What the microbiome actually is
The human gut contains a dense community of bacteria, archaea, fungi and viruses, the overwhelming majority living in the colon. The common estimate is around a hundred trillion microorganisms. The figure is approximate, and worth holding loosely — the well-known claim that microbial cells outnumber human cells ten to one was revised in 2016 to something closer to one to one.
That correction is a useful entry point, because the gap between what is established and what is repeated is unusually wide in this field. The microbiome is genuinely important. It is also the subject of more confident marketing than the underlying science currently supports.
A few things are not in dispute. The community is highly individual — two healthy people can share relatively little at species level while functioning identically. It is established early in life, influenced by birth mode, early feeding and antibiotic exposure. And it is responsive: it changes with what you eat, and it changes fast.
Where it lives
Microbial density increases sharply along the digestive tract. The stomach is relatively sparse, held in check by acidity. The small intestine carries more but is a fast-moving environment focused on nutrient absorption. The colon is where the population concentrates, and where fermentation happens. When people say gut microbiome, they are almost always talking about the colon.
What it does
Three functions have solid support.
Fermenting fibre
Humans cannot digest most dietary fibre. Gut bacteria can. When they ferment it they produce short-chain fatty acids — principally butyrate, propionate and acetate. Butyrate is the primary fuel for colonocytes, the cells lining the colon wall.
This is the mechanism worth understanding, because it explains the rest. Eating fibre is not simply about bulk moving through the system. It is feeding a population that produces a compound your intestinal lining runs on.
Training immunity
A large share of immune tissue sits in and around the gut. The microbiome interacts with it continuously, contributing to the calibration between responding to genuine threats and tolerating harmless material. This is one reason early-life microbial exposure attracts so much research attention.
Competitive exclusion
An established community occupies space and consumes resources, making it harder for pathogens to gain a foothold. This is part of why antibiotic courses sometimes precede opportunistic infections: clear the incumbents and you free the niche.
These three functions are well supported. Much of what circulates beyond them — specific bacteria for specific moods, personality-linked microbial signatures, precision probiotics — sits somewhere between early research and speculation.
The composition question
Most gut bacteria fall into a small number of phyla. The proportions vary widely between healthy individuals, which is precisely why interpreting a single readout is difficult.
| Phylum | Typical share | Associated role | What to know |
|---|---|---|---|
| Firmicutes | ~40–60% | Fibre fermentation, butyrate production | Includes many butyrate producers. The ratio to Bacteroidetes was once linked to obesity; that finding has largely not replicated. |
| Bacteroidetes | ~20–40% | Carbohydrate and protein breakdown | Highly versatile. Tends to rise on plant-rich, fibre-heavy diets. |
| Actinobacteria | ~5–10% | Includes Bifidobacterium | Prominent in infancy, declines with age. Common target of probiotic products. |
| Proteobacteria | ~5–10% | Diverse; usually a small share | Sustained expansion is often treated as a marker of disturbance rather than a cause. |
| Verrucomicrobia | ~1–3% | Mucus layer maintenance | Includes Akkermansia muciniphila, an active research subject. Promising, not settled. |
Note how many entries carry a caveat. That is the honest state of the field. The Firmicutes-to-Bacteroidetes ratio in particular was widely reported as an obesity marker and has since largely failed to hold up — a good example of a finding entering popular health writing and staying there long after the evidence moved.
Dysbiosis and its limits
Dysbiosis describes a microbial community that has shifted away from a typical healthy pattern. It is associated with irritable bowel syndrome, inflammatory bowel disease, metabolic conditions and more.
But it is a descriptive term, not a diagnosis. There is no agreed definition, no threshold, and no single test that confirms it. Much of the research is cross-sectional, meaning it observes a difference without establishing whether the microbial state caused the condition or resulted from it. Illness changes what people eat, and diet changes the microbiome.
If a product promises to correct your dysbiosis, ask what it is measuring against. There is no established reference range for a healthy microbiome, because healthy microbiomes differ enormously.
What genuinely shapes it
| Factor | Strength of effect | Speed | Under your control |
|---|---|---|---|
| Diet | Large | Days | Yes — the main lever |
| Antibiotics | Large | Immediate | Sometimes; never refuse a necessary course |
| Fibre and plant diversity | Large | Days to weeks | Yes |
| Sleep and circadian rhythm | Moderate | Weeks | Largely |
| Chronic stress | Moderate | Weeks | Partly |
| Physical activity | Modest but consistent | Weeks | Yes |
| Age | Moderate | Decades | No |
| Birth mode and infancy | Moderate, early | Fixed | No |
| Geography and household | Moderate | Years | Rarely |
Diet dominates the list of things you can actually influence. That is the practical conclusion, and it is unglamorous: the intervention with the best evidence is eating more plants, more often, of more kinds.
