What are Horvath's clock, GrimAge, and PhenoAge? We explain the science of DNA methylation clocks and how they are changing the way we understand, measure, and — crucially — intervene in the aging process.
What are Horvath's clock, GrimAge, and PhenoAge? We explain the science of DNA methylation clocks and how they are changing the way we understand, measure, and — crucially — intervene in the aging process.
What are epigenetic clocks?
Epigenetic clocks are mathematical models that predict biological age from patterns of DNA methylation — chemical modifications to DNA that regulate gene expression without altering the underlying genetic code. These clocks exploit the observation that methylation patterns at specific sites across the genome change in a highly predictable manner as we age.
Unlike genetic sequence (which remains fixed throughout life), the epigenome is dynamic — shaped by lifestyle, environment, stress, nutrition, and disease. This makes epigenetic clocks not just mirrors of the past, but potentially modifiable windows into the biological future.
Horvath's pan-tissue clock (2013)
Steve Horvath, then at UCLA, published the first multi-tissue epigenetic clock in 2013. By analysing DNA methylation at 353 CpG sites, Horvath's clock could predict chronological age across virtually all human tissue types with a mean absolute error of just 3.6 years — a remarkable achievement.
The pan-tissue property was critical: a single model could analyse methylation in blood, saliva, brain, or biopsy tissue and yield a consistent biological age estimate. This universality suggested that epigenetic aging is a fundamental biological process, not a tissue-specific artefact.
GrimAge: predicting death, not birthdays
Subsequent clock developers recognised that predicting chronological age — however accurate — was less clinically useful than predicting health outcomes. GrimAge, developed by Horvath and colleagues in 2019, was trained to predict lifespan directly.
GrimAge analyses plasma proteins (captured via DNA methylation proxies) alongside smoking pack-years. It predicts time to death and time to disease onset more accurately than any previous clock, and its "GrimAge acceleration" (biological age minus chronological age) is a powerful predictor of cancer risk, cardiovascular disease, and all-cause mortality.
PhenoAge: clinical biomarkers as training signal
PhenoAge, developed by Morgan Levine and colleagues, takes a different approach. Rather than methylation patterns at specific sites, PhenoAge starts from a composite of nine standard clinical biomarkers (albumin, creatinine, glucose, CRP, lymphocyte percentage, mean cell volume, red blood cell distribution width, alkaline phosphatase, and white blood cell count) combined in a weighted formula.
A second-stage DNA methylation clock was then trained to predict this composite phenotypic age — creating a methylation-based surrogate that captures the clinical health information embedded in standard blood tests.
"Epigenetic clocks have given us, for the first time, an objective, quantitative measure of biological age — transforming aging from a philosophical concept into a measurable biological variable." — Longevity World Group
DunedinPACE: measuring the speed of aging
While most clocks estimate a static biological age, DunedinPACE measures the pace of aging — how fast an individual is biologically aging at a given moment. Developed from the Dunedin Study (a 50-year longitudinal study of New Zealanders), DunedinPACE correlates strongly with physical and cognitive decline, and responds measurably to interventions — making it a powerful tool for clinical trials.
Can epigenetic clocks be reversed?
This is perhaps the most consequential question in longevity science today. Several interventions have demonstrated measurable reductions in epigenetic age:
- Caloric restriction — consistent slowing of epigenetic aging in multiple species
- Exercise — particularly aerobic and resistance training in combination
- DHEA supplementation — modest reversal observed in the TRIIM trial
- The Horvath "cocktail" — GH/DHEA/metformin combination showed ~2.5 year epigenetic age reversal in the TRIIM-X trial
- Lifestyle optimisation — the DNAMAGE study showed ~3 years reduction via diet, sleep, exercise, and stress management
These findings remain preliminary, but they establish proof of concept that epigenetic age is not destiny — it can be influenced by interventions accessible today, and potentially by more powerful pharmacological tools in development.
