Researchers discovered that some gut bacterial populations spread between people and across continents within decades. The finding suggests your gut microbiome is partly shaped by whoever you spend time with.
The human gut microbiome has been studied intensively for two decades, and a broad consensus has emerged about what shapes it. Diet is the dominant factor: what you eat determines which bacteria thrive and which decline. Medications, particularly antibiotics, can dramatically reshape the microbial community. Age, genetics, early-life exposures, and geography all leave fingerprints. The picture that emerges from this research is of a personal ecosystem, something fundamentally individual that reflects your particular biology and life history.
A new study published in Nature has added a dimension to that picture that changes how the microbiome should be understood. Some gut bacterial populations, the research shows, are spreading between people, and they are doing so far more rapidly and extensively than the field had assumed. The transmission patterns the researchers found resemble what scientists had previously associated mainly with infectious pathogens, not with the ordinary commensal bacteria that make up the healthy human gut.
The research was led by Xiaoqian Annie Yu at the University of Vienna’s Centre for Microbiology and Environmental Systems Science, working with a team that included Adrian Tett, Martin Polz, and Nicola Segata. Their approach diverged from the standard framework of microbiome research in a specific and important way.
Most studies of the gut microbiome work at the species level, counting and comparing the relative abundances of different bacterial species across people. This approach has been productive but has a significant blind spot: it treats all members of a given species as interchangeable. A bacterium belonging to the species Bacteroides vulgatus in one person is counted as equivalent to a B. vulgatus in another person, even if their evolutionary histories, metabolic capabilities, and disease associations differ substantially.
The Vienna team used a method called reverse ecology to look inside species rather than simply counting them. By analyzing the genomic signatures of thousands of bacterial genomes, they identified evolutionarily distinct sub-populations within common gut species, groups of bacteria that have adapted to different niches within the gut environment and that have taken separate evolutionary paths over time. The method essentially reconstructs the family tree of bacteria within a species and identifies which branches of that tree behave differently from others.
What those branches revealed was unexpected.
How some gut bacteria travel between people and across continents
In ecology, a selective sweep refers to the rapid spread of a genetic variant through a population because it carries a fitness advantage. Selective sweeps are typically dramatic and fast relative to normal evolutionary change, leaving a distinctive signature in the genomes of organisms that have undergone them.
The researchers found evidence of selective sweeps within human gut bacterial species. Certain bacterial sub-populations, having acquired adaptations that made them more competitive within the human gut environment, had spread through human populations rapidly enough to leave these signatures. Some had expanded across continents within a period of decades.
This finding is notable because rapid continental spread of this kind had previously been documented primarily for pathogens: disease-causing bacteria that spread through transmission between hosts. The selective sweep pattern had not been expected in the ordinary bacteria of the healthy gut microbiome, which researchers had generally assumed to be shaped more by local diet and environment than by transmission between people.
“Even within the same bacterial species, some populations occur more frequently than others in certain diseases. When all are considered together, this often remains hidden,” said lead author Xiaoqian Annie Yu. “If you don’t just count species but take evolutionary adaptation into account, you can identify the biologically relevant units in the microbiome much more accurately.”
The specific bacteria showing the strongest transmission signals were not randomly distributed across the microbial community. They were enriched in sub-populations previously linked to disease outcomes, which connects the transmission finding directly to clinical significance.
Which bacteria are spreading and what diseases they are linked to
The reverse ecology analysis identified hidden sub-populations within gut bacterial species that differ not just in their evolutionary history but in their disease associations. Some populations within a species were enriched in healthy individuals. Others were consistently more prevalent in people with specific conditions.
Bacterial sub-populations associated with colorectal cancer showed some of the strongest evidence of active spread and competitive displacement of other populations within the same species. Colorectal cancer has one of the strongest documented relationships with the gut microbiome of any cancer, with several bacterial species, including Fusobacterium nucleatum and certain strains of Escherichia coli, appearing at elevated levels in tumor tissue. The new findings suggest that within-species heterogeneity may be critical: it is not simply the presence of the species but the specific sub-population that determines whether the bacterium is associated with disease or health.
Sub-populations linked to inflammatory bowel disease, including Crohn’s disease and ulcerative colitis, also showed distinct transmission patterns and ecological signatures. IBD-associated bacterial populations appeared to occupy different niches within the gut environment compared to their disease-neutral counterparts within the same species, suggesting that the disease association reflects genuine differences in how these bacterial populations interact with the host.
For type 2 diabetes, the analysis identified bacterial sub-populations whose distribution across human populations followed patterns consistent with active spread rather than simple local environmental variation. Sub-populations associated with accelerated biological aging showed similar signatures.
