Deciphering the Microbial Complexity of Chinese Traditional Sourdough through Integrated Learning

Sourdough fermentation relies on a dynamic microbial consortium of fungi and bacteria, shaped by interspecies interactions and environmental cues. We characterised the diversity and functional traits of microbial communities in 115 sourdough samples across 10 Chinese provinces, integrating high-throughput sequencing with physicochemical analyses (moisture, pH, and total acidity). In dry samples, Pediococcus(P.) pentosaceus and Lactobacillus brevis dominated, whereas wetter samples were enriched with Fructilactobacillus(F.) sanfranciscensis and Kazachstania(K.) humilis. Ensemble learning and abundance ranking identified Saccharomyces cerevisiae as the core fungal species across both groups, alongside variable taxa such as Cladosporium delicatulum(C.) and Aspergillus(A.) penicillioides. Microbial community structure correlated strongly with physicochemical conditions, such as moisture, pH, and acidity, highlighting the importance of designing stable synthetic consortia to enhance the consistency and functional performance of sourdough fermentations. However, current limitations in sampling scale and the depth of functional annotation restrict broader generalisation. This study establishes a foundation for rational manipulation of sourdough microbiomes, offering both theoretical and technical insights to guide future research.