Integrating stool chemistry measurements with metagenomics in adults reveals distinct gut health profiles

Abstract

Stool chemistry metrics, including calprotectin, lactoferrin, and secretory IgA, are used to assess gut inflammation in clinical or research settings. However, these measures in isolation provide a limited view of gut physiology and lack microbial context. Shotgun metagenomics offers a comprehensive assessment of the gut microbiome's taxonomic composition and functional potential, yet its integration with stool chemistry metrics remains underexplored. We analyzed 877 adult stool samples with paired shotgun metagenomic sequencing and chemistry measurements to integrate the microbiome with 12 stool chemistry metrics. Calprotectin, lactoferrin, and secretory IgA were associated with shifts in microbiome composition (beta diversity), while deviations from typical levels of short-chain fatty acids (SCFAs), including butyrate, were associated with both changes in alpha diversity and microbiome composition. Network analysis revealed two distinct microbial neighborhoods: one associated with higher butyrate and lower calprotectin and lactoferrin and another associated with lower butyrate and higher calprotectin and lactoferrin. Machine learning models trained on taxonomic and functional data showed modest discrimination of stool chemistry states, performing best for SCFA deficiency and less well for host-derived inflammatory metrics. Together, these findings systematically characterize microbiome-stool chemistry relationships across a large paired dataset, providing a framework for interpreting stool chemistry metrics in the context of shotgun metagenomic profiles.