What is epigenetics? It is the study of changes in how genes are used that do not alter the DNA sequence itself. The letters stay the same; what changes is which genes are switched on, how strongly, and in which cells. That one idea explains why a liver cell and a neuron can contain identical DNA and behave nothing like each other.

It also explains why the field attracts so much hype. If the environment can leave marks on how genes behave, the thought goes, then diet, stress, and exercise must be rewriting your genetic destiny, and maybe your children’s. Parts of that are supported by evidence. Other parts are speculation dressed up in lab vocabulary. This article tries to keep them apart.

What is epigenetics in plain terms?

Think of DNA as a very large cookbook shared by every cell in your body. Epigenetics decides which recipes each cell reads. The text of the book does not change, but sticky notes, bookmarks, and folded corners guide the cook.

The term goes back to 1942, when C. H. Waddington coined it. A fuller modern definition, from Robin Holliday in 1990, describes the study of the temporal and spatial control of gene activity during development. In everyday use today, the word covers chemical tags on DNA and on the proteins it wraps around, plus several kinds of regulatory RNA.

The CDC describes the key feature this way: unlike mutations, epigenetic changes can be reversible, and they act as switches that turn genes on and off. The agency adds that the reversibility holds only in some cases.

How does DNA methylation work?

Methylation is the best-studied tag. A small chemical group called a methyl group is attached to a cytosine, mostly at spots where a cytosine sits next to a guanine (called CpG sites). Methylation in a gene’s promoter region, the part that starts it up, often silences the gene. Methylation inside a gene body is more often linked to higher expression, so the effect depends on where the tag lands.

Cells copy the pattern when they divide. An enzyme called DNMT1 acts as a “maintenance” methyltransferase: after DNA replication it recognizes the half-methylated new strand and adds the missing tags. Other enzymes, the TET family, remove methyl groups. Hundreds of such removals occur in neurons during learning and memory formation.

That is a key point. Epigenetic marks are dynamic. They shift during development, with age, and in response to the environment.

What are histones and why do they matter?

DNA in your cells is wrapped around spool-like proteins called histones. Chemical modifications to the tails of those proteins change how tightly the DNA is packed, and so how reachable it is. Acetylation of histone H3 at certain positions is generally linked to active genes, while methylation at the position called K9 is linked to silencing.

An older idea held that these modifications form a “histone code,” a kind of digital script. That notion has been largely set aside. The marks interact in messier ways, and how histone states are passed on when cells divide is less well understood than for DNA methylation.

Do diet and lifestyle really change your epigenome?

Some do, in measurable ways. The cleanest example comes from smoking. Smokers tend to have less methylation at a gene called AHRR than nonsmokers do. After quitting, former smokers can gradually drift back toward nonsmoker levels, sometimes within a year, depending on how long and how heavily they smoked. That makes AHRR a useful marker, though a marker is not a mechanism, and the tag is a signal that smoking occurred, not proof that it caused harm through methylation.

Other findings are looser. Reviews suggest exercise builds an “epigenetic memory” that may influence long-term brain function, and lab studies of certain foods and plant extracts show effects on markers of DNA damage over hours or weeks. These results are typically small, short, or based on animals or cell cultures. They do not support the claim that a particular diet reprograms your genes in a lasting way.

Chemicals and infection matter too. Benzene, styrene, and several solvents cause considerable DNA hypomethylation, meaning a loss of methyl tags. And tuberculosis bacteria can silence a gene called IL-12B in immune cells, which weakens the immune response.

Many findings in this area are associations. Two things moving together do not show that one caused the other.

What did the Dutch Hunger Winter show?

It is the most cited natural experiment in the field. During the winter of 1944 to 1945, part of the Netherlands was left severely short of food. Researchers later followed people who had been in the womb at the time.

The Dutch Famine Birth Cohort Study found that children of mothers who were pregnant during the famine were more susceptible to diabetes, obesity, cardiovascular disease, and other problems. Timing mattered. Later work on exposure in the second trimester linked it to higher rates of schizophrenia and neurological defects. And the CDC notes that people whose mothers were pregnant during the famine showed different methylation patterns decades later, which may help explain higher risks of heart disease and type 2 diabetes.

Be careful with the word “may.” The authors of this research describe the epigenetic link as a possible explanation, not an established mechanism. Prenatal malnutrition can reshape development in several ways, and methylation is one candidate.

What are epigenetic clocks?

