EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Bac2Feature: an easy-to-use interface to predict prokaryotic traits from 16S rRNA gene sequences

Masaki Fujiyoshi iD, Takao K Suzuki, Chikara Furusawa iD, Motomu Matsui iD

Bioinformatics Advances · 07.01.2026 · 6 (1)

DOI: 10.1093/bioadv/vbaf136 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Bioinformatics Advances, online erschienen am 14.06.2025. Verfasst von Masaki Fujiyoshi und 3 weiteren Autor:innen. Beteiligte Einrichtung: Department of Integrated Biosciences, Graduate School of Frontier Sciences, The University of Tokyo, Kashiwa, Chiba 277-8561, Japan. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 52 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: JSPS KAKENHI, JST DICP, GteX.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Motivation Annotating physiological, morphological, or genomic features to diverse bacterial and archaeal species is essential for the trait-based interpretations of complex microbiomes in community ecology. One promising way for high-throughput trait annotation is taxonomic marker gene-based trait prediction, where traits are predicted from 16S rRNA gene sequences based on their homology, taxonomy, or phylogeny. Many tools have been developed using different prediction methods and trait datasets; however, this situation lacks a unified framework of different prediction methods and their systematic evaluation, resulting in low accessibility to preferred methods. Results Here, we proposed Bac2Feature, an easy-to-use web tool that integrates different prediction methods in standardized trait dataset and systematically evaluated the prediction performance. We found that the prediction method based on 16S rRNA gene phylogenetic tree displayed superior prediction accuracy than that of other methods. The prediction accuracy of each trait was strongly associated with its phylogenetic signal. Furthermore, we set prediction thresholds for phylogenetic distance to avoid spurious trait prediction. Applied to the infant gut microbiome, Bac2Feature could reproduce trait-based shifts during primary succession. Our unified framework widens access to the trait-based approach and its applications to other fields, such as medical research. Availability and implementation Bac2Feature is available at https://github.com/fuyo780/Bac2Feature (command line interface) and https://www.genome.jp/bac2feature/ (web interface).

Autor:innen und Institutionen

  • Department of Integrated Biosciences, Graduate School of Frontier Sciences, The University of Tokyo, Kashiwa, Chiba 277-8561, Japan.
  • Department of Biochemistry and Systems Biomedicine, Juntendo University Graduate School of Medicine, 2-2-1, Hongo, Bunkyo-ku, Tokyo 113-8421, Japan.
  • Universal Biology Institute, Graduate School of Science, The University of Tokyo, Tokyo 113-0033, Japan.
  • Center for Biosystems Dynamics Research (BDR), RIKEN, Kobe, Hyogo 650-0047, Japan.
  • Institute for Chemical Research, Kyoto University, Uji, Kyoto 611-0011, Japan.

ORCID-Profile: Masaki Fujiyoshi, Chikara Furusawa, Motomu Matsui

Förderung

  • JSPS KAKENHI
  • JST DICP
  • GteX
  • World-Leading Innovative Graduate Study Program in Proactive Environmental Studies
  • The University of Tokyo

Zitieren

APA

Fujiyoshi, M., Suzuki, T. K., Furusawa, C., & Matsui, M. (2026). Bac2Feature: an easy-to-use interface to predict prokaryotic traits from 16S rRNA gene sequences. Bioinformatics Advances, 6(1). https://doi.org/10.1093/bioadv/vbaf136

Vancouver

Fujiyoshi M, Suzuki TK, Furusawa C, Matsui M. Bac2Feature: an easy-to-use interface to predict prokaryotic traits from 16S rRNA gene sequences. Bioinformatics Advances. 2026;6(1). doi:10.1093/bioadv/vbaf136

BibTeX
@article{Fujiyoshi2026,
  author = {Fujiyoshi, Masaki and Suzuki, Takao K and Furusawa, Chikara and Matsui, Motomu},
  title = {Bac2Feature: an easy-to-use interface to predict prokaryotic traits from 16S rRNA gene sequences},
  journal = {Bioinformatics Advances},
  year = {2026},
  volume = {6},
  number = {1},
  doi = {10.1093/bioadv/vbaf136}
}
Kostenlos starten →