Dietary Protein Source Shapes Gut Microbial Structure and Predicted Functional Potential: a Systematic Integrative Re-Analysis Using Machine Learning

Abstract

Dietary proteins shape gut microbial ecology, yet the taxonomic and functional consequences of plant- compared with animal-based proteins remain poorly defined. Although digestibility and fermentation profiles differ by protein type, a systematic evaluation of how these differences influence microbial diversity, community structure, and metabolic capacity is lacking. This study represents a systematic integrative re-analysis of raw 16S rRNA sequencing datasets derived from independent controlled animal feeding studies. Following PRISMA guidelines, we analyzed 16S rRNA sequencing data from 10 murine studies (n = 187) comparing plant- and animal-protein diets. Alpha diversity was assessed using Shannon, Inverse Simpson, and Chao1 indices, and beta diversity with Aitchison distances. Differentially abundant taxa were identified using linear discriminant analysis, effect size, and class-weighted Random Forest (RF) models. Functional potential was inferred with phylogenetic investigation of communities by reconstruction of unobserved states, and taxon-pathway relationships were explored using correlation and network analyses. Plant-protein diets increased gut microbial diversity across all alpha diversity metrics and were associated with higher representation of saccharolytic and nitrogen-recycling genera such as Bacteroides, Muribaculaceae, and Allobaculum. Animal-protein diets favored proteolytic taxa, including Clostridium sensu stricto 1 and Colidextribacter. Microbial community structure differed significantly between diets (analysis of similarities R = 0.663, P < 0.001). RF models achieved >88% accuracy (area under the curve = 0.995) in predicting dietary groups, and linear discriminant analysis effect size identified consistent discriminating taxa. Functional profiling showed that plant-based diets enriched pathways linked to short-chain fatty acid and aromatic amino acid metabolism, whereas animal-based diets favored sulfur- and branched-chain amino acid-associated pathways. Network analysis identified Muribaculaceae as a plant-associated hub and Lactobacillus as an animal-associated hub. Dietary protein source significantly influences gut microbiota composition and functional potential in mice. Plant- and animal-based proteins generate distinct metabolic signatures with implications for nitrogen cycling, sulfur metabolism, and microbial ecology. Future controlled dietary studies that harmonize protein source with other macronutrient variables are needed to isolate protein-specific effects.