EUVIMEDEUVIMED entdecken
Open Access

GenNA: Conditional generation of nucleotide sequences guided by natural language annotations

Yi Shen iD, Guangshuo Cao iD, Jianghong Wu iD, Dijun Chen iD, Cong Feng iD, Ming Chen iD

Science Advances · 20.11.2026 · 12 (47)

DOI: 10.1126/sciadv.aej1481 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Science Advances, erschienen am 20.11.2026. Verfasst von Yi Shen und 5 weiteren Autor:innen. Beteiligte Einrichtung: Department of Bioinformatics, College of Life Sciences, Zhejiang University, Hangzhou 310058, China. Hauptthemen laut MEDLINE: Molecular Sequence Annotation, Natural Language Processing, Nucleotides, Computational Biology. Die Arbeit nennt 71 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: National Natural Science Foundation of China, National Key Research and Development Program of China, Key R&D Program of Ningbo.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Deciphering the mapping between biomolecular sequences and complex biological functions is a central challenge in genomics. While existing generative nucleotide models capture sequence distributions, they lack explicit access to gene-associated biological semantics, limiting semantics-guided design. Here, we present GenNA, a generative nucleotide foundation model guided by natural language annotations. Pretrained on a multimodal nucleotide-text corpus spanning 2221 eukaryotic species (∼416 billion characters), GenNA internalizes relationships between sequence syntax and functional semantics within a unified autoregressive framework. Without explicit rule-based supervision, GenNA yields distinguishable perplexity responses to sequence-annotation mismatches, mutational variants, and species-label perturbations. Furthermore, across diverse generation tasks, GenNA produces statistically plausible nucleotide sequences consistent with target semantics, structural constraints, and species contexts. Overall, GenNA provides a versatile framework for controllable nucleotide generation, establishing a practical route to integrate functional descriptions with sequence design.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Molecular Sequence AnnotationNatural Language ProcessingNucleotidesComputational BiologyBase SequenceGenomicsHumansSemanticsGenerative Artificial Intelligence

Substanzen

Nucleotides

Autor:innen und Institutionen

  • Department of Bioinformatics, College of Life Sciences, Zhejiang University, Hangzhou 310058, China.
  • State Key Laboratory of Pharmaceutical Biotechnology, School of Life Sciences, Nanjing University, Nanjing 210023, China.
  • College of Animal Science and Technology, Inner Mongolia Minzu University, Tongliao 028000, China.
  • Zhejiang Key Laboratory of Multi-omics Precision Diagnosis and Treatment of Liver Diseases, Department of General Surgery, Sir Run-Run Shaw Hospital, Zhejiang University School of Medicine, Hangzhou 310016, China.
  • State Key Laboratory of Vegetation Structure, Function and Construction, College of Life Sciences, Zhejiang University, Hangzhou 310058, China.

ORCID-Profile: Yi Shen, Guangshuo Cao, Jianghong Wu, Dijun Chen, Cong Feng, Ming Chen

Förderung

  • National Natural Science Foundation of China
  • National Key Research and Development Program of China
  • Key R&D Program of Ningbo
  • Science and Technology Innovation Leader of Zhejiang Province

Zitieren

APA

Shen, Y., Cao, G., Wu, J., Chen, D., Feng, C., & Chen, M. (2026). GenNA: Conditional generation of nucleotide sequences guided by natural language annotations. Science Advances, 12(47). https://doi.org/10.1126/sciadv.aej1481

Vancouver

Shen Y, Cao G, Wu J, Chen D, Feng C, Chen M. GenNA: Conditional generation of nucleotide sequences guided by natural language annotations. Science Advances. 2026;12(47). doi:10.1126/sciadv.aej1481

BibTeX
@article{Shen2026,
  author = {Shen, Yi and Cao, Guangshuo and Wu, Jianghong and Chen, Dijun and Feng, Cong and Chen, Ming},
  title = {GenNA: Conditional generation of nucleotide sequences guided by natural language annotations},
  journal = {Science Advances},
  year = {2026},
  volume = {12},
  number = {47},
  doi = {10.1126/sciadv.aej1481}
}
Kostenlos starten →