*2.3. Data Analysis and Statistics*

Metagenomic shotgun sequencing read quality was assessed using FastQC and MultiQC [21,22], and host reads were filtered using kneaddata with low quality (Phred scores < 28) ends and reads trimmed for downstream analyses. Functional pathways of associated microbes were determined using omePath [23]. Functional associations between metabolites, clinical phenotypes, and microbes were assessed using Tweedievers [24]. Data are presented as the mean ± SEM, and statistical analyses were performed using GraphPad Prism 9.4.1. Data were analyzed by non-parametric Wilcoxon matched-pairs signed rank test. Multiple corrections of statistical tests were applied using the Benjamini and Hochberg false discovery rate (FDR), and differences were considered significant at *p* ≤ 0.05 unless otherwise noted.
