Analysis of gut bacterial communities leveraging AI methods unveils novel relations between species in Alzheimers disease

Abstract

The gut microbiome has been increasingly implicated in Alzheimers disease (AD), with studies reporting numerous species- and genus-level differences. These findings established a growing catalog of AD-associated taxa, yet they typically evaluate taxa individually or in small sets rather than across microbial communities. Gut microorganisms act collectively through cross-feeding, competition, and metabolic exchange. Hence, jointly analyzing co-occurring species can reveal community structure and biological insights that a taxon-by-taxon analysis may miss. In 274 stool metagenomes from 119 older adults (18 with AD), we used our Alzheimers disease Analysis Model (ADAM) framework to run Latent Dirichlet Allocation (LDA) 1,000 times with other bioinformatics tools, decomposing species abundances into communities and aligning them into 22 reproducible ones. Of the 50 species that define these communities, 15 showed an AD-associated shift in relative abundance (10 depleted, 5 enriched; Cohens d from -0.91 to +0.67, each 95% CI excluding zero), spanning the depletion of Phocaeicola vulgatus (d - 0.91, 95% CI [-1.23, -0.59]) and the enrichment of Bacteroides fragilis (d +0.67, 95% CI [0.35, 0.99]), the two ends of an AD-associated balance. Separately, P. vulgatus competitively excludes its congener Phocaeicola dorei. In AlzBiom, an independent amyloid-defined cohort, the exclusion reproduced (within the Bacteroidaceae, r = -0.43 vs -0.57 in GAINS) and held in both control and AD participants, a conserved, disease-independent property. The P. vulgatus/B. fragilis balance also reproduced but more modestly (d = -0.24, permutation p = 0.038), whereas the substitution toward P. dorei did not. IMPORTANCEThe gut microbiome, the collection of bacteria living in the human gut, is organized into interacting communities whose members rise and fall together. Reading them jointly is more faithful but yields complex, high-dimensional patterns that standard analysis cannot resolve. Making sense of them means weighing how species move together, what they do, and what the literature reports, a task suited to artificial intelligence. We used our previously developed Alzheimers disease Analysis Model, an artificial intelligence framework customized here for species-community analysis, to integrate this evidence and turn patterns into biological findings. These findings are relationships, not single species, and they separate conserved ecology, shared with and without the disease, such as the competition between Phocaeicola vulgatus and its relatives, from shifts specific to Alzheimers disease. This keeps a general relationship from being mistaken for a disease marker, because the signal lies in the community, not in a single microbe.