Two women can be 47 years old and not be living in the same body.

One sleeps relatively well, maintains stable energy, still has informative cycles and feels few symptoms. The other starts waking up hot, loses recovery, has brain fog, unpredictable cycles and the feeling that her body changed before anyone can point to a clear diagnosis.

The calendar age is identical. The hormonal experience is not.

That difference is one reason why talking about perimenopause only as "what happens around a certain age" falls short. The body does not age as one single block. The ovary, nervous system, metabolism, sleep and inflammation can lose stability at different speeds.

The useful question is not only how old you are. It is how coherent your hormonal signals are for the stage you are in.

Chronological age does not capture the whole transition

Chronological age helps orient average risk. It is true that ovarian reserve changes over the years and that the probability of the menopausal transition increases with age.

But a real woman does not live in averages. She lives in specific nights, specific meals, specific symptoms and weeks where her energy changes without the calendar explaining enough.

Functional ovarian age depends on several layers:

how many follicles remain and how they respond;
how the granulosa cells that surround the egg are working;
how stable the estradiol and progesterone signal is;
how loaded the stress axis is;
how you sleep and recover;
what level of inflammation or cardiometabolic load the body is carrying;
how coherent your rhythm of sleep, food, light and activity is.

That is why two 47-year-old women can have very different hormonal stories. One may be in a smooth transition. Another may feel that the whole system has become more fragile.

The difference is not imagination. It is individual biology.

What epigenetics has to do with it

Epigenetics is one way to explain how the body stores part of its history.

It does not change the DNA sequence. It changes chemical marks that influence which genes are activated, silenced or become louder. One of those marks is DNA methylation.

Epigenetic clocks, like the Horvath clock, use methylation patterns to estimate biological age in tissues. They do not measure hormones directly. They do not say "your estradiol is low". What they try to read is an accumulated footprint of aging, inflammation, metabolism, stress and cellular repair.

In population studies, menopause and reproductive history appear associated with biological age. A classic study published in PNAS found that greater epigenetic acceleration was associated with earlier menopause, bilateral oophorectomy and more time since menopause. More recent research in NHANES and in cohorts of women over 50 has found associations between age at menopause, reproductive life and methylation clocks.

But there is an important warning: association is not individual destiny.

A Mendelian randomization analysis published in 2024 did not find robust evidence that systemic epigenetic clocks directly cause age at menopause. Put simply: we cannot say "fast epigenetic clock = early menopause" as if it were a straight line.

The relationship seems more interesting and more complex: ovarian aging, systemic aging, inflammation and metabolism are coupled, but they are not the same thing.

The ovary does not age like blood

Many epigenetic clocks are measured in blood. That is practical, but it does not always answer the ovarian question.

The ovary has its own biology. Granulosa cells, which support the follicle and participate in hormone production, do not age the same way leukocytes do. They respond to FSH and LH, produce signals for the egg, participate in estradiol synthesis and live inside a very specific follicular microenvironment.

A study in Human Reproduction showed that granulosa cells had a different epigenetic trajectory than leukocytes: many more age-associated methylation regions in granulosa than in blood. Another 2024 study found that epigenetic acceleration in follicular fluid was associated with lower peak estradiol, fewer total oocytes and fewer mature oocytes in women undergoing IVF.

This does not mean everyone should measure their follicular fluid. It means something more conceptual:

ovarian age is not read well with a single metric.

It needs context. Ovarian reserve, symptoms, sleep, recovery, rhythm, inflammation and individual response.

Perimenopause can be felt before it shows up on the calendar

The popular conversation often waits for the cycle to become irregular. But many women feel changes before that:

they sleep, but do not recover;
they wake up hot;
they feel morning fatigue;
they have brain fog;
they tolerate alcohol, coffee or late dinners less well;
they have more anxiety or irritability;
they feel that one bad night costs them three days;
they notice that their energy no longer comes back the same way.

None of this diagnoses perimenopause on its own. But it can form a pattern.

