Integrative Multiomics Analysis of Gut Microbial Dysbiosis Associated with Immune Dysfunction in Human Populations

Abstract

Gut microbial dysbiosis is an essential factor dependent on immune regulation; however, complex interactions among microbial communities, metabolites and immune responses remain inefficiently characterized. This research develops an integrative multi-omics analysis of gut microbial dysbiosis associated with immune dysfunction in human populations. Prospective cohort research involving 132 participants, aged 18–65, comprising 66 immune dysfunction patients and 66 healthy controls, was conducted. Clinical characteristics, gut microbial profiles, transcriptomic and immune-related biomarkers included in this research were systematically analyzed. Microbial diversity and community composition were characterized using alpha-and beta-diversity analyses, while differential abundance analysis identified dysbiosis-associated microbial taxa and functional alterations. Differential metabolites, immune-related genes, proteins, cytokines, and chemokines were identified across the population groups using statistical comparisons. Data Integration Analysis for Biomarker discovery using Latent variable approaches for Omics studies (DIABLO), Multi-omics factor analysis (MOFA) and correlation-based integration were employed to characterize relationships among microbial, metabolic, transcriptomic, and immune features. Statistical association analyses identified significant microbe–metabolite– gene–immune relationships, and network analysis highlights hub features. Functional and pathway enrichment analyses characterized biological mechanisms involving toll-like receptor, oxidative-stress, bile-acid, cytokine, T-cell receptor, tryptophan-related, short-chain-fatty-acid and Janus Kinase–Signal Transducer and Activator of Transcription (JAK–STAT) pathways. Multivariate statistical analyses, Support Vector Machine (SVM) modeling, Extreme Gradient Boosting (XGBoost) based imputation, and Partial Least Squares Discriminant Analysis (PLS-DA), were further used to evaluate group distinctions. MOFA identified microbial-metabolic-immune dysregulation provides 28.4% variance, with Metabolite A exhibited Variable Importance in Projection (VIP) of 0.42. The integrated analysis provides a comprehensive characterization of dysbiosis-associated microbial, metabolic, genetic, proteomic, and immune signatures, as well as potential biomarkers underlying immune dysfunction across human populations