Fibre and the diversity question
Different fibres feed different bacteria. Inulin from onions and chicory, beta-glucan from oats and barley, resistant starch from cooled cooked rice and potatoes, pectin from apples, and the varied fibres in legumes each support different populations.
This is why plant diversity appears to matter more than total fibre grams. The American Gut Project reported that people eating more than thirty different plant types weekly had measurably more diverse microbiomes than those eating fewer than ten. It was observational, so causation is not established, but the finding is consistent with the mechanism.
Thirty sounds demanding until you count properly. Herbs, spices, nuts, seeds, wholegrains, legumes, tea and coffee all count. A bowl of mixed grains with three vegetables, seeds and two herbs is eight or nine in a single meal.
Increase slowly
A sudden large increase in fibre typically produces bloating and gas. This is fermentation working, not damage, but it is uncomfortable and it is the usual reason people abandon the change. Increase over several weeks and drink enough water. If you have IBS, a rapid increase in fermentable fibre may worsen symptoms — that is a case for professional guidance rather than a general protocol.
Fermented foods and the Stanford trial
In 2021 a Stanford team ran a ten-week randomised trial comparing a high-fibre diet against a high-fermented-food diet in healthy adults. It is among the more informative pieces of human evidence available, because it was interventional rather than observational.
The fermented-food group showed increased microbial diversity and decreases across a panel of inflammatory markers. The high-fibre group did not show the same diversity increase within that window, and inflammatory response varied by participant — appearing to depend partly on the microbiome each person started with.
That result was not what many expected, and it deserves care rather than enthusiasm. It was one trial, over ten weeks, in a modest number of healthy adults. It does not demonstrate that fibre is unimportant; decades of evidence link fibre intake to gut and cardiovascular outcomes. What it suggests is that fermented foods do something distinct, and possibly faster.
| Food | Origin | Live cultures | Notes |
|---|---|---|---|
| Miso | Japan | Yes, if unpasteurised | Add after cooking — boiling kills the cultures. High in salt. |
| Kimchi | Korea | Yes | Fibre plus live cultures. Look for refrigerated, not shelf-stable. |
| Natto | Japan | Yes | Bacillus subtilis. Notably high in vitamin K2. |
| Sauerkraut | Central Europe | Only if unpasteurised | Shelf-stable jars are usually heat-treated and contain no live cultures. |
| Kefir | Caucasus | Yes | Broader strain range than most yoghurt. |
| Yoghurt | Widespread | Yes | Check for live active cultures. Many flavoured products are high in sugar. |
| Nukazuke | Japan | Yes | Vegetables fermented in a rice-bran bed, tended daily. |
| Kombucha | East Asia | Yes | Modest culture content. Often high in sugar. |
Most supermarket sauerkraut, pickles and shelf-stable kimchi are pasteurised for shelf life, which kills the organisms you are buying them for. If it sits unrefrigerated and has a long date, assume no live cultures.
The gut-brain axis, carefully
The gut and brain communicate continuously, through the vagus nerve, immune signalling, and metabolites produced by gut bacteria. That much is established, and it is genuinely interesting.
Two claims need correcting, because both circulate widely.
The first: ninety percent of your serotonin is made in the gut. True. But that serotonin does not cross the blood-brain barrier. It regulates intestinal motility. It is not the pool that antidepressants act on, and the implied conclusion — that eating for your gut directly raises brain serotonin — does not follow.
The second: specific bacteria control mood. The striking results here come largely from rodent studies, where transferring microbiota between animals of differing temperament has produced behavioural change. Human evidence is thinner, mostly observational, and complicated by the obvious problem that depression changes appetite and diet, which changes the microbiome.
None of this means the connection is unreal. It means the direction of causation is unresolved in humans, and confident claims about curing anxiety through gut bacteria are ahead of the data.
An Eastern reading of digestion
Chinese medicine has no concept of the microbiome. It does have a detailed framework for digestion, built on observation over roughly two millennia, and the overlap is interesting in places.
