Summary: The “aging clock” is a series of models that redefines age measurement. It improves upon the ordinary measure of age, which fails to capture a person’s overall health and appearance. Pioneered in 2011, the original aging clock model used DNA methylation patterns to predict biological age, offering a more accurate reflection of a person’s health. Progress since then has led to more advanced models, further impacting health by enabling earlier detection of age-related diseases.


When asked about one’s age, we tend to define it as years since birth. Yet, this number does not reflect a person’s overall health and appearance. Enter the “aging clock,” a groundbreaking model that redefines the measuring of age.

The concept of the “aging clock” was first developed in 2011 at the University of California, Los Angeles, where biostatistician Steve Horvath made a pivotal discovery. Horvath found that patterns of methylation, or changes to DNA activity, could predict a person’s age, since they highlight changes in the genes that control aging. DNA methylation patterns mostly reflect a person’s age, but they can also affect such genes to make people look older or younger than they are. Horvath took this knowledge and created an age predictor model exclusively using synthesized DNA from samples of body tissue and cells. He dubbed it the “Horvath Clock” – one of the first aging clocks developed that is still used in studies today.

Unlike traditional clocks, these clocks are regression-based models that model the relationship between independent and dependent variables. They use machine-learning tools to analyze DNA methylation data and generate a deeper understanding of aging.

But why are these clocks important? Since Horvath’s first aging clock, researchers have significantly advanced the capabilities of these clocks. The initial models primarily worked to predict chronological age, because they were limited to DNA methylation patterns as training data. As time has passed, researchers have expanded these models to predict biological age, using blood samples as well as cheek swabs to include more biomarkers beyond DNA methylation patterns.

Unlike chronological age, which only gives the number of years since birth, biological age reveals how well our bodies are functioning relative to our chronological age. Thus, measuring biological age revolutionizes health care by identifying individuals at risk of age-related diseases earlier, so they gain immediate access to treatments and preventive measures.

Recently, Horvath and his colleague Morgan Levine from Yale School of Medicine created two different clocks named DNAm PhenoAge and GrimAge, using glucose and white blood cells from a sample of the US population. These two clocks have been effective in predicting mortality and biological age. For example, Horvath used them to test 6,000 cancer samples from 32 datasets. In analyzing the results, he found that regardless of cancer type, the samples all showed signs of aging faster than normal. This is just one of the many studies using aging clocks that provide insights on how diseases progress in the human body.

The applications of such clocks are vast. According to Horvath, pairing the clocks alongside blood pressure and cholesterol tests at the clinical level provides a specific value of individual health, leading to early detection and intervention if necessary.

For young adults and children, having specific health advice tailored to them from the aging clocks is paramount in developing habits for a healthy lifestyle. At this stage, they are being applied in health care to improve the well-being of patients by accompanying existing forms of testing health, including heart tests. Thus, these clocks contribute to improving overall productivity and well-being of society.

However, questions remain about how these aging clocks are developed. Most of the clocks created by Horvath and other researchers have been based on data from blood proteins or bulk tissue, which contain many cells clustered together. From there, researchers have taken these clusters as data for the models. This method of development is good for predicting biological age. However, because the data are cell clusters, it is unclear how each individual cell type contributes to the prediction.

To address this problem, researchers within the Department of Genetics at Stanford have used RNA sequencing technologies to develop single cell aging clocks. These clocks can predict biological age through data from cell samples by utilizing RNA-sequencing technologies, which helps researchers analyze distinct aging patterns between different cell types to create a well-rounded prediction of biological age.

Besides the sampling questions, the practical application of these clocks to health care beyond clinics is still in development. Currently, these models are too nuanced for home use as individual units, although they can potentially complement at-home blood pressure and cholesterol tests. Ongoing research and global collaboration are crucial in advancing aging clocks from a prototype to practical tools in health care. The goal for these models is becoming not only a complement but a mainstay in health testing.

The current and future work on aging clocks is helping us understand the aging process better. In the short period of time since Horvath’s initial clock, researchers have continued to find greater methods of testing and measuring both chronological and biological age. This advancement highlights how much researchers can achieve in studying the complexity of aging and health. Through an expanded focus on biological age, researchers are redefining how we understand and manage aging and overall well-being.