A team of researchers at the Hebrew University of Jerusalem has developed an exceptionally accurate method for predicting chronological age from DNA, based on two short genomic regions. The findings of the study were published in the journal Cell Reports.

Unlike other methods that require the analysis of large amounts of genetic material, MAgeNet evaluates only two small regions of DNA, where so-called methylation "tags" accumulate.

This process consists in the attachment of methyl groups to DNA molecules, which does not change the sequence of genes, but regulates their activity. With age, the "picture" of methylation becomes more characteristic, and this enables to determine how "worn out" the human body is.

Researchers used artificial intelligence to decipher the changes, teaching neural networks to read the molecular “signatures” of aging. This achieved age predictions with a median error as low as 1.36 years in individuals under 50.

This is not just an epigenetic clock, but a complex mechanism where random factors combine with clear patterns, explained Professor Yuval Dor, a supervisor of the study,.

Moreover, researchers found that the aging process in DNA is recorded not linearly, but in “jumps,” and it is as if cells periodically check their internal clock.

According to the researchers, the use of this new methodology in medicine will help doctors more accurately assess the biological age of a patient, identify the risks of developing diabetes, dementia or cardiovascular diseases.

In gerontology, MAgeNet can make it possible to find out how lifestyle, diet, or medication affect cellular processes.

Another advantage of this study is its low cost. Since it requires the analysis of only two regions of DNA, it is considerably cheaper than sequencing the entire genome. However, so far it has only been tested on healthy volunteers. Researchers still need to test how accurately MAgeNet works with the genomes of people with serious diseases, in which the methylation pattern may be disrupted.