innovation Center for Genomics (ICG), Center for Bioinformati…

ORCID-Profile: Tong Zhou, Yang Ding, Ge Gao

Förderung

  • National Science and Technology Major Project
  • State Key Laboratory of Gene Function and Modulation Research
  • Beijing Advanced Innovation Center for Genomics at Peking University

Zitieren

APA

Liu, T., Li, J. Y., Chen, Z., Gu, C., Zhou, T., Yang, S. Q., Ding, Y., & Gao, G. (2026). K-attention: a biologically informed attention operator for data-efficient sequence-based omics modeling. Briefings in Bioinformatics, 27(4). https://doi.org/10.1093/bib/bbag351

Vancouver

Liu T, Li JY, Chen Z, Gu C, Zhou T, Yang SQ, et al. K-attention: a biologically informed attention operator for data-efficient sequence-based omics modeling. Briefings in Bioinformatics. 2026;27(4). doi:10.1093/bib/bbag351

BibTeX
@article{Liu2026,
  author = {Liu, Tao and Li, Jing-Yi and Chen, Ziyu and Gu, Chan and Zhou, Tong and Yang, Shi-Qi and Ding, Yang and Gao, Ge},
  title = {K-attention: a biologically informed attention operator for data-efficient sequence-based omics modeling},
  journal = {Briefings in Bioinformatics},
  year = {2026},
  volume = {27},
  number = {4},
  doi = {10.1093/bib/bbag351}
}
Kostenlos starten →