A different perspective on nutrition: DNA-based diets & Nutrigenomics

Diet according to DNA: personalized nutrition or fortune-telling?

Why does one of two people following the same diet lose weight quickly while the other struggles? Why does the same meal send one person's blood sugar soaring while having a milder effect on another? "DNA-based diets", which seek answers to these questions in our genes, have become a large market in recent years. Tests claiming to identify "the best diet for you" from a saliva sample are also widely sold in Türkiye. But what does science say? The short answer: Personalized nutrition is a real and important goal; however, there is no evidence that this goal can be achieved with a DNA test alone. In this article, we discuss what nutrigenomics is, what randomized studies actually show, which genetic tests are truly useful in nutrition, and why different types of data need to be evaluated together.

Terms: nutrigenetics, nutrigenomics and epigenetics

• Nutrigenetics: Studies how our genetic differences affect our responses to foods. For example, "Does a person with this genotype break down caffeine more slowly?"
• Nutrigenomics: Studies how foods affect gene activity. In everyday language, all "DNA-based diet" tests are marketed under this heading, but most are actually nutrigenetic tests.
• Epigenetics: Studies the mechanisms determining when and how much genes function without changing the DNA sequence. One striking example is the Dutch Hunger Winter: people who were in the womb during the famine of 1944-45 had chemical marks on the growth-related IGF2 gene that differed from their siblings even six decades later. We explained the measurement of epigenetic age in this article .

Different responses to the same meal: real, but genetics is a small part

It is now well established that people respond differently to the same foods. In 2020, Nature Medicinepublished the PREDICT study, in which 1.002 people (including twins) received the same standardized meals. Differences between individuals in post-meal responses were very large: variability of %103 for blood lipids (triglycerides), %68 for blood sugar and %59 for insulin.

However, the picture changes when we ask how much of these differences is genetic:

Thus, "different responses to the same meal" are real; however, most of this variation comes from our gut bacteria, when and under what conditions we eat, our sleep and our activity, rather than our genes. We discussed meal timing in detail in our chrono-nutrition article .

Do diets based on DNA work? Randomized studies

The most reliable way to establish whether an approach actually works is to assign people randomly to groups and compare them. The key studies of DNA-based diets are as follows:

The Food4Me result is particularly instructive: personalized nutrition works, but this effect largely comes from reviewing a person's current diet and providing specific feedback accordingly; genetic information adds nothing to this effect. We also discussed the place of genetic testing in nutrition in our article on DNA-based genetic tests for nutrition .

Where do the examples in the old article stand today?

The FTO gene and weight

FTO is the gene whose association with obesity is best known. In a study involving 38 thousand people in 2007, those carrying two copies of the risk genotype (AA), approximately %16 of adults, weighed an average of 3 kg more and had 1,67 times higher odds of obesity than people without this variant. The finding mentioned in the old article, "AA carriers gain more weight when eating a high-fat, low-carbohydrate diet", comes from an observational study conducted at a single time point in Sweden; it measured body mass index at that moment rather than weight gain. A meta-analysis of randomized studies, however, shows that FTO genotype does not affect weight-loss success. Thus, healthy eating and physical activity work regardless of your FTO gene. For more information on obesity and the healthiest diets, see this article .

Caffeine and CYP1A2

The CYP1A2 enzyme breaks down caffeine in the liver. For the variant called rs762551, people with the AA genotype break down caffeine quickly, while carriers of the C allele (AC and CC) do so more slowly. In a Costa Rican study of 2.014 heart attack patients and 2.014 healthy people, among slow caffeine metabolizers, drinking 2-3 cups of coffee daily was associated with a %36 increase in heart attack risk, and 4 or more cups daily with a %64 increase; no such increase was observed among fast metabolizers. This association was more pronounced below age 59. However, this is a case-control study, and genotype-based coffee advice has not yet entered guidelines. The European Food Safety Authority (EFSA) recommendations, regardless of genotype, are as follows: up to 200 mg at once and 400 mg daily for healthy adults, and up to 200 mg of caffeine daily for pregnant and breastfeeding women are safe.

MTHFR, folate, alcohol and colorectal cancer

MTHFR is an enzyme involved in folate metabolism. The old article stated that the C677T variant increased colorectal cancer risk; however, the evidence shows a different picture. In a meta-analysis of 77 studies, the TT genotype, compared with CC, was associated with a slightly lower risk (odds ratio 0,88). This protective effect is particularly seen in people with adequate folate intake; the advantage disappears with folate deficiency and excessive alcohol consumption. Regardless of genotype, a folate-rich diet was associated with lower risk in both groups.

NAT2, well-done red meat and cancer

Some carcinogens formed in meat cooked at high temperatures are processed by the NAT2 enzyme. Small studies had reported increased risk among smokers with the "rapid acetylator" genotype who preferred well-done red meat. However, in a pooled analysis of 11 studies and more than 17 thousand people, high red meat intake was associated with a similar increase in colorectal cancer risk (%38-43) in both slow and rapid acetylators; genotype did not modify the effect. Thus, the message is the same for everyone: limit red and processed meat. We discussed foods that reduce colorectal cancer risk in this article, and the impact of lifestyle on this cancer in this article .

