Study shows that biological age tests measure different processes and are not interchangeable
Researchers discovered that five popularly used biological aging clocks track very distinct cellular and molecular processes. This finding, released on August 14, 2026, suggests that the tools should not be seen as interchangeable for issues related to aging.
A new methodology based on gene expression, developed from these discoveries, has demonstrated greater accuracy in some cases in predicting health outcomes such as diabetes, heart problems, lung problems, frailty and mortality. However, performance has varied, and original watches retain value for certain outcomes.
Although no Reactome pathway was shared among all five clocks, a broader Gene Ontology analysis identified overlap in fundamental biological processes. Immune system activity and inflammation, for example, were present in four of the five measuring instruments.
The comprehensive study, detailing that biological clocks track something surprisingly different at the cellular level, was published in the journal npj Aging. The new gene-based approach showed stronger associations with some health outcomes, although results varied across outcomes and external datasets.
More on this story: Estrogen hormone therapy in older women may reduce signs of Alzheimer’s
People born in the same year may have significant differences in appearance, sensation and function, as if they belonged to different generations. To measure this variation more accurately, scientists have developed tools known as “epigenetic clocks” over decades. These clocks interpret chemical markers in DNA, patterns that change as the body ages, and convert them into an estimate of how quickly a person’s body ages compared to their chronological age. Despite widespread use in research and growing presence in clinical and commercial settings, what these clocks actually measure internally was, until then, a mystery. The recent study began to provide explanations on this topic.
The diversity of approaches in five biological clocks
The research focused on five biological aging clocks: Horvath, Hannum, PhenoAge, GrimAge and DunedinPACE. Each was designed with a particular strategy. Some were calibrated to estimate chronological age, while others were created to predict disease risk, remaining lifespan, or the rate of physical and mental decline over time. Researchers and healthcare professionals use all of these clocks, often without a clear understanding of which would be most appropriate for a specific issue.
Scientists from the University of Southern California, UCLA, Indiana University and the University of Minnesota examined data from 3,227 elderly Americans. By combining two types of biological information obtained from the same blood samples – measurements of chemical markers in DNA and of genes that were being turned on or off – they were able to map the gene activity and biological pathways associated with each clock. Clocks, which are often treated as equivalent measurements, turned out to be, in fact, quite distinct biological indicators.
Information was collected from an ongoing, nationally representative study of Americans over 50. Participants had an average age of 70, were 58% female, and represented a variety of racial and ethnic backgrounds. Blood samples were collected in 2016, and health and survival data were monitored until 2022.
Learn more: Daily multivitamin can slow down biological aging in the elderly, study reveals
When the team investigated which genes were activated or deactivated in people with higher scores in each biological clock – indicating accelerated biological aging – the differences were significant. The Horvath clock was associated with changes in just 49 genes, while the DunedinPACE was linked to more than 3,200 genes. A crucial aspect was the near absence of overlap: no gene was common to all five clocks.
Understanding what each biological clock tracks in the body
Grouping individual genetic signals into broader biological themes provided clarity about what each clock appears to capture.
Horvath, the oldest of the five and designed to estimate age in different types of tissue, was shown to be linked to metabolic processes and basic cellular communication. Their signals indicated essential cellular maintenance functions rather than specific diseases.
Hannum’s clock, also older and developed specifically from blood samples, showed stronger connections to immune system function, inflammation and blood vessel activity. These patterns are consistent with how the immune system gradually degrades and becomes more susceptible to chronic inflammation as we age.
See Also
Poor diets increase microplastic damage to the liver, research reveals
Connection between a tiny area of the brain and the quality of deep sleep discovered
Research points to link between virgin olive oil and brain protection in the elderly
PhenoAge, created to predict a set of clinical health measures related to mortality risk, revealed associations with cellular aging, stress responses and signs of uncontrolled cell division. This aligns with its purpose of capturing broad, systemic biological stress.