In that tradition the Spleen and Stomach form the body’s transformation centre, converting food into usable substance. The recommendations that follow are consistent: eat warm cooked food rather than cold raw food, keep regular meal times, stop before full, avoid eating late, and do not eat while agitated.
| Traditional guidance | Modern reading | Evidence |
|---|---|---|
| Eat warm cooked food | Cooking softens plant cell walls and can increase availability of some nutrients | Partial — varies by food; some nutrients decrease with heat |
| Regular meal timing | Digestive function follows circadian rhythms; the microbiome shows daily oscillation | Reasonable support |
| Stop at seventy percent full | Satiety signalling lags the stomach by 15–20 minutes | Well supported |
| Do not eat late | Late eating is associated with poorer metabolic markers | Growing support |
| Do not eat when agitated | Stress alters motility and permeability via the gut-brain axis | Well supported |
| Fermented and pickled staples | Delivers live cultures and fibre together | Supported by the 2021 trial |
The convergence should not be overstated. A tradition that recommends regular warm meals and fermented vegetables was not describing butyrate production. But arriving at similar practical advice from close observation of how people felt is not nothing, and it is a reasonable reason to take the practices seriously without adopting the metaphysics.
A practical protocol
Four weeks, one change at a time. Nothing here requires a purchase beyond food.
| Week | Change | Why |
|---|---|---|
| 1 | Count your plants for seven days. Do not change anything. | You need a baseline. Most people guess high and are surprised. |
| 2 | Add one fermented food daily — a spoon of kimchi, a bowl of miso soup, live yoghurt. | Direct live cultures. This is the change with the fastest measured effect. |
| 3 | Raise plant variety toward thirty a week. Herbs, spices, nuts and seeds count. | Diversity of fibre feeds diversity of bacteria. |
| 4 | Fix an eating window — regular times, nothing substantial within three hours of sleep. | Digestive function and microbial activity both follow daily rhythms. |
Practical tips
- Keep a jar of mixed seeds on the table — four plant types added to anything without thought.
- Cook rice or potatoes and cool them before eating; cooling forms resistant starch, a useful fibre.
- Buy fermented foods from the refrigerated section, not the shelf.
- Add miso after the heat is off, never to boiling liquid.
- Frozen vegetables count fully. They are picked ripe and often more nutritious than tired fresh produce.
- Tinned legumes are the cheapest fibre diversity available. Rinse and add to anything.
- Herbs and spices count toward plant diversity. A spice-heavy dish can add five in one go.
What does not work
- Colon cleanses. No demonstrated microbiome benefit. Can strip beneficial bacteria and disturb electrolytes.
- Most probiotic supplements, for most healthy people. Evidence is strain-specific. A study showing benefit for one strain in one condition says nothing about a different product.
- Consumer microbiome tests. They report what is present. The science cannot yet convert that into reliable personal advice, and labs disagree with each other.
- Detox teas. Frequently contain laxatives. Bowel movement is not microbial improvement.
- Eliminating whole food groups without cause. Restriction tends to reduce plant diversity, which is the opposite of the goal.
Frequently asked questions
How much does the gut microbiome weigh?
Estimates commonly cited put it at roughly 0.2 to 2 kg, though the figure is uncertain and often overstated. The frequently repeated claim that bacteria outnumber human cells ten to one was revised downward in 2016 to roughly one to one. The microbiome matters regardless of which number is correct.
Can I change my microbiome through diet?
Yes, and quickly. Composition shifts measurably within days of a substantial dietary change. Whether those shifts persist depends entirely on whether the diet does. Return to your old eating pattern and the microbiome largely follows.
Is fibre or fermented food more effective?
A 2021 Stanford randomised trial compared both over ten weeks. The fermented-food group showed increased microbial diversity and decreased inflammatory markers. The high-fibre group did not show the same diversity increase over that period, though fibre remains strongly associated with gut health in other research. The honest reading is that both matter and the fermented result was somewhat unexpected.
Do I need a probiotic supplement?
For most healthy people, no. Evidence is strain-specific rather than general, which means a product showing benefit for one condition tells you nothing about a different product for a different condition. Fermented foods deliver live cultures alongside fibre and other compounds, usually at lower cost.
What is dysbiosis?
A general term for a microbial community that has shifted away from a typical healthy pattern. It is descriptive rather than diagnostic. There is no agreed clinical definition and no single test that confirms it, which is why commercial microbiome tests should be treated with caution.
