A decade of evidence linking gut microbiome dysbiosis to depression: a computational meta-analysis of mechanistic pathways: 2014–2024

OBJECTIVES: The gut microbiome-gut-brain axis (MGBA) has been associated in the pathophysiology of depression; however, the expanding literature remains fragmented across metabolic signalling, immune-inflammatory pathways, stress physiology and dysbiosis outcomes. METHODS: Abstracts were retrieved from bibliographic databases (Lens.org, PubMed, DOAJ, Europe PMC) for studies published between 2014 and 2024 investigating associations between the gut microbiome and depression using 16S rRNA sequencing. Following text preprocessing, Latent Dirichlet Allocation (LDA) was applied to identify latent thematic topics. Topic proportions were subsequently embedded using principal component analysis (PCA), t-distributed stochastic neighbour embedding (t-SNE), and uniform manifold approximation and projection (UMAP). RESULTS: Topic modelling revealed distinct interconnected thematic domains within the gut microbiome depression literature, encompassing metabolic and short chain fatty acid pathways, immune inflammatory mechanisms, stress and hypothalamic pituitary adrenal (HPA) axis regulation, probiotic and interventional work, microbial diversity and compositional metrics, neurochemical and neuroplasticity, developmental cohorts, and sequencing- or methodology-focused research. CONCLUSIONS: Computational synthesis indicates that research on the gut microbiome depression axis is structured around multiple convergent mechanistic themes. This thematic landscape highlights dominant areas of mechanistic focus, providing a conceptual framework to guide future experimental design, mechanistic validation and translational research.