The convergence of transmission evidence and disease association in the same bacterial sub-populations is the most clinically significant aspect of the study. If bacteria linked to cancer and metabolic disease are spreading between people, the implications for understanding disease risk extend beyond individual diet and genetics to include social contact patterns.
What this means for how you think about your microbiome
The conventional advice about gut health focuses entirely on things you do alone: what you eat, whether you take probiotics, whether you avoid unnecessary antibiotics, how much fiber you consume. The implicit assumption behind all of this advice is that your microbiome is a product of your individual choices and your individual biology.
The new research adds a social dimension that this advice ignores. If bacterial sub-populations spread between people at rates comparable to those seen in pathogens, then the microbial composition of the people you spend time with is relevant to your own gut health. Household members, romantic partners, close friends, and family members may all be contributing to each other’s microbiomes through normal social contact.
Prior research has documented microbiome similarity between people who live together. Long-term partners who have cohabited for years tend to share more gut bacterial strains than unrelated individuals. Household members show higher microbiome similarity than people who live separately. The conventional interpretation of these findings has been that shared diet and shared environment explain the similarity.
The Vienna study suggests that active transmission between people may be a more important driver of this household-level microbiome convergence than previously thought. The bacteria are not just responding to the same dietary environment. Some sub-populations are spreading directly between hosts.
The research team is careful to note that “contagious” carries implications that the study cannot fully support. The evidence for continental spread is found in genomic signatures rather than in direct observation of specific transmission events. The study shows that bacterial populations have spread, not precisely how they spread in everyday life. Whether the main routes involve shared food, physical contact, airborne particles, or other vectors requires further investigation.
What the hidden structure inside species changes about disease research
The methodological contribution of the study may ultimately matter as much as the specific biological findings. The reverse ecology approach reveals a level of organization within gut bacterial species that conventional species-level analysis cannot detect.
If disease associations are concentrated in specific sub-populations within a species rather than distributed uniformly across the species, then studies that count species without resolving sub-population structure are likely to miss or blur those associations. A species might appear unrelated to a disease at the population level because disease-associated and health-associated sub-populations are both present in large numbers and cancel each other out statistically.
“Pinpointing these specific strains could make microbiome-based diagnosis and treatment far more precise,” the researchers noted. Microbiome-based diagnostics, which attempt to identify which bacteria are present as markers for disease risk or disease state, have so far shown only modest clinical utility. The resolution problem the Vienna team identified may be one reason: species-level counting is too coarse to distinguish the sub-populations that actually carry the disease signal.
The same problem affects microbiome-based therapeutics, particularly the fecal microbiome transplants used to treat recurrent Clostridioides difficile infection and being investigated for other conditions. If the therapeutic effect of a transplant depends on the specific sub-population of bacteria transferred rather than simply the species present, then donor selection and transplant protocols may need to account for sub-population composition rather than only species composition.
What the study does not establish
The study’s reverse ecology method identifies sub-populations based on genomic signatures and infers their ecological properties from those signatures rather than directly observing their behavior. The disease associations identified are derived from comparing sub-population distributions across published datasets and require validation in prospective studies that directly test whether specific sub-population composition predicts disease outcomes.
The continental spread patterns are inferred from the geographic distribution of genomic signatures, not from tracking specific bacterial populations across populations over time. The decades-long timescales suggested by the sweep patterns are estimates based on genomic clock models, which carry their own uncertainties.
The study does not establish specific transmission routes or quantify how much of an individual’s microbiome composition is shaped by person-to-person transmission versus diet, environment, and other factors. The relative contribution of transmission versus lifestyle factors to microbiome composition remains an open question.
What it establishes, at a level of resolution not previously applied to the healthy gut microbiome at scale, is that the species-level picture of the gut microbiome is a significant simplification. Inside each species is hidden structure, and that hidden structure contains disease information, transmission history, and evolutionary dynamics that the conventional approach to microbiome research has largely missed.
The study, “Genome-wide sweeps create ecological units in the human gut microbiome”, was authored by Xiaoqian Annie Yu, Cameron R. Strachan, Craig W. Herbold, Michaela Lang, Christoph Gasche, Athanasios Makristathis, Nicola Segata, Shaul Pollak, Adrian Tett, and Martin F. Polz at the University of Vienna and collaborating institutions, and published September 7, 2026 in Nature.
Source: University of Vienna / Centre for Microbiology and Environmental Systems Science. DOI: 10.1038/s41586-026-10476-w
