Organs do not age in lockstep. The liver peaks at 40, the brain at 50.
For generations, aging has been observed but rarely measured with precision — a process felt in the body long before it could be quantified in a laboratory. A consortium of Chinese researchers has now built a computational framework called the Digital Aging Twin, published in Cell in May 2026, that maps how individual organs age at different speeds within the same person, and identifies specific liver-derived proteins not merely as signs of aging, but as its active architects. In doing so, science has moved from the language of correlation into the harder, more consequential language of cause.
- Aging has never been a single process — two people born the same year can have organs a decade apart in biological age, and medicine has lacked the tools to see this clearly until now.
- Researchers recruited over 2,000 healthy adults across four Chinese cities, gathering more than a billion data points per person to build aging clocks precise enough to detect organ-level divergence.
- The liver ages critically around 40, the brain around 50, and two major systemic waves of decline strike between 40–50 and again between 60–70 — a rhythm that explains why health trajectories diverge so sharply in midlife.
- Liver-derived coagulation proteins F13B, F9, and F10 were shown in cell cultures and animal models to actively accelerate vascular and tissue aging — not passive markers, but causal agents.
- The full framework has been distilled to a blood test measuring just 100–108 plasma proteins, bringing organ-specific aging assessment within reach of routine clinical practice.
Two people born the same year can age at entirely different speeds — one with the cardiovascular system of someone a decade younger, another whose liver has aged far ahead of schedule. This uneven, organ-by-organ reality has long frustrated researchers who could describe aging but struggled to explain or predict it in any individual. A consortium of Chinese institutions has now built a system called the Digital Aging Twin that attempts to do both.
Published in Cell in May 2026, the framework draws on data from 2,019 healthy adults between 18 and 91, recruited across four Chinese cities. Each participant contributed 240 measurements — clinical tests, brain and retinal scans, gait analysis, and multiple layers of molecular data — producing a dataset exceeding one billion individual data points. From this, researchers constructed three tiers of aging clocks: one integrating all physiological markers into a single measure of decline, one combining molecular layers through deep learning to predict chronological age within 3.87 years, and a third generating organ-specific clocks for the brain, liver, lungs, muscles, blood vessels, and skin.
The organ clocks revealed a striking asynchrony: the liver hits a critical aging inflection around age 40, the brain around 50, with two broad waves of systemic change appearing between 40–50 and again between 60–70. But the deeper finding came when researchers moved from correlation to causation. Through analysis of blood proteins, liver tissue, human cell cultures, and animal models, they identified liver-derived coagulation factors — particularly F13B, alongside F9 and F10 — as active drivers of vascular and systemic aging. Exposing human aortic cells to these proteins triggered cellular senescence, impaired blood vessel formation, and elevated inflammation. Injecting F13B into mice accelerated aging across the liver, heart, aorta, and kidney.
For practical use, the team found that just 100 to 108 plasma proteins — measurable through a standard blood draw — could replicate the accuracy of the full framework. Lifestyle factors also registered clearly: fruit consumption, consistent sleep, and moderate walking slowed biological aging, while smoking, sleep deprivation, and frequent eating accelerated it. The Digital Aging Twin is the first proof-of-concept for China's national X-Age Project, and while the current data is cross-sectional, longitudinal studies are underway. What has shifted, fundamentally, is the register of aging science — from description to prediction, from observation to intervention.
Two people born in the same year can age at entirely different speeds. One might have the cardiovascular system of someone ten years younger while their liver deteriorates like someone a decade older. This fundamental truth about human aging—that it is radically uneven, organ by organ—has long frustrated researchers trying to measure it. They could identify what aging looked like. They struggled to explain what caused it, or how to predict it in any individual person.
Now a consortium of Chinese researchers has built a computational system called the Digital Aging Twin that does something different: it measures not just whether someone is aging, but how fast, and which parts of their body are aging fastest. The work, published in Cell in May 2026, represents a shift from describing aging to quantifying it—and from finding correlations to identifying what actually drives the process.
The foundation is massive. Researchers from the Institute of Zoology, the China National Center for Bioinformation, Xuanwu Hospital, and seven other institutions recruited 2,019 healthy people between ages 18 and 91 from four Chinese cities. For each person, they collected 240 different measurements: standard clinical tests, cognitive and motor assessments, brain and retinal scans, gait analysis, and multiple layers of molecular data including DNA methylation patterns, RNA transcripts, proteins, metabolites, and gut microbiome composition. The resulting dataset contains more than a billion individual data points.
