EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Extraction of biological terms using large language models enhances the usability of metadata in the BioSample database

Shuya Ikeda iD, Zhaonan Zou iD, Hidemasa Bono iD, Yuki Moriya iD, Shuichi Kawashima iD, Toshiaki Katayama iD, Shinya Oki iD, Tazro Ohta iD

GigaScience · 2025

DOI: 10.1093/gigascience/giaf070 ↗

Kurz gefasst

Fachartikel in der Zeitschrift GigaScience, 2025. Verfasst von Shuya Ikeda und 7 weiteren Autor:innen. Beteiligte Einrichtung: Database Center for Life Science, Joint Support-Center for Data Science Research, Research Organization of Information and Systems, Univ. of Tokyo Ka…. Hauptthemen laut MEDLINE: Metadata, Natural Language Processing, Databases, Factual, Computational Biology. Die Arbeit wurde bisher 22-mal zitiert (Crossref) und nennt 34 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: JST, JSPS.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

BioSample is a repository of experimental sample metadata. It is a comprehensive archive that enables searches of experiments, regardless of type. However, there is substantial variability in the submitted metadata due to the difficulty in defining comprehensive rules for describing them and the limited user awareness of best practices in creating them. This inconsistency poses considerable challenges to the findability and reusability of archived data. Given the scale of BioSample, which hosts over 40 million records, manual curation is impractical. Automatic rule-based ontology mapping methods have been proposed to address this issue, but their effectiveness is limited by the heterogeneity of the metadata. Recently, large language models (LLMs) have gained attention in natural language processing and are promising tools for automating metadata curation. In this study, we evaluated the performance of LLMs in extracting cell line names from BioSample descriptions using a gold-standard dataset derived from ChIP-Atlas, a secondary database of epigenomics experiment data in which samples were manually curated. The LLM-assisted methods outperformed traditional approaches, achieving higher accuracy and coverage. We further extended them to extract information about experimentally manipulated genes from metadata when manual curation had not yet been applied in ChIP-Atlas. This also yielded successful results, including the facilitation of more precise filtering of the data and the prevention of possible misinterpretations caused by the inclusion of unintended data. These findings underscore the potential of LLMs in improving the findability and reusability of experimental data in general, which would considerably reduce the user workload and enable more effective scientific data management.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

MetadataNatural Language ProcessingDatabases, FactualComputational BiologyHumansData CurationData MiningLarge Language Models

Stichwörter der Autor:innen

automatic data curationbiological samplelarge language model

Autor:innen und Institutionen

  • Database Center for Life Science, Joint Support-Center for Data Science Research, Research Organization of Information and Systems, Univ. of Tokyo Kashiwanoha-campus Station Satellite 6F. 178-4-4 Wakashiba, Kashiwa-shi,…
  • Graduate School of Integrated Sciences for Life, Hiroshima University, 1-4-4 Kagamiyama, Higashihiroshima-shi, Hiroshima 739-8528, JAPAN.
  • Institute of Resource Development and Analysis, Kumamoto University, Gene Technology Center 6F. 2-2-1 Honjo, Chuo-ku, Kumamoto-shi, Kumamoto 860-0811, JAPAN.
  • Genome Editing Innovation Center, Hiroshima University, Hiroshima University Innovation Plaza 3-10-23 Kagamiyama, Higashihiroshima-shi, Hiroshima 739-0046, JAPAN.
  • BioData Science Initiative, Joint Support-Center for Data Science Research, Research Organization of Information and Systems , Univ. of Tokyo Kashiwanoha-campus Station Satellite 6F. 178-4-4 Wakashiba, Kashiwa-shi, Chib…
  • Department of Artificial Intelligence Medicine, Graduate School of Medicine, Chiba University, 1-8-1 Inohana, Chuo, Chiba, Chiba 260-8670, JAPAN.

ORCID-Profile: Shuya Ikeda, Zhaonan Zou, Hidemasa Bono, Yuki Moriya, Shuichi Kawashima, Toshiaki Katayama, Shinya Oki, Tazro Ohta

Förderung

  • JST
  • JSPS

Zitieren

APA

Ikeda, S., Zou, Z., Bono, H., Moriya, Y., Kawashima, S., Katayama, T., Oki, S., & Ohta, T. (2025). Extraction of biological terms using large language models enhances the usability of metadata in the BioSample database. GigaScience. https://doi.org/10.1093/gigascience/giaf070

Vancouver

Ikeda S, Zou Z, Bono H, Moriya Y, Kawashima S, Katayama T, et al. Extraction of biological terms using large language models enhances the usability of metadata in the BioSample database. GigaScience. 2025. doi:10.1093/gigascience/giaf070

BibTeX
@article{Ikeda2025,
  author = {Ikeda, Shuya and Zou, Zhaonan and Bono, Hidemasa and Moriya, Yuki and Kawashima, Shuichi and Katayama, Toshiaki and Oki, Shinya and Ohta, Tazro},
  title = {Extraction of biological terms using large language models enhances the usability of metadata in the BioSample database},
  journal = {GigaScience},
  year = {2025},
  doi = {10.1093/gigascience/giaf070}
}
Kostenlos starten →