GrimAge, which has consistently been shown to be the most effective predictor of mortality in published studies, has been linked to immune surveillance pathways. Although it was related to more changes in gene activity than PhenoAge, it mapped a narrower set of biological themes. This pattern suggests a signal of concentrated stress, formed in part by markers of smoking exposure and inflammation.
DunedinPACE, the most recent of the five and designed to measure the speed of aging rather than a static biological age, exhibited the greatest range of all. It covered protein metabolism, immune signaling, development of the nervous system, cellular energy production, and reduced activity in the pathways that cells use to repair DNA damage.
On the same topic: Vitamin D deficiency and abdominal fat double the risk of death after 50 years
Despite these distinctions, some themes were repeated across multiple watches. Immune system activity, especially the type of chronic, low-grade inflammation that tends to increase with age, has emerged recurrently. Pathways linked to the body’s response to infection and internal damage were active in four of the five clocks, making immune activity one of the most obvious signals shared between them.
How biological clocks compare in their genetic activities
The study provided a detailed overview of how the five biological age clocks differ in their genetic activities and associated biological pathways.
Innovation: a new gene-based score with greater accuracy
Based on these findings, scientists developed a new scoring system called Transcriptomic Aging Gene Scores, or TAGS. Instead of reading chemical markers in DNA, TAGS directly measure the intensity of expression of genes linked to each aging clock at a given time. This approach is analogous to differentiating between a rotary knob set to a specific position and checking whether the switch it controls is on or off.
The TAGS were developed using 80% of the sample and subsequently evaluated on the remaining 20%, a group of 645 individuals whose data were reserved specifically for this purpose. For several health outcomes measured until 2022, including deaths, frailty, difficulty performing daily activities, diabetes diagnosis, heart and lung problems and levels of an inflammatory marker such as interleukin-6, TAGS scores showed stronger associations with some outcomes than the original epigenetic clocks.
More on this story: Research indicates that low fiber intake can lead microbes to attack intestinal lining
For diabetes, heart problems, lung problems, and markers of inflammation, TAGS scores remained significant predictors even when directly compared to their parental clocks in the same statistical model, although performance varied by clock and outcome. It is notable that when evaluated in the independent test sample, all five TAGS scores predicted all-cause mortality, while the two oldest clocks, Horvath and Hannum, did not do so significantly in isolation. At the same time, specifically for mortality and frailty, GrimAge and DunedinPACE still showed distinct contributions along with the TAGS scores. This serves as a reminder that the original clocks retain real value for certain outcomes, even as the new scores add predictive power for others. The researchers also tested their TAGS scores on three independent external data sets, with varying results. The associations were strongest in groups with Parkinson’s disease and early-stage cardiovascular disease, but weakest in a group with major depressive disorder.
One area in which neither the original watches nor the new TAGS showed reliable signals was psychological health. Associations with mental health outcomes were consistently weak across all analyzes performed.
Choosing the right biological clock for each application
Because each biological clock monitors different biology, scientists emphasize the importance of choosing the correct clock. For example, they suggest that GrimAge, connected to immune and inflammatory activity, could be more appropriate for an anti-inflammatory study than Horvath, which appears to be largely disconnected from immune processes. DunedinPACE, with its broad coverage across multiple body systems, may be better suited to monitoring whether interventions such as physical exercise are slowing the overall rate of decline.
These findings come at a time when biological clocks of aging are rapidly being translated from research laboratories to clinical settings and consumer products. In some of these contexts, the five clocks are treated as roughly equivalent windows for biological age, which is a mistake. This study strongly argues that they are not interchangeable, and treating them as such can lead to erroneous conclusions about the biology of aging, affecting the accuracy of diagnoses and the effectiveness of interventions.
For anyone who has ever received an “epigenetic age” number from their doctor or research team, the message of this study is both comforting and complex. It’s comforting because genetic clocks are linked to real, measurable differences in gene activity. However, it is complex, as what exactly they measure depends largely on which watch was used, a detail that, until now, has remained largely unexplained to the public and even many professionals.