From this, they built three tiers of aging clocks. The first integrates all 240 physiological markers into a single measure of overall functional decline. The second, powered by deep learning algorithms that identify which data types matter most, combines all the molecular layers and predicts chronological age with a mean absolute error of just 3.87 years—better than any single-layer approach. The third tier creates organ-specific clocks for the brain, liver, lungs, muscles, blood vessels, and skin, each built from clinical markers, blood proteins, and imaging data.
What emerged was striking: organs do not age in lockstep. The liver hits a critical aging inflection point around age 40. The brain's aging accelerates around age 50. Across the entire population, two major waves of aging-related change appear—one between 40 and 50, another between 60 and 70. This asynchrony explains why two 65-year-olds can have such different health profiles.
But the real breakthrough came when researchers moved beyond correlation to causation. They analyzed blood proteins, examined liver tissue from donors, and ran experiments in human cell cultures and animal models. They found that liver-derived coagulation factors—particularly a protein called F13B, along with F9 and F10—accumulate with age and directly drive vascular and systemic aging. When human aortic endothelial cells were exposed to these factors, they showed clear signs of cellular senescence: elevated aging markers, impaired ability to form blood vessel tubes, and increased inflammation. When researchers injected F13B into mice, aging accelerated across multiple tissues including liver, heart, aorta, and kidney, accompanied by immune cell infiltration and inflammatory signals. These were not passive markers of aging. They were active drivers.
For clinical use, the team simplified the system. They found that just 100 to 108 plasma proteins—measurable in a standard blood test—could predict aging as accurately as the far more complex full framework. They also identified lifestyle factors: greater fruit consumption, consistent sleep, and moderate walking slowed biological aging, while smoking, sleep deprivation, and frequent eating accelerated it.
This work is the first proof-of-concept for China's X-Age Project, a national initiative to build comprehensive aging clocks for the Chinese population. The current framework uses cross-sectional data—snapshots in time—but researchers are refining it with longitudinal follow-up and larger, more diverse populations. What matters now is that aging science has moved from description to prediction, from identifying what correlates with aging to understanding what drives it. For the first time, researchers have a standardized system that can tell how fast a person is aging, which organs need attention, and where intervention might work.
Notable Quotes
Organs age asynchronously—the liver reaches a critical aging inflection point around age 40, whereas the brain's aging accelerates at around age 50.— Digital Aging Twin research findings
Greater fruit intake, consistent sleep routines, and moderate walking were linked to slower aging, while smoking, insufficient sleep, and high meal frequency were associated with accelerated aging.— Study analysis of lifestyle factors
The Hearth Conversation Another angle on the story
Why does it matter that organs age at different rates? Couldn't we just measure overall aging?
Because it changes where you intervene. If someone's liver is aging at 60 but their brain at 45, you'd target the liver. A single number misses that entirely.
The coagulation factors—F13B and the others—they're proteins your liver makes?
Yes. They accumulate over time, and when they circulate in the blood, they seem to trigger aging in blood vessels and other tissues. The researchers showed this isn't just correlation. They injected the protein into mice and aging accelerated.
So theoretically, if you could block those factors, you could slow aging?
That's the implication, though it's early. First you have to prove it's safe and effective in humans. But yes—moving from "this marker goes up with age" to "this marker causes aging" opens a completely different door.
The blood test using 100 proteins—how soon could that be available?
The framework exists now. The practical question is cost and accessibility. A simple blood test is much more feasible than brain imaging and genetic sequencing for everyone.
What about the lifestyle factors—fruit, sleep, walking? Are those new discoveries?
Not entirely new, but this is the first time they've been quantified within a system that also identifies molecular drivers. It's one thing to say "sleep is good." It's another to measure exactly how much it slows your biological clock.
The cross-sectional limitation—they took a snapshot of 2,000 people at different ages rather than following the same people over time?
Right. That's the next phase. They need to follow people longitudinally to see if the predictions hold up and refine the clocks further. Cross-sectional gives you the architecture. Longitudinal proves it works.