In 2013, two teams published methods that estimate age from DNA methylation. Steve Horvath at UCLA built a multi-tissue clock using 353 CpG sites, trained on about 8,000 samples from 51 healthy tissues and cell types. It tracks chronological age closely (a correlation of 0.96), with a median error of 3.6 years. The Hannum clock, from teams led by Trey Ideker and Kang Zhang at UC San Diego, uses 71 markers in blood.

Later “second-generation” clocks, such as PhenoAge from 2018 and GrimAge, were trained to predict health and mortality risk rather than age alone, and GrimAge incorporates smoking-related measures.

Two cautions apply. First, nobody knows exactly what a methylation clock measures. Horvath himself described it as an emergent property of the epigenome, and the specific CpG sites likely have no direct causal effect on aging. Second, clocks have limits: they underestimate age in older people, and accuracy depends somewhat on where the DNA came from.

As for lifestyle, associations between epigenetic age acceleration and habits are weak in blood samples. Cross-sectional studies link education, a plant-heavy diet, and physical activity to slower epigenetic aging, while heavy drinking, higher body mass index, and metabolic syndrome markers are linked to faster aging. These are correlations in populations. A commercial “biological age” test is a different thing from a validated medical tool.

Can epigenetic changes be inherited?

Within a lifetime, yes: cells pass their marks to daughter cells, which is how a liver stays a liver. Across generations is where the debate gets heated.

In mammals, most marks are erased twice, once after fertilization and once in the developing germ cells that will become eggs and sperm. That makes passing an acquired mark to children, let alone grandchildren, harder than the popular accounts imply. Some marks resist erasure, and imprinted genes, where the mark depends on which parent supplied the gene, are a well-established exception.

Evidence for inheritance is strongest in plants and simple animals. In the roundworm C. elegans, starvation effects have been passed on through small RNAs. In maize, a silenced state at the b1 gene has persisted across generations for fifty years of experiments. Rice exposed to simulated drought showed greater drought tolerance after eleven generations, tied to heritable methylation changes.

In humans, the evidence remains limited. Some analyses of the Dutch famine and a Swedish study of men exposed to famine in the 19th century found effects in grandchildren. These findings are often attributed to developmental programming in the womb, not proven epigenetic inheritance. A 2018 paper by Bernhard Horsthemke, titled “A critical view on transgenerational epigenetic inheritance in humans,” argues for skepticism, and a 2026 review of war and genocide cohorts concluded that human observational studies cannot establish true epigenetic inheritance, citing small samples and shared social and cultural influences.

A mouse study on odor and fear in offspring attracted critics, who cited low statistical power. The honest status of transgenerational inheritance in people is “unproven.”

Is epigenetics used in medicine?

In a few places, yes. Abnormal methylation patterns are associated with cancer and can help distinguish cancer types. The stool-based Cologuard screening test looks for abnormal methylation as one of its markers in colorectal cancer, although a positive result requires a colonoscopy to confirm. The CDC notes that epigenetics alone cannot diagnose cancer.

In cancer cells, some genes are more methylated, yet overall methylation is lower than in normal cells. The picture is a rearrangement of the pattern, not a simple rise or fall.

Common questions

Does epigenetics mean my genes can change?

Not in the sense of the sequence. The DNA letters stay the same, but how genes are used can shift. Some of those shifts are reversible.

Can I reverse aging with diet or exercise?

Not on current evidence. Habits are associated with slower epigenetic aging in population studies, but an association is not a treatment. Be cautious about products that promise to turn back your biological clock.

Is it true trauma is passed down in the genes?

It is not established in humans. Studies of descendants of famine and war report effects, but critics say the research cannot separate epigenetics from family environment, culture, and other shared influences.

Are epigenetic tests reliable?

Research clocks have measurable accuracy for chronological age. Commercial tests vary, and what the results mean for your health is still being worked out. Discuss anything that affects health decisions with a clinician.

What is the difference between a mutation and an epigenetic change?

A mutation alters the DNA sequence and is permanent in that cell line. An epigenetic change alters how the sequence is read, and it can sometimes be undone.

What to remember

Epigenetics is real, well documented in development, aging, and disease, and genuinely interesting. Cells use it to decide what they are. Smoking leaves a measurable trace. Famine before birth is linked to health effects decades later.

The leap from there to “you can rewrite your genes with a green smoothie,” or to inheriting your grandparents’ trauma, runs well ahead of the data. Stay with the claims that name a mechanism and a measurement, and treat the rest as open questions.

Further reading

Accessible popular-science introduction to the field. See it on Amazon (affiliate link)

As an Amazon Associate I earn from qualifying purchases. A commission may be earned if you buy through this link, at no extra cost to you.

Sources