Dr. Alexandre Almeida had spent years building something almost no one had attempted before: a comprehensive reference catalogue of every genome in the human gut microbiome. The work required analyzing hundreds of thousands of microbial sequences, developing new computational tools to untangle them, and confronting a basic problem that has defined microbiome science since its beginning. More than 60 percent of the bacterial species living in the human gut have never been successfully grown in a laboratory. Scientists know they exist because their genetic fingerprints appear in sequencing data. What they actually do, and whether they matter, has largely been a mystery.
On February 9, 2026, Almeida and his team at the University of Cambridge’s Department of Veterinary Medicine published their answer in Cell Host & Microbe. Across 11,115 gut microbiome samples collected from people in 39 countries, one group of bacteria appeared again and again in the guts of healthy individuals and was consistently lower in people living with chronic disease. The group had a name, CAG-170, and almost no scientific literature behind it. It had been there all along.
The Bacteria Nobody Was Studying
The history of microbiome research is, in large part, a history of the bacteria scientists could grow. Culturing bacteria in the laboratory is the foundational method: take a sample, grow it on a plate, and study what grows. The bacteria that thrive in that environment are the ones that get studied. The bacteria that cannot survive outside the specific conditions of the human gut do not get studied, because nobody can grow them.
This is not a small gap. The University of Cambridge announcement of the study noted that over 60 percent of gut species remain uncultured and inaccessible to experimental manipulation. The microbiome field has been building its understanding of gut health on the fraction of bacterial life it can reach, which means it has been working with a minority of the available data since the field began.
Almeida’s previous work had already started to map the uncultured majority. Using metagenomics to analyze the genomes of all microbes in a gut sample simultaneously, then computationally separating them into individual species, he and colleagues constructed the Unified Human Gastrointestinal Genome catalogue. That catalogue revealed over 4,600 bacterial species, including over 3,000 that had not previously been documented in the human gut at all.
CAG-170 came from that catalogue. It was not one of the bacteria researchers typically discuss when they talk about gut health. It was not Akkermansia muciniphila, the well-studied species that lines the intestinal wall. It was not Faecalibacterium prausnitzii, the producer of butyrate, that appeared in hundreds of published studies. It was something far less visible, known only from its genetic fingerprint, impossible to study in isolation because nobody had figured out how to grow it yet.
11,115 People, 39 Countries, One Pattern
The scale of the analysis is what made the finding credible. A pattern that appears in one population might reflect diet, geography, or genetics rather than a universal feature of human health. A pattern that persists across 39 countries and 13 different diseases is something else.
Almeida’s team analyzed gut microbiome data from 11,115 people across 39 countries, including healthy individuals and those with 13 noncommunicable diseases. The diseases included inflammatory bowel disease, Crohn’s disease, obesity, chronic fatigue syndrome, and others. The researchers used multiple computational approaches to search for CAG-170’s genetic signature across the full dataset, cross-checking the results to rule out methodological artifacts.
The finding was consistent. People without chronic disease had higher levels of CAG-170 in their guts. People with chronic disease had lower levels. The pattern held across different countries, different diets, different demographics, and different disease categories. “We looked at the gut microbes of thousands of people across 39 countries and 13 different diseases, including Crohn’s and obesity,” Almeida said in the Cambridge press release. “We consistently found that people with these diseases had lower levels of CAG-170 bacteria in their gut.”
That consistency across disease types is the genuinely novel part. Most microbiome research identifies bacteria associated with specific conditions: a species linked to IBD, a different species linked to obesity, and another to depression. What Almeida’s study found was a group of bacteria whose presence correlates with health across disease categories, a potential common denominator beneath conditions that have very different clinical presentations.
What CAG-170 Might Actually Do
The mechanism is still being worked out, which the researchers are careful to acknowledge. CAG-170 cannot be grown in the lab, which means the experimental tools scientists normally use to establish causation, dosing cells with bacteria, removing bacteria from animal models, and running controlled trials are not yet available. The finding is a robust correlation from an unusually large dataset. Causation has not been established.
What the genetic analysis suggests is that CAG-170 bacteria are involved in digesting the main structural components of food and in maintaining the broader microbial community of the gut. Almeida described them as “likely helping us to digest the main components of our food and keeping the whole microbiome running smoothly.” The functional picture that emerges from the genetic fingerprints is of a group of bacteria that serve as generalists, supporting the ecosystem of the gut rather than performing a single specialized function.
