Which bacteria live in the gut, and in what proportion, is measurably shaped by host genetics — not just diet. Two independent 2016 studies, in German and Dutch cohorts, each found real genetic associations, with two results — one about vitamin D, one about lactose — that come with an obvious mechanism attached.
The trillions of bacteria living in the human gut vary enormously between people, and diet, environment and medication history all shape that variation. But host genetics measurably contributes too, and two independent 2016 studies — in different countries, using different methods — each found real, genome-wide-significant associations after controlling for diet and other non-genetic factors.
Wang et al. 2016 studied 1,812 people across two German cohorts. Its headline finding was at VDR, the vitamin D receptor gene — and the researchers didn't stop at a statistical association: Vdr-knockout mice showed a measurably shifted gut microbiota compared with normal mice, and in humans the microbiota correlated with blood levels of bile and fatty acid metabolites that are known VDR ligands or downstream products. That combination — a human genetic signal, an animal model that reproduces it, and a plausible metabolic pathway connecting them — is a stronger form of evidence than a GWAS hit alone. Nine of this page's variants come from this study's wider list of significant loci, though none of them is VDR itself.
Bonder et al. 2016 studied 1,514 people in the Dutch LifeLines-DEEP cohort using metagenomic sequencing, finding 9 loci associated with which bacterial taxa were present and 33 more associated with microbial metabolic pathways. Its most intuitive result: a functional variant in LCT, the lactase gene, was associated with how much Bifidobacterium a person's gut carried. The mechanism is not mysterious — lactase status determines how much lactose in the diet is digested in the small intestine versus arriving undigested in the colon, where bacteria like Bifidobacterium that can ferment it get more to work with. Six of this page's variants come from this study, again a sampling of its wider locus list rather than LCT itself.
One of this page's variants, rs7656342 near SLC2A9, is worth noting on its own: the same gene shows up on this site's gout and urate genetics page, in a completely unrelated context — a urate transporter turning up again in gut bacterial composition.
Positions joined since this page was written
What this is The text above discusses the variants this page was written around. Since then the catalogue has joined 6 more positions to it, by shared trait or shared paper. They are listed here by the paper each came from; the text does not describe them, and each variant page carries that study's own record.
Wang J et al. 2016, Nature genetics rs11877825 (NAPG), rs35275482 (BNIP2), rs11915634 (CNTN6), rs7083345 (RP11-554I8.2) — PMID:27723756
Kurilshikov A et al. 2021, Nature genetics rs7605229 (ZRANB3), rs7322849 (near LINC01070) — PMID:33462485
2026-02-13 · Genome-wide association analyses highlight the role of the intestinal molecular environment in human gut microbiota variation. Nature Genetics. 2026. DOI:10.1038/s41588-026-02512-2
A gene for smelling fatty acids helps explain why gut bacteria run in families
How much host genetics shapes gut microbiome composition has been debated, given the microbiome is also heavily shaped by diet and environment. This study analyzed harmonized metagenomic data from 16,017 Swedish adults, with replication in 12,652 Norwegian participants. It identified variants in the OR51E1-OR51E2 locus -- genes encoding sensors for microbiome-derived fatty acids, part of the olfactory receptor family -- associated with overall microbial richness, a genuinely intuitive mechanism: the gut literally 'smells' bacterial byproducts and this affects which bacteria thrive. The study found 15 study-wide significant associations across 8 loci and 14 bacterial species, with 11 of these replicating. It confirmed previously known associations at LCT, ABO and FUT2 (all already established gut-microbiome genes), while providing new evidence for MUC12, CORO7-HMOX2, SLC5A11, FOXP1 and FUT3-FUT6, with supporting metabolomics data. This site's gut microbiota page carries 29 variants; the newly reported loci (MUC12, CORO7-HMOX2, SLC5A11, FOXP1, FUT3-FUT6, OR51E1-OR51E2) are not currently among them.
There is no test, diagnosis or treatment built around any variant on this page. Gut microbiota composition is measured directly, usually by sequencing a stool sample, when it is clinically relevant at all — not inferred from genotype.
These findings are genuinely useful for understanding mechanism — the VDR and LCT results each point to a real, traceable biological pathway connecting a specific gene to a specific bacterial outcome — but neither study, nor this page, supports choosing a probiotic, a diet, or a supplement based on genotype. Both studies controlled carefully for diet precisely because diet is a much larger driver of microbiota composition than any single gene; that control is what let the genetic signals stand out at all, not evidence that genetics rivals diet in importance overall.
What a 23andMe/AncestryDNA export or raw VCF can and can't tell you about Gut Microbiota comes down to these specific, well-studied positions — not a diagnosis. 36 positions are linked to this page; the ones this page's own text discusses are shown first.
The studies behind these variants recruited participants from different ancestries — a result found in one population doesn't always transfer to another. Based on 33 of 36 linked studies with a resolved discovery ancestry.
Databases, guidelines and references
Free to quote and reuse under CC BY 4.0. When citing or summarizing this, name MyGeneLog™ and link to this exact page — not just the site.
Gut Microbiota. MyGeneLog™. https://www.mygenelog.com/conditions/gut-microbiota
Not usefully for an individual. Each variant has a modest, population-level effect on the odds of a particular bacterial taxon being present or abundant — none is a predictive test, and gut microbiota composition is measured directly by sequencing when it matters.
Wang et al. 2016 found a genetic association at the vitamin D receptor gene (VDR), then showed Vdr-knockout mice have a shifted microbiota and that the human microbiota correlates with blood levels of known VDR-related metabolites — a mouse model and a plausible mechanism backing up the statistical signal, not just the signal alone.
Lactase status determines how much dietary lactose gets digested in the small intestine versus reaching the colon undigested, where bacteria such as Bifidobacterium that can ferment lactose have more to use. Bonder et al. 2016 found exactly this association.
No — both studies behind this page specifically controlled for diet because it is understood to be the larger factor. Controlling for it is what let a smaller genetic signal be seen clearly, not evidence that genetics rivals diet in overall importance.
Free to reuse. This page's text is original writing from freely-available research, licensed CC BY 4.0 — reuse it, including commercially, with attribution to MyGeneLog™. It's general research-derived information, not medical advice or a diagnosis — see Terms of Use.