Network-Based Quantitative Analysis Of Microbiome–Metabolome Interactions Regulating Human Intestinal Homeostasis

Abstract

The human gut microbiome and metabolome are important factors in maintaining intestinal homeostasis by complex interactions amongst the microbes and by metabolic regulation. But the conventional analytical methods tend to focus on microbial communities or metabolites separately which do not enable the understanding of the entire functional networks among microbes. In this study, a network-based, quantitative framework, which combines the microbiome and metabolome datasets, is introduced for the characterization of interaction patterns that govern intestinal homeostasis in humane. The framework relies on the identification of key microbial taxa and metabolites, as well as microbial regulatory modules, using the integration of microbial co-occurrence analysis, microbiome–metabolome association modeling, graph-theoretical network analysis and functional pathway enrichment. Network analysis quantitatively identified 67 key metabolites and 42 hub microbial species involved in strongly connected network metabolic pathways. The proposed model successfully predicted functional interaction modules with an accuracy of 95.2%, precision of 94.6%, recall of 93.8%, F1-score of 94.2% and ROC–AUC value of 0.972. Functional enrichment showed certain associations with short-chain fatty acid metabolism, bile acid biosynthesis and amino acid metabolism. The results confirm the validity of the proposed platform for understanding microbiome–metabolome interactions and offer key insights for the discovery of new markers and for precision medicine approaches for intestinal wellbeing.