Temperature fluctuations drove the interactions dynamic of keystone microbes for Chinese baijiu fermentation

Abstract

Temperature is an important abiotic factor to drive the spontaneous food fermentations, however, the effects of temperature on these processes still remain unclear. In this study, we revealed the dynamics of temperature, microbiota and metabolites in the typical Chinese baijiu fermentation. Moreover, we established a specific data-driven correlation analysis model based on machine learning. We revealed 13 high-abundance and 6 low-abundance species as the keystone microbes in this fermentation, by overall considering abundance, function and frequency of fermented microorganisms. In addition, the increasing of heating rate (from 2 to 5 ℃/day) strengthened the bacterial-bacterial interactions and weakened the bacterial-fungal interactions among these keystone microbes, subsequently affecting the metabolite shifts in baijiu fermentation. Therefore, we proposed a data-driven keystone microbe’s identification framework, and revealed the effects of temperature fluctuation on the interaction and metabolism of keystone microbes. Moreover, this study would pave the way for the identification of keystone microbes within complex microbial communities, and guide the optimization of fermentation processes for improving the quality of fermented products.