How much do your genes shape your microbiome?
Less than most people think. In a study of 1,046 people published in Nature in 2018, researchers found no significant link between genetic ancestry and the make-up of the gut microbiome, while diet, medicines and body measurements were associated with more than 20% of the differences between people. People who shared a home resembled each other even when they were not related.
That does not mean genes play no part. When very large groups are studied gene by gene, a few clear signals appear. The largest analysis so far, the MiBioGen consortium (18,340 people in 24 cohorts, Nature Genetics 2021), found 31 genetic regions linked to gut bacteria, and only one passed its strictest threshold: the lactase gene. A German study of 8,956 people published the same year confirmed the lactase signal and added two more: FUT2, which sets your secretor status, and the ABO blood group gene.
The three genes with the clearest effect
| Gene | What it does | Link with gut bacteria | How common |
|---|---|---|---|
| MCM6 / LCT (lactase) | Decides whether you keep digesting milk sugar after childhood | People who stop making lactase carry more Bifidobacterium, especially when they still eat dairy: the undigested lactose feeds them | Most adults worldwide stop making lactase; most northern Europeans keep it |
| FUT2 (secretor status) | Puts fucose-rich sugars on your gut lining and in body fluids | Non-secretors have a different mix of bacteria, often with fewer bifidobacteria | About 1 in 5 people of European ancestry are non-secretors |
| DAO (AOC1) | Breaks down the histamine in food, in the gut | Does not change which bacteria you carry, but changes how you react to histamine from fermented foods and some bacteria | Slower variants are common; symptoms decide whether they matter |
Lactase: the strongest link
Lactase persistence is read mainly from one variant, rs4988235 (also called −13910 C/T). One or two copies of the persistence version usually mean you digest milk sugar as an adult. With none, lactase production usually falls after childhood. The milk sugar you do not digest then reaches the colon, where bifidobacteria ferment it. That is why the link between the lactase gene and Bifidobacterium is strongest in people who still eat some dairy. This variant is the European one: in parts of Africa and the Middle East other variants do the same job, so a "non-persistent" result is less certain for people with that ancestry.
FUT2: what your gut lining offers
FUT2 adds fucose, a sugar, to the mucus that lines your gut. Many bacteria use it as food or as an anchor. About 1 in 5 people of European ancestry carry two non-working copies (rs601338 AA): they are "non-secretors". Several studies have found a different gut bacteria profile in non-secretors, with fewer bifidobacteria in some of them. Non-secretors also have higher vitamin B12 in their blood on average (Nature Genetics, 2008), and they are much less likely to catch the most common norovirus strains, which need those sugars to attach.
DAO: histamine, not bacteria
Fermented foods and some gut bacteria produce histamine, and DAO is the enzyme that breaks it down in the gut. Slower DAO variants can make aged cheese, wine, cured meats and long-fermented foods harder to tolerate for some people. This does not change your bacteria, but it does change which "gut-friendly" fermented foods suit you.
Blood group and secretor status
The German study of 8,956 people found that people with blood group A, B or AB who are also secretors were more likely to carry one group of Bacteroides than people with blood group O or non-secretors. Blood group was also linked with Faecalibacterium. The reason is the same as for FUT2: blood-group sugars are displayed on the gut lining only in secretors, and bacteria can feed on them.
This has nothing to do with the "blood type diet". A Canadian study of 1,455 people (2014) found that the benefits of those diets had nothing to do with the person's blood group. We do not read your blood group from your DNA file: consumer chips cannot reliably tell group A from group B, and most people already know theirs.
What about stool microbiome tests?
Stool tests try to list the bacteria in your gut, which DNA cannot do. The idea is sound, but in February 2026 a team from the US National Institute of Standards and Technology (NIST) sent identical portions of one stool sample to seven consumer testing services, three kits each, and published the results in Communications Biology:
- Across all reports, 1,208 different bacterial groups were named, but only 3 genera appeared in every result.
- Three services reported Clostridioides difficile; four did not.
- One service labeled the same sample "healthy" twice and "unhealthy" once, with different diet advice.
- For most common genera, the differences between services were as large as the natural differences between people.
The authors advise against basing diet or health decisions on these reports for now. We go through the study in our article on microbiome test reliability.
Eating for your gut profile
Whatever your genes, the advice with the best evidence is the same: eat many different plants. In the American Gut Project, people who ate more than 30 different plants a week had more diverse gut bacteria than those who ate fewer than 10. Your genes can then fine-tune a few choices:
| If your result is | Worth knowing | Food ideas |
|---|---|---|
| Non-secretor (FUT2) | Your gut lining offers bifidobacteria less fucose | Feed them with fibers they ferment: oats, legumes, onions, leeks, garlic, asparagus, chicory, slightly green bananas |
| Secretor (FUT2) | Your gut lining already feeds fucose-loving bacteria | Fiber variety does the rest |
| Lactase non-persistent | Undigested lactose feeds bifidobacteria | If you tolerate it, small portions of yogurt, kefir or hard cheese rather than no dairy at all |
| Slower DAO | Fermented foods also bring histamine | Fresh fermented foods in small portions; go easy on aged cheese, wine, cured meats and long-fermented sauerkraut |
None of this replaces medical advice. Persistent digestive symptoms call for a visit to a doctor, not a gene test.
How to check your own profile
If you have already tested with 23andMe, AncestryDNA, MyHeritage, FamilyTreeDNA or Living DNA, your raw data file contains these markers (how to download it). Our free test reads your secretor status, lactase and DAO in your browser, without sending the file anywhere. MyHeritage files lack the main FUT2 marker (rs601338), so we read rs492602, which is inherited with it: on the real files we checked, the two gave the same answer every time (49 of 49).
The Ultimate report has a Gut & Microbiome section with these results, what they mean for your gut bacteria and food ideas for your profile, alongside the full nutrition report and the medication genes.