Maotai-flavor Baijiu, a traditional Chinese liquor produced via solid-state fermentation, exhibits diverse base Baijiu types due to variations in fermentation rounds, styles, and grades. While crucial for flavor complexity, current manual identification methods hinder blending efficiency and quality control. This study employed GC-FID and machine learning to analyze 410 base Baijiu samples. Decision Tree (74.36 %), XGBoost (92.9 %), and Random Forest (62.3 %) emerged as optimal classifiers for fermentation rounds, typical styles, and Chuntian grades, respectively. SHAP analysis revealed: (1) esters as primary markers for fermentation rounds, (2) ester-trimethylbutanol combinations for grade differentiation, and (3) multi-compound signatures (butyric acid, tetramethylpyrazine, 2-butanol et al.,) for style discrimination. Notably, marker compounds' flavor properties - beyond mere concentration - critically influenced their discriminative power, as evidenced by correlations with nine sensory dimensions.

Pickle like odor (PLO) is one of the main defective flavors of Maotai flavor baijiu (MFB). Understanding and controlling the PLO compounds producing strains not only solves the problem of PLO from the source, but also ensures the high-quality production of MFB. However, the relevant research on PLO compounds producing strains has not been reported in MFB. In this study, we identified a Bacillus amyloliquefaciens ZZ7 with high yield of PLO compounds in the fermented grains of MFB, and measured its physiological characteristics. It produces 627 volatile compounds and 1,507 non-volatile compounds. There are 7 volatile sulfur compounds that cause the PLO, the content of dimethyl disulfide, dimethyl trisulfide, and dimethyl sulfur is relatively high, accounting for 89.43% of the total volatile sulfur compounds. The genome size of B. amyloliquefaciens ZZ7 is 3,902,720 bp with a GC content of 46.09%, and a total of 3,948 protein coding genes were predicted. Moreover, the functional annotation of coding genes and an assessment of the metabolic pathways were performed by genome annotation, showing it has strong ability to transport and metabolize amino acids and carbohydrates. Comprehensive genomic and metabolomic analysis, the metabolic pathway of PLO compounds of B. amyloliquefaciens ZZ7 was revealed, which mainly involves 12 enzymes including sulfate adenylyltransferase, cysteine synthase, cystathionine γ-synthase, etc. This work provides biological information support at both genetic and metabolic levels for the mechanism of B. amyloliquefaciens ZZ7 to synthesize PLO compounds, and provides a direction for the subsequent genetic modification of ZZ7 to solve PLO from the source in the MFB.

The production of Chinese Baijiu relies on the synergistic metabolism of multi-species microbial communities in an open environment. Its intricate microbial succession and flavor formation mechanisms have long been considered complex systems that are difficult to fully deconstruct. Traditional culture-dependent techniques inherently fail to comprehensively capture the actual functional roles and dynamic regulation of “viable but non-culturable” (VBNC) microorganisms within this complex system. In recent years, the rapid advancement of multi-omics technologies has offered a novel perspective for elucidating the underlying fermentation mechanisms of Baijiu. This paper systematically reviews the recent progress in the application of metagenomics, metatranscriptomics, metaproteomics, and metabolomics in Baijiu research. Specific focus is placed on the unique contributions of these tools to resolving microbial community structural diversity, mining key functional genes and enzymes, uncovering microbial stress response mechanisms under environmental fluctuations, identifying phages and spoilage microorganisms, and tracing the metabolic pathways of flavor substances. Furthermore, the pivotal role of multi-omics integration strategies in constructing “microbe–metabolite” regulatory networks is highlighted. Finally, current challenges regarding standardization and data integration are discussed, with an outlook on leveraging omics big data to promote digital monitoring and intelligent brewing in the Baijiu industry.