
How Does Sleep Change Across Your Lifespan?
18 min read
ReadA new study rewrites the heritability of lifespan — and what it means for the 45% you can still control

A 2026 study in Science found genetics accounts for ~55% of intrinsic aging variation
Previous estimates of 20-25% were skewed by including extrinsic causes of death
The Calico 2018 study placed heritability below 10% due to assortative mating effects
The remaining 45% is modifiable through sleep, exercise, nutrition, and stress management
Precision tracking matters more when the controllable window narrows to 45%
For years, the longevity community has been working from a tidy assumption: your genes control somewhere between 20 and 25 percent of how long you live. The rest — 75 to 80 percent — is lifestyle, environment, and luck. That figure became the foundation for an entire industry built around the idea that if you just sleep better, eat cleaner, move more, and manage stress, you can effectively outrun your DNA.
Then, in 2018, Calico Life Sciences — the Alphabet-backed longevity company — published a large-scale genealogical study that put genetic heritability of lifespan at below 10 percent, suggesting assortative mating (the tendency to choose partners with similar lifestyles and socioeconomic backgrounds) had inflated even the already-modest estimates.
It seemed settled. Genetics was a footnote. Lifestyle was the story.
But a study published on January 29, 2026, in Science — one of the most rigorous peer-reviewed journals in the world — has reopened the question entirely.
The research, led by senior author Uri Alon, a systems biologist at the Sagol Institute for Longevity Research at the Weizmann Institute of Science in Israel, approached lifespan heritability from a different angle.
The core insight is deceptively simple: previous heritability studies had been measuring the wrong thing.
When earlier researchers estimated that genetics controlled only 20–25% of lifespan variation, they were including every cause of death in their analysis — including deaths from accidents, acute infections, violence, and other extrinsic causes that have nothing to do with the biological machinery of aging. A 30-year-old who dies in a car accident tells us nothing about their longevity genes, but in a traditional heritability study, that data point pulls the genetic estimate downward.
Alon and his team developed mathematical models of aging that separated intrinsic mortality (death driven by the body's own biological aging processes) from extrinsic mortality (death caused by external factors). They validated these models against twin and sibling data from Scandinavian population registries and US datasets.
When you isolate intrinsic aging — the biological process of growing older — genetics accounts for approximately 55% of the variation in lifespan. That is more than double the traditional estimate.
The explanation is elegant. Imagine you are studying two identical twins. They share 100% of their DNA. If one twin dies at 40 in a skiing accident and the other lives to 90, a naive heritability analysis would record a 50-year difference in lifespan between two people with identical genes — dragging the heritability estimate toward zero.
Previous studies did not adequately account for this problem. By lumping extrinsic deaths into the analysis, they were essentially measuring the heritability of "surviving everything life throws at you" rather than the heritability of biological aging itself.
The Calico study's finding of below 10% heritability was particularly influenced by assortative mating effects, but as Graham Ruby of Calico noted in response to the new Weizmann findings, controlling for extrinsic factors "changes the trait from overall lifespan to lifespan under specific conditions." This is a fair methodological point — the 55% figure applies specifically to intrinsic aging, not to lifespan as measured in the messy real world.
But that is precisely the point. If we are interested in understanding and intervening in the biology of aging — which is what the longevity field is actually trying to do — then intrinsic aging is the relevant trait.
In an accompanying commentary published in the same issue of Science, Daniela Bakula and Morten Scheibye-Knudsen from the University of Copenhagen wrote that these findings justify "large-scale efforts to identify longevity-associated variants." In other words, the case for investing in genomic longevity research just got significantly stronger.
This does not mean previous lifestyle research was wrong. It means the picture is more nuanced than the "genetics barely matters" narrative suggested. The research actually shows that both genetics and lifestyle are substantial contributors to how we age — and neither should be dismissed.
Here is where the skeptical voice lands: even if genetics controls 55% of intrinsic aging, that leaves 45% that is modifiable through lifestyle and environmental factors.
Forty-five percent of your aging trajectory is determined by:
Forty-five percent is not a small number. It is an enormous lever. The difference between someone who actively manages these four pillars and someone who neglects them can be measured in years — potentially decades — of healthspan.
If the modifiable portion of aging is 45% rather than 80%, the implication is not that lifestyle doesn't matter. The implication is that precision matters more.
When you had 80% to work with, you could afford to be imprecise. Broad strokes — eat better, move more — might have been enough. But when the controllable window narrows to 45%, every percentage point counts. You need to know where you stand. You need to see your trends. You need data.
This is where wellness scores come in.
Your wellness scores in LongevLab track the 45% you CAN control. They aggregate signals from your daily habits — sleep, activity, recovery — into a clear picture of how you are performing on the modifiable side of your health equation. They are not a medical diagnosis. They are a directional tool that helps you see patterns, identify weak spots, and course-correct before small issues become large ones.
No subscription required. Your wellness scores are part of the core LongevLab experience.
The Weizmann Institute study does not tell you that your genes are your destiny. It tells you that genetics plays a larger role than we previously thought — and that the 45% you do control deserves more focus, more precision, and more tracking than ever before.
The research actually shows that both sides of the equation matter. Your genes set the playing field. Your daily choices determine how you play the game.