The strong-flavor Baijiu (SFB), a Chinese liquor made by anaerobic fermentation, accounts for about 51 % of total liquor consumption in China (Liu et al., 2023). SFB is fermented for 60–90 days in mud cellars and distilled using fermented grain known as Zaopei. In Zaopei, microbes, including bacteria, fungi, and archaea, together determine the flavor and quality of SFB (Hu et al., 2021; Li et al., 2022; Liu et al., 2023; Qian et al., 2021; Tan et al., 2019; Xu et al., 2022). Several molds, such as Aspergillus and Rhizopus, secrete enzymes that contribute to saccharification, including amylase, glucoseamylase, and cellulase (Tu et al., 2022; Wang et al., 2020). Yeast, such as Saccharomyces cerevisiae, contributes to ethanol fermentation (Wu et al., 2016). A major function of bacteria (e.g., Clostridium, Bacillus, and Lactobacillus) is to produce flavor precursors or components, such as hexanoic acid, ethyl hexanoate, and ethyl lactate, which are crucial for the aromatic properties of SFB (Liu et al., 2020; Tu et al., 2022). SFB fermentation can be divided into two phases, with 14–21 days as the cut-off date, depending on the dynamics of the bacterial communities in Zaopei. The bacterial community undergoes drastic changes in the early stages, but then becomes more stable (Hu et al., 2021; Liu et al., 2023; Tan et al., 2019; Xu et al., 2022). Previous studies described microbial metabolic activity during peak bacterial diversity on the seventh day of fermentation (Hu et al., 2020; Liu et al., 2023). However, the changes in structure and function of bacterial community during SFB fermentation have not been extensively explored.

Many efforts to study the Baijiu brewing bacterial communities have relied on sequencing of 16S rRNA gene amplicons, particularly those in variable regions V3-V4 (Hu et al., 2021; Liu et al., 2023; Qian et al., 2021; Tan et al., 2019; Xu et al., 2022). Although this second-generation sequencing technology provides a comprehensive picture of bacterial diversity, its phylogenetic resolution is limited by short read lengths (Jeong et al., 2021; Myer et al., 2016; Yu et al., 2022). In recent years, the PacBio platform for sequencing full-length 16S rRNA genes (V1-V9) has been used to characterize the bacterial communities of Baijiu-producing environments (Dong et al., 2022; Li et al., 2022). In various settings, PacBio platforms have been shown to produce higher-resolution profiles of bacterial communities compared to sequencing only a few variable regions (Jeong et al., 2021; Myer et al., 2016; Yu et al., 2022). However, it is unclear whether PacBio platforms are better suited than V3-V4 platforms for study of the Baijiu brewing bacterial communities. In addition, full-length 16S rRNA gene sequencing is not without limitations, including its high cost per read, making large-scale sequencing a challenge (Myer et al., 2016; Yu et al., 2022).

Another powerful tool used to explore the Baijiu microbiome is whole metagenomic shotgun (WMS) sequencing, an approach that reduces amplification bias and allows strain-level analysis (Mthethwa et al., 2021). Despite recent cost reductions in sequencing, WMS datasets remain more expensive to produce, require more computing power to analyze, store, and process, and are difficult to analyze for low-biomass samples (Mthethwa et al., 2021). A major advantage of WMS is its ability to extract gene content related to microbial function (Mthethwa et al., 2021). However, functional genetic profiles can be predicted from 16S rRNA sequences using PICRUSt (Phylogenetic investigation of communities by reconstruction of unobserved states), at tool that predicts gene families based on reference genomes (Douglas et al., 2020). There has been increased recent use of PICRUSt to predict microbial activities in the Baijiu-making environment (Chai et al., 2019; Liu et al., 2020; Liu et al., 2023; Qian et al., 2021). Nevertheless, this consistency between theoretical and actual microbiological functionality has not yet been evaluated in the brewing environment.

In this study, we examined the bacterial dynamics in Zaopei during SFB brewing using 16S rRNA gene sequencing for variable regions V3-V4 and V1-V9. In addition, WMS analysis was performed on the same samples to compare the ability of 16S rRNA gene-based sequencing methods to effectively characterize the microecology of Zaopei and determine the functions of the bacterial population. The annotation of 16S rRNA gene sequences was done using two widely used databases, SILVA and NCBI, since different reference databases vary in the reported structures of microbial communities (Balvočiūtė and Huson, 2017). The purpose of this study was to quantify the effects of sequencing methods on Baijiu fermentation microbiology surveys. The findings may help guide the optimal selection of methodologies for future investigations of Baijiu fermentation as well as for other anaerobic fermentation systems.