DNABERT-S: pioneering species differentiation with species-aware DNA embeddings
Bioinformatics · 01.07.2025 · 41 (Supplement_1) · i255-i264
DOI: 10.1093/bioinformatics/btaf188 ↗Kurz gefasst
Fachartikel in der Zeitschrift Bioinformatics, online erschienen am 15.07.2025. Verfasst von Zhihan Zhou und 7 weiteren Autor:innen. Beteiligte Einrichtung: Department of Computer Science, Northwestern University, Evanston, IL 60208, United States. Hauptthemen laut MEDLINE: Sequence Analysis, DNA, DNA, Genomics. Die Arbeit wurde bisher 43-mal zitiert (Crossref) und nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NSF, NIH HHS, Department of Energy.
Abstract (Originalsprache)
Summary We introduce DNABERT-S, a tailored genome model that develops species-aware embeddings to naturally cluster and segregate DNA sequences of different species in the embedding space. Differentiating species from genomic sequences (i.e. DNA and RNA) is vital yet challenging, since many real-world species remain uncharacterized, lacking known genomes for reference. Embedding-based methods are therefore used to differentiate species in an unsupervised manner. DNABERT-S builds upon a pre-trained genome foundation model named DNABERT-2. To encourage effective embeddings to error-prone long-read DNA sequences, we introduce Manifold Instance Mixup (MI-Mix), a contrastive objective that mixes the hidden representations of DNA sequences at randomly selected layers and trains the model to recognize and differentiate these mixed proportions at the output layer. We further enhance it with the proposed Curriculum Contrastive Learning (C2LR) strategy. Empirical results on 28 diverse datasets show DNABERT-S’s effectiveness, especially in realistic label-scarce scenarios. For example, it identifies twice more species from a mixture of unlabeled genomic sequences, doubles the Adjusted Rand Index (ARI) in species clustering, and outperforms the top baseline’s performance in 10-shot species classification with just a 2-shot training. Availability and implementation Model, codes, and data are publically available at https://github.com/MAGICS-LAB/DNABERT_S.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Substanzen
Autor:innen und Institutionen
- Department of Computer Science, Northwestern University, Evanston, IL 60208, United States.
- Center for Foundation Models and Generative AI, Northwestern University, Evanston, IL 60208, United States.
- School of Natural Science, University of California at Merced, Merced, CA 95343, United States.
- Department of Energy Joint Genome Institute, Lawrence Berkeley National Laboratory, Berkeley, CA 94720, United States.
- Department of Statistics and Data Science, Northwestern University, Evanston, IL 60208, United States.
- Department of Biomedical Informatics, Stony Brook University, Stony Brook, NY 11794, United States.
ORCID-Profile: Zhihan Zhou, Weimin Wu, Harrison Ho, Lizhen Shi, Ramana V Davuluri, Zhong Wang, Han Liu
Förderung
- NSF
- NIH HHS
- Department of Energy
- DOE Office of Science User Facility
- Office of Science
- NLM NIH HHS
- Department of Energy Joint Genome Institute
- NIH
Zitieren
APA
Zhou, Z., Wu, W., Ho, H., Wang, J., Shi, L., Davuluri, R. V., Wang, Z., & Liu, H. (2025). DNABERT-S: pioneering species differentiation with species-aware DNA embeddings. Bioinformatics, 41(Supplement_1), i255-i264. https://doi.org/10.1093/bioinformatics/btaf188
Vancouver
Zhou Z, Wu W, Ho H, Wang J, Shi L, Davuluri RV, et al. DNABERT-S: pioneering species differentiation with species-aware DNA embeddings. Bioinformatics. 2025;41(Supplement_1):i255-i264. doi:10.1093/bioinformatics/btaf188
BibTeX
@article{Zhou2025,
author = {Zhou, Zhihan and Wu, Weimin and Ho, Harrison and Wang, Jiayi and Shi, Lizhen and Davuluri, Ramana V and Wang, Zhong and Liu, Han},
title = {DNABERT-S: pioneering species differentiation with species-aware DNA embeddings},
journal = {Bioinformatics},
year = {2025},
volume = {41},
number = {Supplement_1},
pages = {i255--i264},
doi = {10.1093/bioinformatics/btaf188}
}
EUVIMED