Researchers analysed 11,000 gene samples from four different species, including humans, and found the same handful of genes ageing in the exact same way in every one of them. Then they built a clock out of it.
Aging research has long searched for a universal signature, something that ages the same way across species, that could be measured and used to track how fast an individual is actually aging, separate from how many birthdays they have had.
Epigenetic clocks, based on chemical modifications to DNA, were the first major tool to attempt this. They estimate biological age reasonably well, but they reveal little about what is actually happening inside the cell.
In a study published in Nature in late May 2026, researchers led by Alexander Tyshkovskiy and Vadim Gladyshev at Harvard Medical School and Brigham and Women's Hospital, with collaborators at Tohoku University, took a different approach. They analysed more than 11,000 transcriptomes, complete readouts of which genes are actively turned on, from over 25 tissues across four mammals: mice, rats, crab-eating macaques, and humans.
They found that specific genes age in nearly identical patterns across all four species. Genes involved in inflammation and immune activation consistently increase with age. Genes involved in mitochondrial energy production, wound healing, and extracellular matrix maintenance consistently decrease. Three genes in particular, GPNMB, CDKN1A, and LGALS3, emerged as universal markers, changing in almost identical ways during aging in every species studied.
The researchers then built what they call transcriptomic clocks. Unlike previous tools, these clocks do not produce a single number. Aging, the study found, is not one process. It is organised into separate modules, distinct groups of genes that handle specific biological functions. The team built individual clocks for each module, allowing them to measure how quickly different parts of a single body are aging relative to each other.
The clocks were validated against a substantial test: RNA sequencing data from genetically diverse mice exposed to 20 different pharmacological interventions from the Interventions Testing Program, including rapamycin and canagliflozin, drugs already known to extend lifespan in mice. The clocks correctly registered the effects of these interventions on biological age.
In human blood samples, the transcriptomic clocks correlated with established DNA methylation clocks, particularly within chromatin-related pathways, providing independent confirmation across two entirely different molecular measurement methods.
The study is correlational. It cannot establish that these gene expression changes cause aging rather than result from it. What it establishes is a shared molecular vocabulary of aging that exists, largely unchanged, across mice, rats, monkeys, and humans.
Four species. 11,000 samples. The same genes, aging the same way, in all of them.
Source: Nature, Volume 654, pages 173 to 188, 2026. DOI 10.1038/s41586-026-10542-3. Harvard Medical School and Brigham and Women's Hospital, with Tohoku University. Lead author: Alexander Tyshkovskiy. Senior author: Vadim N. Gladyshev.