Are microbiome testing kits worth it?
Currently they tell you what is present but not reliably what it means. The science cannot yet translate most compositional readouts into specific actionable advice. Two labs can return different results from the same sample. Save the money and spend it on vegetables.
How do antibiotics affect the gut?
They reduce both harmful and beneficial bacteria. Most people recover substantially within weeks to months, though some taxa may not return to baseline. This is a reason to avoid unnecessary antibiotics, not a reason to refuse necessary ones — untreated bacterial infection is considerably more dangerous.
What are short-chain fatty acids?
Compounds including butyrate, propionate and acetate produced when gut bacteria ferment dietary fibre. Butyrate is the primary energy source for the cells lining the colon. This is the clearest mechanism by which eating fibre translates into a physical effect on the gut wall.
Is the gut really a second brain?
The gut has its own extensive nervous system and communicates with the brain via the vagus nerve, immune signalling and microbial metabolites. That two-way link is well established. The stronger popular claims about mood being controlled by gut bacteria run considerably ahead of the human evidence, most of which is observational or from animal studies.
Does most serotonin come from the gut?
Roughly ninety percent of the body's serotonin is produced in the gut, which is true and frequently quoted. What is usually omitted is that this serotonin does not cross the blood-brain barrier. It regulates gut motility. It is not the same pool that antidepressants act on.
How long does it take to improve gut health?
Composition changes within days. Symptomatic change, where it happens, is usually reported over weeks. Anyone promising a transformation in three days is selling something.
Does stress affect the gut?
Yes, through well-documented pathways involving the vagus nerve, gut motility and intestinal permeability. This is one point where traditional Eastern frameworks and modern physiology broadly agree, having arrived from very different directions.
Should I do a gut cleanse or detox?
No. Colon cleanses have no demonstrated benefit for microbiome health and can disrupt electrolyte balance and strip beneficial bacteria. The organs that handle clearance are the liver and kidneys, and they do not require assistance from a product.
Are artificial sweeteners bad for the gut?
Some studies suggest certain sweeteners alter microbial composition and glucose response, but findings are inconsistent and often involve doses beyond typical consumption. The honest answer is that this remains unresolved.
What single change matters most?
Increase plant diversity. The number of different plants eaten weekly is more strongly associated with microbial diversity than total fibre quantity. Thirty different plants a week is a commonly cited target, counting herbs, spices, nuts, seeds and grains.
Key takeaways
- Diet is the lever you control, and it works within days.
- Plant variety beats fibre quantity. Aim for thirty types a week.
- Fermented foods showed measurable benefit in a controlled trial — buy refrigerated, not shelf-stable.
- Increase fibre gradually or you will abandon it.
- The gut-brain link is real; claims that gut bacteria govern mood are ahead of the human evidence.
- Skip the cleanses, the detox teas and, for most people, the supplements.
- Persistent symptoms are a reason to see a clinician, not to buy a protocol.
Related books
Read next
References and further reading
- Wastyk HC, Fragiadakis GK, Perelman D, et al. Gut-microbiota-targeted diets modulate human immune status. Cell 2021;184(16):4137–4153. The Stanford fermented-food versus fibre trial.
- Sender R, Fuchs S, Milo R. Revised estimates for the number of human and bacteria cells in the body. PLoS Biol 2016;14(8):e1002533. Corrects the ten-to-one claim.
- McDonald D, Hyde E, Debelius JW, et al. American Gut: an open platform for citizen science microbiome research. mSystems 2018;3(3):e00031-18. Origin of the thirty-plants observation.
- Koh A, De Vadder F, Kovatcheva-Datchary P, Bäckhed F. From dietary fiber to host physiology: short-chain fatty acids as key bacterial metabolites. Cell 2016;165(6):1332–1345.
- Cryan JF, O'Riordan KJ, Cowan CSM, et al. The microbiota-gut-brain axis. Physiol Rev 2019;99(4):1877–2013. Comprehensive and appropriately cautious.
- Unschuld PU (2003). Huang Di Nei Jing Su Wen. University of California Press.
A note on health. This article is educational and is not medical advice, diagnosis or treatment. Traditional Chinese medical concepts are presented as a historical framework, not as biomedical fact. Persistent digestive symptoms — bleeding, unexplained weight loss, ongoing pain or change in bowel habit — warrant assessment by a qualified clinician. Do not delay medical care on the basis of anything written here.