Scientists have learned how to determine the age of DNA with an accuracy of 1.36 years
A method has been developed that makes it possible to determine the chronological age of a person with amazing accuracy — the error is only 1.36 years for people under 50. At the same time, the algorithm does not depend on gender, weight, smoking or other factors. All you need is a little blood, and artificial intelligence will tell you how old you really are. Not “according to the passport”, but from the point of view of molecules. This is not science fiction, but the result of a new study conducted by a team from the Hebrew University of Jerusalem.
The method is based on the analysis of DNA methylation, a process in which methyl groups (ch₃) are attached to DNA sections. These “chemical tags” do not change the genetic information itself, but affect its activity. And the most interesting thing is that methylation changes with age, leaving a kind of “time stamp” on the molecules.
After studying just two short regions in the genome, the team used deep neural networks (AI) to learn how to read these changes with an accuracy previously unavailable to any existing technique. The model was named MAgeNet (from the Methylation Age Network).
What is unique about this study?:
Only two DNA sections are used instead of hundreds or thousands, as in other methods.
Neural networks are used that work with individual molecules, rather than averaged data.
The accuracy is up to 1.36 years, which makes MAgeNet the most accurate tool today.
The result does not depend on smoking, BMI, gender or other external factors.
It is suitable for analyzing both chronological and biological age.
Understanding the exact biological age of a patient allows doctors to better select treatments, predict disease risks, and evaluate the effectiveness of anti-aging and anti-aging programs. Observers also note that MAgeNet can become a revolutionary tool in criminology: a DNA trace is enough to establish the age of an alleged criminal — without a passport, photos or witnesses.
The study found that the cells “record” age both in a random form and in the form of synchronous “flashes” of changes. This brings us closer to unraveling the molecular mechanisms of aging: “It turned out that time really leaves traces on our DNA. And now we can decipher them accurately,” explains Professor Tommy Kaplan, co—author of the study.
“This is a new window into the nature of aging at the cellular level. A vivid example of what happens when biology meets AI,” adds Professor Yuval Dor.
The ability to determine age by DNA opens the door to personalized medicine, accurate diagnosis, and a revolution in legal expertise. It also leads to a deeper understanding of how the body counts down time. Not according to the calendar, but according to its internal, molecular clock.
Published
July, 2025
Duration of reading
2-3 minutes
Category
Science
Share
Don’t miss the most important science and health updates!
Subscribe to our newsletter and get the most important news straight to your inbox