Kept apart from the drug pages on purpose. Every drug here stands on a prescribing guideline that says what to do differently at a given genotype. Almost nothing sold as a supplement has that, and listing the two together would lend the second the authority of the first.
So the evidence is stated on every entry, and on most of them the useful answer is that no guideline exists. Nothing here recommends taking anything.
A commercial genetic test told people with one eye genotype to skip zinc and another to skip antioxidants. Three independent teams re-ran the numbers behind it and found the claim did not hold up — but a separate, real trial found something narrower that might.
The most important thing on this page is not a genotype. It is that two large trials gave these to smokers and counted more cancers, not fewer.
A professional body says genetic variation may explain why the same coffee helps one person and keeps another awake. It stops short of telling anyone to test.
Genetics has run the experiment nobody could run deliberately — lifelong high milk intake, assigned at conception. It did not reduce hip fractures.
Not on most people's radar as a nutrient, and one variant explains a real, measured dose-response in how much of it a body actually needs — 13%, 43% or 80% risk of symptoms on a deficient diet, by genotype.
Sold almost entirely to one group of people — statin takers with aching muscles — on a mechanism that the trial which measured it directly could not find.
MTHFR is the most overstated gene on the internet. The variant is real, its effect on an enzyme is real, and almost everything sold on the back of it is not.
The one supplement on this page where a genotype changes what a doctor would say — and it changes it towards caution, not towards a purchase.
A real, replicated genetic effect on blood magnesium, found in a channel the kidney uses to reclaim it — and still not a reason to skip the blood test.
The gene cluster that decides whether your body actually turns flaxseed oil into the omega-3s fish already contain — and no guideline tests it before recommending a source.
The same MTHFR variant behind the folate page, in a randomised trial testing a completely different question: whether riboflavin lowers blood pressure — and in one genotype, it measurably did.
A confirmed genetic locus for blood selenium — and, in the one trial built to test whether raising it prevents cancer, a result no better than placebo, with a non-significant hint of harm.
Two common variants in the enzyme that turns plant beta-carotene into usable vitamin A — carriers of both convert 69% less efficiently, which is a real reason "eat more carrots" does not work the same for everyone.
One of the clearest gene–nutrient links there is, and it works through the gut lining rather than through anything on the label. It still does not tell anyone to take a supplement.
The same 2009 study that found the FUT2 link to B12 on this site found a second one, in a gene most people have never heard of, for B6 — and it clears the vitamin rather than absorbing it.
Ordinary vitamin C is not the concern. High-dose vitamin C — the kind pushed as an infusion — is a documented cause of severe hemolysis in one specific genotype, and every reported case with that genotype has had it happen.
Genotype explains part of why two people in the same sunlight end up with different blood levels. It does not tell either of them what to take.
A 2024 Cochrane review found zinc barely prevents a cold at all, but may shorten one already underway by about two days — with more side effects than placebo either way.
One of three things, stated in full rather than as a badge a reader has to interpret. "A guideline says something specific" means a professional body has specific prescribing-adjacent advice. "The association is established; acting on it is not advised" means the genetics is real and replicated, but no guideline sets a dose or a choice from it. "Studies disagree — unsettled" means the research itself has not converged on an answer.
No. Nothing on this site recommends starting, stopping or choosing a supplement. Each page states what the genetics and the clinical trials actually found, including where a supplement was tested directly and did not help — or, in two cases here, was tested and caused measurably more harm than placebo.
Every drug on this site stands on a prescribing guideline that says what to do differently at a given genotype. Almost nothing sold as a supplement has that, and listing the two together would lend the second the authority of the first — so supplements are kept as their own axis, with the evidence level stated on every entry.
By cross-referencing genes with a documented role in a supplement’s absorption, metabolism or blood level against variants already published on this site, then verifying the primary study directly rather than trusting a gene name or a marketing claim. That is a narrower list than what is sold in a pharmacy, on purpose.