When ovarian reserve declines, AMH can drop, FSH can rise and the estradiol and progesterone signal can become more variable. When that variability combines with fragmented sleep, more erratic cortisol, low-grade inflammation or loss of circadian rhythm, the subjective experience changes a lot.

It is not just "age". It is loss of coherence.

OEDI: a Lua Labs hypothesis

At Lua Labs we are working on a hypothesis called OEDI: Ovarian Epigenetic Discordance Index.

It is not a clinical test. It does not estimate real methylation. It does not say "your ovary is 53 years old". That would be a poor way to communicate science.

The idea is different: to observe whether a woman's longitudinal signals behave as if her hormonal system carried more load than expected for her stage.

An index like this would have to look at several layers:

age and hormonal stage;
cycle variability;
vasomotor symptoms such as hot flashes or night sweats;
sleep, energy, HRV and resting heart rate;
recovery after stress;
regularity of food, sleep, light and activity;
inflammatory food load or late dinners;
logging consistency and data confidence.

The right output would not be an alarmist number. It would be something more useful:

"This week your signals show more load than expected for your stage. The main contributors were fragmented sleep, heat-related awakenings and low recovery."

That does not replace a doctor. But it can help you arrive at a consultation with a clearer longitudinal story.

Why food enters this story

Lua is not a food app. But food is one of the signals the body uses to coordinate metabolism, inflammation, sleep and hormones.

A late dinner, alcohol or ultra-processed foods do not have the same effect on all women or in all phases. The impact depends on sleep, stress, hormonal stage, metabolic sensitivity, microbiome, timing and personal baseline.

In a woman with high circadian resilience, an imperfect night may be just a bad night. In a woman in hormonal transition, with more fragile sleep and low recovery, the same night can amplify hot flashes, fatigue or irritability.

That is why longitudinal logs matter. Not to blame you for what you eat. To understand when your body becomes more sensitive to certain combinations.

The question is not "is this food good or bad?". The question is:

what pattern appears when this food, at this time, in this stage, with this sleep and these symptoms, repeats?

Medicine needs context; so do you

A medical consultation usually has little time. Many women arrive with phrases like:

"I do not know what is happening to me."

"My tests come back normal."

"They tell me it is stress."

"I do not know if it is age, cycle, sleep or food."

The problem is that the body does not present its changes in separate columns. Perimenopause can appear mixed with work stress, broken sleep, digestion, cravings, temperature changes, pain, anxiety and tiredness.

Lua aims to build that longitudinal context: not a quick label, but a sequence.

What changed. When it changed. What repeated. What coincided with food, sleep, phase, symptoms and recovery.

Because a woman does not need another app telling her what day of her cycle she is on. She needs a tool that helps her understand why her body responds differently now.

What we can honestly say

Ovarian age is not the same as chronological age.

Epigenetic clocks are not a crystal ball.

Blood does not tell the whole story of the ovary.

Perimenopause does not always begin when the calendar becomes obvious.

And two women of the same age may need completely different explanations.

That is Lua's territory: longitudinal hormonal intelligence. Not universal tips. Not fear. Not automatic diagnosis.

Just a more serious question:

what is your body trying to say when its signals stop moving the way they used to?

Sources

Horvath S. Genome Biology (2013). DNA methylation age of human tissues and cell types.
Levine ME et al. PNAS (2016). Menopause accelerates biological aging.
Olsen KW et al. Human Reproduction (2020). A distinctive epigenetic ageing profile in human granulosa cells.
Hood RB et al. Human Reproduction (2024). Epigenetic age acceleration in follicular fluid and markers of ovarian response among women undergoing IVF.
Wang L et al. Frontiers in Endocrinology (2024). Exploring the causal association between epigenetic clocks and menopause age.
Daredia S et al. Clinical Epigenetics (2025). Timing of menarche and menopause and epigenetic aging among U.S. adults.
Herweck AM et al. Fertility and Sterility (2025). Prediction of aging pace and health risks by granulosa cell DNA methylation.
Xu YQ et al. npj Aging (2026). Reproductive life events and biological aging in women over 50.