Functional multi-axis decomposition of the human gut microbiome: an operational definition of eubiosis and dysbiosis, and a clean-reference framework for disease stratification

Abstract

Taxonomy gives only a partial description of a gut microbiome, since phylogenetically distinct communities encode much the same metabolism and composition is confounded by lifestyle and geography. Here we introduce TAGMOS, an annotation-verified enzymatic architecture aimed at reading gut-microbiome function, built on a fermentative engine and a set of host-facing channels. The engine reports whether the community disposes of fermentative hydrogen through anaerobic sinks or through respiration, and it is the causal driver of the state: antibiotics drive it to an oxidised, respiratory state while faecal transplantation restores the fermentative milieu, and across 73 cohorts the oxidised state is where facultatively anaerobic Enterobacteriaceae bloom. We observe that the engines disease signal lies in the tail of a population, and a threshold criterion detects disease that a comparison of means overlooks, whereas the channels describe condition-specific mechanistic signatures. Engine and channels transfer across pipelines and populations where taxon-based indices fail. Combining engine and specific channels defines a portable, quantitative measure of eubiosis and dysbiosis, and shows that this state is causally downstream of the communitys metabolic structure.