Deterministic selection and compositional turnover in Parkinson's disease-associated gut dysbiosis

The scientific understanding of links between Parkinson's disease (PD) and gut microbiome dysbiosis has advanced significantly, yet the ecological mechanisms driving these microbial changes remain poorly understood. To address this gap, we postulated that PD-associated gut dysbiosis arises as harmful microbes outcompete beneficial bacteria, and consequently any therapeutic strategies must both suppress opportunistic pathogens and restore protective, fiber-degrading microbes to effectively rebalance the gut microbiome in PD. To evaluate ecological patterns consistent with this hypothesis, we apply Sloan's near-neutral model (SNM), Ning et al.'s stochasticity framework, and ecological network analysis to reanalyze six published gut microbiome datasets (1957 samples total: 804 healthy controls, 1153 PD cases). We first applied SNM to categorize bacterial species as neutral, positively selected, or negatively selected. While the overall proportions of these categories were similar between groups (neutral: ~ 40%, positively selected: ~ 47%, negatively selected: ~ 13%), stochasticity framework analysis revealed significantly stronger deterministic selection in PD microbiomes. Shared species analysis (SSA) resolved this apparent paradox by demonstrating substantial compositional shifts within each species category, indicating that while classification frequencies remained stable, the specific microbes occupying these ecological niches changed significantly in PD. This divergence between SNM category proportions and NSR values highlights a subtle yet critical aspect of community assembly dynamics. Ecological network analysis further revealed that neutral species had fewer antagonistic co-occurrence links, consistent with their ecological equivalence, while negatively selected species maintained higher relative abundances in both groups. Together, these findings indicate that PD-associated gut microbiomes are characterized by stronger deterministic assembly signatures and substantial compositional turnover within near-neutral ecological categories. These patterns are consistent with altered ecological assembly signatures in PD-associated dysbiosis.