Genetic tests that are genuinely useful for nutrition

Genetic tests cannot be said to have no place in nutrition. Tests that address a specific medical question and whose result changes treatment are useful:

• Lactose intolerance: The LCT -13910 variant near the lactase gene determines whether the lactase enzyme persists in adulthood. It can be used in diagnosis alongside methods such as breath testing.
• Celiac disease: If both HLA-DQ2 and DQ8 genotypes are absent, celiac disease is practically excluded. This test is particularly useful in people who have already started a gluten-free diet or whose blood tests and biopsy results disagree. However, carrying these genotypes does not mean having celiac disease; a substantial proportion of the population carries these genes.
• Inherited diseases: In conditions such as metabolic diseases identified by newborn screening, including phenylketonuria, hereditary iron accumulation (hemochromatosis) and familial high cholesterol, genetic diagnosis directly determines the dietary and treatment plan.

Personalized nutrition: multiple data rather than a single test

The idea of personalized nutrition is sound; the mistaken assumption is that it can be achieved with a single DNA test. Today's most promising approaches combine multiple types of data:

• Personalized blood sugar prediction: A study of 800 people in Israel, published in 2015 in Cell, showed that an algorithm combining blood values, diet, body measurements, physical activity and gut microbiota could predict post-meal blood sugar responses, and that a diet prepared accordingly reduced blood sugar fluctuations.
• The ZOE METHOD study: In a randomized study published in 2024 that followed 347 people for 18 weeks, a personalized nutrition program based on blood lipid, blood sugar and microbiota data reduced triglycerides more than standard dietary advice (0,13 mmol/L) and produced approximately 2,5 kg more weight loss; there was no difference in LDL cholesterol. It should also be noted that the study was conducted by a commercial company, %86 of participants were women, and the findings have been criticized.
• Continuous glucose monitors: Their use in people without diabetes is becoming more widespread, but evidence that they improve health outcomes remains limited.

The market, regulation and data privacy

• The market: Commercial reports suggest that the nutrigenomics market is growing rapidly; estimates range for 2023 from approximately 450 million dollars to estimates for 2025 of approximately 610 million dollars, and for 2035 of approximately 3 billion dollars. Estimates vary between reports and are not independent scientific sources.
• Regulation: The United States Food and Drug Administration (FDA) does not review non-medical, "general wellness" direct-to-consumer tests before they enter the market, and advises consumers to consult a healthcare professional before making dietary or health decisions based on these tests. An international consensus report published in 2017 also emphasizes that nutrigenetic tests are unregulated and lack defined standards.
• Advertising oversight: In 2021, the United Kingdom Advertising Standards Authority ruled that an advertisement for a DNA diet test claiming to "unlock the secret of your ideal diet" was unsupported by evidence.
• Data privacy: Genetic data are permanent and concern your relatives as well as you. The company 23andMe, which holds the genetic data of millions of people, filed for bankruptcy in March 2025 and was sold to a nonprofit research organization in July 2025. This process raised the question of what happens to genetic data when companies change hands.

What should you do in practice?

• Start with the basics: A balanced diet rich in vegetables, fruit, legumes and whole grains, with little processed food, applies to everyone. We compiled the basic tips for healthy eating in a separate article.
• Start personalization with your current diet: Reviewing your current diet with a dietitian and receiving specific feedback is more useful than a DNA test.
• If you have a specific medical question (lactose intolerance, suspected celiac disease, a family history of high cholesterol), consult your physician; medical genetic testing may be ordered if necessary.
• If you are considering a DNA diet test: Ask which genes are examined, what scientific evidence supports each recommendation and how your data will be protected.
• Observe your own response: Your own experience, such as how coffee affects your sleep or palpitations, is often more informative than your genotype.

Frequently asked questions

Can a DNA diet test tell me the best diet for me?

Not with current evidence. Randomized studies have not shown that advice based on genetic information provides additional benefits over general advice or advice tailored to a person's current diet. Nor can your genotype predict whether a low-fat or low-carbohydrate diet will suit you better.

I carry a risk variant in the FTO gene; is losing weight impossible?

No. The FTO risk variant is associated with an average of a few extra kilograms; however, carriers respond to weight-loss programs involving diet, exercise and medication just as well as others.

If I have a slow caffeine-metabolizing genotype, should I stop drinking coffee?

There is no definitive genotype-based recommendation on this issue. General advice is that healthy adults should not exceed 400 mg daily (approximately 3-4 cups of filter coffee). Observing how coffee affects your sleep, palpitations or blood pressure, and consulting your physician if you have heart disease, is more informative.

How do my genes affect my cancer risk and diet?

Apart from some inherited cancer syndromes, findings that genetic differences interact with diet to alter cancer risk are limited and inconsistent. Risk factors such as red and processed meat, alcohol and excess weight matter for everyone, regardless of genotype. We discussed the relationship between nutrition, genetics and cancer in this article previously.

When is nutrition-related genetic testing truly necessary?

For specific medical questions, such as suspected lactose intolerance or celiac disease, familial inherited high cholesterol or iron accumulation disease, genetic tests performed after a physician's assessment are useful.

Are my genetic data safe?

Everyone having a test should ask this question. Genetic data are permanent and also concern your relatives. Under the KVKK, genetic data are special-category personal data; before testing, find out how your data will be stored, with whom they may be shared and whether you can request deletion.

Glossary

Nutrigenetics

The field studying how genetic differences affect responses to foods.

Nutrigenomics

The field studying how foods affect gene activity.

Epigenetics

Mechanisms determining when and how much genes function without changing the DNA sequence.

Polymorphism

Common small differences in DNA sequences between individuals.

Genotype

The combination of the two copies a person carries at a particular gene region (for example, AA, AC or CC).

Microbiota

All the bacteria and other microorganisms living in the gut.

Randomized controlled trial

A type of study that randomly assigns participants to groups and most reliably demonstrates the effect of an approach.