Highly biased DNA sequence reconstruction in DNA storage with multi-scale attention mechanism and contrast learning
Synthetic and Systems Biotechnology · 2026-06 · 12 · 422-432
DOI: 10.1016/j.synbio.2026.01.028 ↗Kurz gefasst
Fachartikel in der Zeitschrift Synthetic and Systems Biotechnology, 2026-06. Verfasst von Xue Li und 8 weiteren Autor:innen. Beteiligte Einrichtung: Key Laboratory of Advanced Design and Intelligent Computing, Ministry of Education, School of Software Engineering, Dalian University, Dalian 116622,…. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 55 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: Overseas Expertise Introduction Center for Discipline Innovation of Food Nutrition and Human Health, National Natural Science Foundation of China, Department of Education of Liaoning Province.
Abstract (Originalsprache)
Base substitution, insertion, and deletion errors due to inherent technical constraints inducing unavoidable sequencing inaccuracies, limiting access to high-quality raw data and biological knowledge. To address this, we propose a deep sequence reconstruction model based on the multi-scale attention mechanism and contrastive learning (MACL), designed to enhance DNA sequence reconstruction under highly error rate conditions. The multi-scale attention mechanism includes base scale, inter-sequence and intra-sequence scale. First, the MSA Transformer fully extracts both global and local features of the base scale from the dimensions of the rows and columns. Furthermore, for the errors between sequences and the substitution errors within sequences, MACL proposes Inter-Sequence and Intra-Sequence Multi-Head Attention Mechanisms, respectively, and handles the insertion and deletion errors through the convolution module. In order to maximize the consistency of positive sample pairs in the representation space, we introduce contrastive learning and design a negative sample construction method and data augmentation that are more suitable for substitution errors in sequencing channels. Experiments on real-world DNA storage and viral genome datasets demonstrate that MACL significantly outperforms existing methods in reconstructing the DNA sequence. In particular, when combined with RS codes, MACL can losslessly reconstruct medical images in highly biased sequence (base error rate = 5%) in DNA storage. In summary, the MACL introduces a novel approach to DNA sequence reconstruction in highly error rate conditions, laying a solid foundation for practical applications in DNA storage and genomics research.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Key Laboratory of Advanced Design and Intelligent Computing, Ministry of Education, School of Software Engineering, Dalian University, Dalian 116622, China.
- School of Computer Science and Technology, Dalian University of Technology, Dalian 116024, China.
- Key Laboratory of Social Computing and Cognitive Intelligence (Dalian University of Technology), Ministry of Education, Dalian 116024, China.
- School of Computing, Engineering and the Built Environment (CEBE), University of Roehampton, London, 133884, UK.
- Department of Accounting and Information Systems, University of Canterbury, Christchurch 8140, New Zealand.
Förderung
- Overseas Expertise Introduction Center for Discipline Innovation of Food Nutrition and Human Health
- National Natural Science Foundation of China
- Department of Education of Liaoning Province
- Liaoning Provincial Natural Science Foundation
Zitieren
APA
Li, X., Zheng, Y., Shao, Q., Wang, J., Li, W., Wang, B., Zhou, S., Cao, B., & Zheng, P. (2026). Highly biased DNA sequence reconstruction in DNA storage with multi-scale attention mechanism and contrast learning. Synthetic and Systems Biotechnology, 12, 422-432. https://doi.org/10.1016/j.synbio.2026.01.028
Vancouver
Li X, Zheng Y, Shao Q, Wang J, Li W, Wang B, et al. Highly biased DNA sequence reconstruction in DNA storage with multi-scale attention mechanism and contrast learning. Synthetic and Systems Biotechnology. 2026;12:422-432. doi:10.1016/j.synbio.2026.01.028
BibTeX
@article{Li2026,
author = {Li, Xue and Zheng, Yanfen and Shao, Qi and Wang, Jiadong and Li, Wei and Wang, Bin and Zhou, Shihua and Cao, Ben and Zheng, Pan},
title = {Highly biased DNA sequence reconstruction in DNA storage with multi-scale attention mechanism and contrast learning},
journal = {Synthetic and Systems Biotechnology},
year = {2026},
volume = {12},
pages = {422--432},
doi = {10.1016/j.synbio.2026.01.028}
}
EUVIMED