2025-09-16 カーディフ大学
Web要約 の発言:
<関連情報>
- https://www.cardiff.ac.uk/news/view/2946284-ai-could-im-prove-sourdough-starters
- https://www.sciencedirect.com/science/article/abs/pii/S0924224425003693
AIとマルチオミクスによるサワードウ発酵の強化:自然の多様性から合成微生物群集へ Enhancing sourdough fermentation with AI and multi-omics: From natural diversity to synthetic microbial community
Yujuan Yu, Jiale Wang, Faizan Ahmed Sadiq, Honghong Cheng, Aowen Liu, Yan Liu, Senmiao Tian, Jingjing Liang, Ling Zhu, Guohua Zhang
Trends in Food Science & Technology Available online: 14 August 2025
DOI:https://doi.org/10.1016/j.tifs.2025.105233
Highlights
- Lactic acid bacteria and yeast interactions stabilize sourdough microbial ecosystems and quality.
- Artificial intelligence and multi-omics optimize microbial interactions in sourdough fermentation.
- Synthetic microbial communities (SynCom) standardize sourdough, enhancing efficiency, flavor, and health benefits.
- Genome-scale metabolic models reveal interactions and guide rational design of SynCom for sourdough.
Abstract
Background
Sourdough fermentation is driven by complex microbial consortia shaped by diverse factors, including flour type, environment, and process conditions. Traditional natural sourdough relies on spontaneous microbial colonization from raw materials and the environment, often resulting in high variability in microbial composition, metabolic activity, and fermentation outcomes. These unpredictable dynamics lead to challenges in maintaining product consistency, controlling fermentation time, ensuring food safety, and scaling up production.
Scope and approach
This review outlines how multi-omics approaches could be applied to characterize sourdough microbiota, understand microbial interactions, and monitor metabolic functions. We further explore the transition from natural fermentations to synthetic microbial communities (SynComs), highlighting how the integration of artificial intelligence (AI) and omics can guide SynCom design and predict community behavior under varying fermentation conditions.
Key findings and conclusion
The integration of AI and multi-omics enables in-depth modeling of microbial interactions and functional outputs in sourdough ecosystems. These approaches provide a rational framework to construct functionally stable SynComs for consistent fermentation performance. Although challenges remain in scaling SynComs for industrial application, AI-driven multi-omics strategies represent a promising avenue to optimize sourdough fermentation and foster innovation in fermented food production. Importantly, these findings have broader implications for the development of community models across clinical and industrial microbiology.
