Epigenetics: why your genes are not your destiny

Epigenetics is the reason "your genes are not your destiny" is partly true, but not a magic rewrite of biology. Your DNA sequence is mostly fixed, while epigenetic marks help control which genes are switched up, down or kept silent in particular cells. Diet, smoking, exercise, sleep, inflammation, age and early-life environment can all be associated with epigenetic changes, but commercial claims that you can precisely reprogramme your genes with a supplement or test are far ahead of the evidence.

What epigenetics actually means

Your genome is the DNA sequence. Your epigenome is a layer of chemical and structural information that helps cells decide which genes are accessible, quiet or active. The National Human Genome Research Institute describes DNA methylation as one major mark, while histone modification changes the proteins around which DNA is wrapped.¹

That distinction matters. Epigenetics is not a mystical override switch. It is part of normal biology: embryo development, cell identity, immune responses, cancer biology, ageing research and adaptation to environment. A muscle cell and a skin cell do not differ because they have different DNA. They differ because different genetic programmes are active.

Some epigenetic marks are responsive. Others are tightly controlled. A single meal, workout or stressful day can change gene expression temporarily, but that is not the same as permanently improving your genome. Most health-relevant claims need to answer three questions: which tissue changed, which mark changed, and whether that change caused a meaningful health outcome.

What changes epigenetic marks

The strongest human examples are not exotic. They are smoking, ageing, body composition, severe early-life exposures, disease, inflammation and cell-type shifts in blood. These influences can leave measurable methylation patterns, but causality can be hard to prove because illness can change the epigenome as well as the other way round.

The Dutch Hunger Winter is a famous example. Heijmans and colleagues studied people prenatally exposed to famine in 1944 to 1945 and found, about six decades later, lower DNA methylation of the imprinted IGF2 gene compared with same-sex siblings who were not exposed in the womb.² This does not mean every stressful pregnancy permanently damages a child. It shows that severe developmental exposures can leave long-lasting molecular traces.

Smoking is clearer at population scale. Joehanes and colleagues analysed blood DNA methylation from 15,907 people across 16 cohorts. Current versus never smokers differed at 2,623 CpG sites using a strict genome-wide threshold, with 18,760 CpG sites at false discovery rate below 0.05.³ That is a large biological footprint, not a vague wellness claim.

Adiposity is also linked with methylation patterns, though cause and effect can run both ways. Wahl and colleagues reported BMI-associated methylation differences and used genetic analyses suggesting many methylation changes were consequences of adiposity rather than primary causes.⁹

Exercise also changes gene regulation in human tissue. Barres and colleagues showed that acute exercise in human skeletal muscle was associated with promoter hypomethylation and increased expression of metabolic genes such as PGC-1 alpha, PDK4 and PPAR delta.⁴ A 2024 systematic review of randomised trials in previously inactive adults concluded that exercise can affect DNA methylation, but the field is still early and dose-response patterns are not settled.¹¹

Epigenetic clocks and biological age

Epigenetic clocks use methylation patterns at selected DNA sites to estimate age-related biology. Steve Horvath's 2013 multi-tissue clock used 353 CpG sites and could estimate DNA methylation age across many human tissues and cell types.⁵ Later clocks tried to predict healthspan, lifespan or pace of ageing rather than calendar age alone.

DNAm PhenoAge, GrimAge and DunedinPACE are examples of newer clocks. Levine and colleagues developed PhenoAge to reflect lifespan and healthspan-related biology across large cohorts.⁶ Lu and colleagues reported that DNAm GrimAge strongly predicted lifespan and healthspan.⁷ Belsky and colleagues developed DunedinPACE as a methylation marker of the pace of ageing, using data from the Dunedin Study and validating against morbidity, disability and mortality-related outcomes.¹²

These tools are useful for research, but an individual result can be noisy. Different clocks answer different questions. A saliva test, blood test and cheek swab may not agree. A two-year change on a report is not the same thing as adding two healthy years to your life.

What lifestyle trials show

The most quoted lifestyle trial is the 2021 pilot study by Fitzgerald and colleagues. It randomised 43 healthy men aged 50 to 72 to an eight-week programme including diet, exercise, sleep guidance, relaxation and selected supplements, versus control. The intervention was associated with a 3.23-year lower Horvath DNA methylation age compared with controls.⁸

That is interesting, but it was small, short, male-only and used a methylation clock outcome. It does not prove fewer heart attacks, less dementia or longer life. The right interpretation is "worth studying", not "buy a methylation stack".

The larger 2025 DO-HEALTH methylation analysis looked at 777 older adults from a three-year trial of vitamin D 2,000 IU per day, omega-3 1 g per day and a home exercise programme. It reported small protective effects of omega-3 on several DNA methylation clocks, with an additive effect of omega-3, vitamin D and exercise on PhenoAge.¹⁰ The effect sizes were small. The public-health lesson is still the combined basics: diet quality, movement, sleep and risk-factor control.

What not to buy

Be cautious with direct-to-consumer epigenetic tests that sell precise age scores, supplement bundles or monthly "methylation optimisation". The measurement may be technically real, but the interpretation can be thin. A high clock result may reflect smoking, inflammation, blood-cell mix, body weight, illness, assay noise or the clock chosen. It is not a diagnosis.

Also be careful with methyl donors. Folate, B12, choline and betaine matter in one-carbon metabolism, but more is not automatically better. High-dose methylation supplements can cause side effects, interact with medicines, mask B12 deficiency in some contexts, or distract from validated care. If you have anaemia, neuropathy, pregnancy, cancer treatment, epilepsy medication, kidney disease, bipolar disorder or complex prescriptions, ask before experimenting.

How to use this practically

The useful version of epigenetics is not "hack your genes". It is a reminder that biology is responsive. You cannot choose your inherited variants, childhood exposures or age, but you can reduce the exposures most likely to damage the system: smoking, inactivity, poor sleep, excess alcohol, uncontrolled blood pressure, untreated diabetes, chronic stress without support and poor-quality diet.

Use Start Here to list your modifiable risks before buying a test. Use the stack builder if you are running a careful experiment, one change at a time. For wider prevention and evidence checks, use the health library and insights section.

Epigenetics is empowering when it makes health feel modifiable. It becomes misleading when it turns complex biology into a personalised sales funnel. Your genes are not destiny, but the fundamentals still beat the hacks.

This article is educational and does not constitute medical advice, diagnosis, or a treatment recommendation. Medication uses described as “off-label” are not licensed for that purpose in the UK and should only be considered under qualified clinical supervision. Always speak to your GP, pharmacist, or a registered specialist before starting, stopping, or changing any treatment. If you have severe or alarm symptoms - unintentional weight loss, blood in your stool, difficulty swallowing, persistent vomiting, a fever, or severe pain - seek urgent medical care.