Sequential active learning incorporating multi-objective optimization for selective bacterial growth
New Biotechnology · 2026-11
DOI: 10.1016/j.nbt.2026.11.005 ↗Kurz gefasst
Fachartikel in der Zeitschrift New Biotechnology, 2026-11. Verfasst von Shuyang Zhang und 1 weiteren Autor:innen. Beteiligte Einrichtung: Graduate School of Life and Environmental Sciences, University of Tsukuba, 1-1-1 Tennodai, Ibaraki, Tsukuba 305-8572, Japan. Die Arbeit nennt 61 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: Japan Science and Technology Agency, Japan Society for the Promotion of Science.
Abstract (Originalsprache)
Developing selective media for bacterial cultivation is challenging because of complex nutritional requirements and the need to suppress competing species. Here, we present a machine learning-based sequential active learning framework that incorporates multi-objective optimization to design selective media from 30 common components for six phylogenetically diverse bacteria. High-throughput monoculture growth assays generated data used to train gradient-boosting decision tree models with multi-objective functions, which predicted media formulations that maximize target growth while minimizing non-target growth. Iterative cycles of prediction and experimental validation refined media compositions, yielding selective media that enhanced target bacterial growth and selectivity. Coculture experiments confirmed the effectiveness of the selective media in promoting specific bacteria within communities, demonstrating that monoculture data can predict community growth dynamics. This data-driven approach enables selective medium development without prior knowledge of bacterial physiology or inhibitors, providing a scalable and systematic framework for microbiology and systems biology applications.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Graduate School of Life and Environmental Sciences, University of Tsukuba, 1-1-1 Tennodai, Ibaraki, Tsukuba 305-8572, Japan.
- Graduate School of Life and Environmental Sciences, University of Tsukuba, 1-1-1 Tennodai, Ibaraki, Tsukuba 305-8572, Japan; MiCS, University of Tsukuba, 1-1-1 Tennodai, Ibaraki, Tsukuba 305-8572, Japan. Electronic addr…
Förderung
- Japan Science and Technology Agency
- Japan Society for the Promotion of Science
Zitieren
APA
Zhang, S., & Ying, B. W. (2026). Sequential active learning incorporating multi-objective optimization for selective bacterial growth. New Biotechnology. https://doi.org/10.1016/j.nbt.2026.11.005
Vancouver
Zhang S, Ying BW. Sequential active learning incorporating multi-objective optimization for selective bacterial growth. New Biotechnology. 2026. doi:10.1016/j.nbt.2026.11.005
BibTeX
@article{Zhang2026,
author = {Zhang, Shuyang and Ying, Bei-Wen},
title = {Sequential active learning incorporating multi-objective optimization for selective bacterial growth},
journal = {New Biotechnology},
year = {2026},
doi = {10.1016/j.nbt.2026.11.005}
}
EUVIMED