DNA diet and biological-age tests: what do they tell you?
Consumer DNA tests may suggest dietary advice from genetic variants, while biological-age tests estimate ageing-related patterns in a sample. These are different tests, and neither provides a useful treatment or eating plan on its own. Before paying for one, ask what decision the result would change, how reliable that change is and what happens to your sample and data.
What a consumer DNA test reads
Many consumer DNA kits analyse saliva or a cheek swab using an array that checks selected genetic variants. Some services use sequencing instead. Check what the product tests: it may not read the whole genome or detect every clinically relevant variant.
The company builds the report by matching a person's variants against published studies that found an association between a variant and a trait. The report is only as strong as those studies, and researchers ran most of them on people of European ancestry. For Chinese, Malay and Indian Singaporeans, some associations transfer well and some do not, and the report rarely says which.
Clinical genetic testing is chosen for a specific medical question and can use several methods, including targeted variants, gene panels or broader sequencing. A clinician or genetic counsellor interprets the result with the person’s history. It does not always involve reading every relevant gene in full.
Can genes identify the best diet for you?
A handful of single-gene traits have a clear link to food, and these are established.
- Lactase persistence. Several genetic variants influence whether lactase production continues into adulthood. Symptoms and tolerance also depend on the amount eaten and the individual. A consumer result does not diagnose every cause of digestive symptoms after milk.
- Alcohol flushing. An ALDH2 variant common in East Asian populations can reduce acetaldehyde breakdown. Drinking with this variant increases certain cancer risks. Flushing is a warning sign, not a complete genetic test or proof that someone without flushing is protected.
- Caffeine clearance. Variants in a liver enzyme change how fast kopi leaves the system. The practical effect is modest, and a person who cannot sleep after a 4 pm kopi already knows.
- Coeliac risk genes. Certain HLA types are present in almost everyone with coeliac disease but also in many people without it. Absence makes the disease very unlikely; presence does not diagnose it. If you suspect coeliac disease, see a clinician before removing gluten, because doing so can affect diagnostic testing.
Researchers use combinations of variants, called polygenic scores, to estimate disease risk. Their usefulness varies by condition, population and clinical setting. A score for disease susceptibility is not the same as evidence that a company’s personalised diet improves health.
The marketing part is the diet plan. Reports that say a person should eat low-carbohydrate or low-fat because of their genes draw on associations far weaker than they imply. A well-known 2018 trial in the United States assigned volunteers to a low-fat or a low-carbohydrate diet and found that a genetic pattern did not predict who would lose more weight. A European trial found that adding genetic information to ordinary personalised dietary advice produced no extra benefit. The same saliva sample sent to two companies can come back with different advice, because each company chooses its own studies.
What biological-age tests estimate
DNA carries chemical tags called methyl groups, and the pattern of those tags shifts with age in a predictable way. In 2013 researchers showed that reading the tags at a few hundred positions could predict a person's age to within a few years. Researchers call these estimators epigenetic clocks. Later teams trained clocks to predict disease and death, and the newest ones estimate the pace of ageing.
The established part is that methylation age is a valid biomarker. Across large groups, people whose clock runs ahead of their birthday have higher rates of disease and earlier death, and the effect survives adjustment for smoking, weight and income. Researchers use the clocks to test whether an intervention slows ageing, and that is promising work.
The marketing part is the individual number sold by post, which is hard to act on.
- Repeat samples can differ. Laboratory methods, the sample and the clock used affect the result. Ask for the test’s own repeatability data and uncertainty range; there is no single error margin that applies to every biological-age product.
- Different clocks disagree. One clock may say a person is five years younger and another five years older, and both are working as designed.
- Recent events move it. An infection, poor sleep or a stressful month can shift the reading.
- There is no agreed threshold. No guideline says what to do with a biological age of 47 in a 42-year-old.
A lower score after a supplement or retreat does not by itself show that ageing slowed or health improved. The change needs to exceed the test’s uncertainty and be supported by evidence connecting that intervention to meaningful health outcomes.
Costs, consent and genetic information
Costs vary by test, follow-up and any subscription. Check the full terms and whether interpretation is included before paying. Consumer wellness testing is different from medically indicated genetic testing, which may have financing support through specific approved clinical pathways.
Genetic information is sensitive and can also reveal information about relatives. Ask where the sample is analysed, whether it will be retained, who may access the data and how requests to withdraw consent or delete an account work. Overseas processing does not, by itself, remove a Singapore organisation’s data-protection responsibilities.
How this compares with routine clinical care
Routine cardiovascular risk assessment uses current information such as blood pressure, glucose, lipids, smoking, age and medical history. Healthier SG Screening supports appropriate screening for eligible adults, commonly from age 40. These results can lead to established prevention or treatment decisions.
Family history is valuable clinical information. Tell your doctor about relatives with early heart disease, diabetes, cancer or a known inherited condition, including their age at diagnosis. It can help determine whether clinical genetic assessment is needed; it does not replace a genetic test when one is indicated.
What to do before or after testing
Consumer DNA and biological-age reports can be interesting, but a result is useful only if its accuracy, meaning and next step are clear. Evidence for selecting a weight-loss diet from common genetic variants is limited. A biological-age score is not a diagnosis or a reason to change medicines.
If you are worried about an inherited condition, raise it with your polyclinic doctor, who can refer you onward. If you have already bought a test and are anxious about the result, bring the report to the same doctor rather than to a search engine.
A practical next step
For a suspected inherited condition, start with your doctor and family history. If you already have a consumer report, bring it to the appointment. Do not change treatment or buy supplements with the sole aim of improving a biological-age score.
Sources and further reading
These references explain the guidance and research discussed above. For advice about your own health, speak to a qualified healthcare professional.
Read next
Blue Zones and longevity diets: what holds up
Five regions
Healthspan and lifespan: what ageing well actually means
Two words that sound alike, one of which is worth planning around