This ecological framing matters. The microbiome is not a collection of independent species operating in isolation. It is a community in which some species provide resources that others depend on, and in which the removal of key members can cascade through the ecosystem in ways that individual-species analysis misses. If CAG-170 bacteria play a maintenance role for the broader microbial community, their consistent absence in chronic disease could be both a marker of disruption and a contributor to it.
The same research community that documented how ultra-processed foods reduce Akkermansia muciniphila and Faecalibacterium prausnitzii is now confronting the possibility that those well-studied species may be part of a larger community whose health depends on bacteria nobody had previously thought to investigate. CAG-170 was not missing from previous research because scientists looked and did not find it. It was missing because the methods used assumed bacteria could be cultured, and CAG-170 cannot.
The Hidden Microbiome and What It Contains
The 3,000-plus bacterial species that Almeida’s catalogue work revealed for the first time in the human gut represent a frontier that microbiome science is only beginning to explore. CAG-170 is one group within that frontier. Others remain entirely uncharacterized.
This has practical implications for every probiotic on the market, every dietary intervention aimed at improving gut health, and every clinical interpretation of microbiome testing. The bacteria that appear on commercial microbiome reports are, by definition, the ones that could be cultured or that appeared prominently enough in early sequencing studies to become reference points. The 60 percent of gut species that cannot be cultured are absent from most of the framework that guides clinical and consumer understanding of what a healthy gut looks like.
Almeida framed the implications directly: “We’re now discovering new groups of bacteria like CAG-170 with important links to our health, and probiotics aimed at supporting them could have a much greater health benefit.” That claim is conditional on developing the capacity to grow and study CAG-170 in the lab, which does not yet exist. But the direction of travel is clear: the next generation of microbiome interventions may target bacteria that current products do not contain and current tests do not measure, because the tools to work with them are still being developed.
The Steelman: What This Study Does Not Prove
Correlation between CAG-170 levels and health is not the same as proof that CAG-170 causes health or that its absence causes disease. This distinction matters, and the researchers acknowledge it clearly.
The study design is observational. It found patterns in existing data. It cannot answer whether people with chronic disease have lower CAG-170 because their disease disrupted the bacteria, or whether lower CAG-170 contributed to the development of disease, or whether both are downstream effects of some shared upstream factor, such as diet, medication use, or early life microbiome development. Establishing that causal direction requires experimental work that the current inability to culture CAG-170 makes impossible.
The 39-country sample, while large for a microbiome study, is also not evenly distributed across global populations. Microbiome composition varies significantly with diet, geography, and ancestry, and the extent to which CAG-170’s association with health generalizes across all human populations has not been fully established.
These are the appropriate limitations of a finding at this stage of the science. They do not reduce the significance of finding a consistent pattern across 11,115 samples. They define what needs to happen next.
What the Definition of a Healthy Gut Might Look Like
The field has been working toward a definition of what a healthy microbiome looks like for years. The challenge is that individual variation is enormous. Every person’s microbiome is unique, shaped by their diet, their early life environment, their medication history, and their geography. Identifying a universal signature of health within that variation has proven difficult.
CAG-170 represents a candidate for that signature, not a complete definition, but a component that persists across variation in diet, geography, and disease type in a way that few bacterial markers have previously demonstrated. The discovery connects to the same questions about gut health and systemic disease that Ozempic research has been probing through a completely different mechanism, suggesting that the gut’s relationship to chronic disease is more interconnected and more fundamental than any single research line has captured.
If CAG-170 can be cultured, studied, and eventually formulated into interventions that support its presence in the gut, the implications extend across every condition in which it was found to be depleted. That is a significant list: IBD, Crohn’s disease, obesity, and chronic fatigue syndrome. Not a niche population. A substantial fraction of people living with persistent illness in every country the study examined.
The Bacteria That Were Always There
Almeida’s catalogue revealed over 3,000 bacterial species in the human gut that had not previously been documented. CAG-170 is one of them, or rather, it was already in the catalogues, overlooked because most of its members cannot be grown in the lab, and its association with health had not been tested at the scale required to see it clearly.
The same pattern of missing something fundamental that was always present runs through the most significant discoveries in science: the mechanism that was there all along, waiting for a dataset large enough and a computational approach sophisticated enough to make it visible. The 11,115 samples from 39 countries were on the scale. The metagenomics tools were the approach. The finding was a group of bacteria that has been living in the guts of healthy people around the world, across every culture and every disease studied, quietly doing something that keeps the gut ecosystem functional.
Nobody had given it a proper look until now. It had been hiding in plain sight, which is the most unnerving place anything can hide.
