*4.5. Microbiome Analyses*

The Microbiome Analyst platform [63,64] was used to calculate alpha and beta diversities and to compare the relative abundance of bacterial taxa. For richness measurements, we used Observed (amount of unique OTUs found in each sample) and Chao1 (also accounting for unobserved species based on low-abundance OTUs). For evenness measurement, Shannon diversity was used, which accounts for both richness and abundance. A *p* value of <0.05 was considered significant. Beta diversity represents the diversity between microbial communities. Bray–Curtis dissimilarity or the Jensen–Shannon distance was calculated to measure beta diversity, and then, principal coordinates analysis (PCoA) was applied for visualization. For differential analysis, DESeq2 was used for samples less than 20 samples per sample; it is computationally intensive but more robust with low false positive rates, and a *p* value of <0.05 was considered significant.

Linear discriminant analysis effect size (LEfSe) was used to identify the key microbial taxa associated with the different locations. This analysis integrates statistical significance with biological consistency (effect size) estimation. It uses a nonparametric factorial Kruskal– Wallis (KW) rank sum test to detect features with significant differential abundance with respect to the class of interest, followed by linear discriminant analysis to estimate the effect size of each differentially abundant feature. The original LEfSe implementation uses original *p* values when determining significant taxa, and an LDA score > 3 (effect size) and a *p* value of <0.05 were considered statistically significant. Meanwhile, the Microbiome Analyst implementation provides the option to use either original or FDR-adjusted *p* value cutoffs to identify significant features.
