How long will we live, and how much of our lifespan is really written in our genes? For decades, the scientific consensus held that genetics played only a relatively minor role in determining human life expectancy, about 20% to 25%, and in some large-scale studies from recent years, even less than 10%.
A new study by researchers at the Weizmann Institute of Science, published Thursday in the journal Science, fundamentally challenges that assumption. According to its findings, genetics has a far greater influence on life expectancy than previously thought, accounting for about 50% of the variation, at least twice the estimates long accepted by science.
The study was conducted in the laboratory of Prof. Uri Alon, head of the Sagol Center for Longevity Research at the Weizmann Institute, and was led by doctoral student Ben Shenhav.
“Life expectancy is the average length of a person’s life,” Shenhav explains. “It has risen dramatically over the past 150 years, by more than 30 years. In the past it was under 50, and today it is over 80 in developed countries like Israel.” The main reason for this dramatic change, he stresses, has little to do with aging itself and much more to do with external factors. “First and foremost, infant and child mortality dropped sharply. Added to that were improvements in sanitation and declines in accidents, violence and especially infectious diseases,” he says. “We are talking about a world without antibiotics. If someone got an infection, there was a good chance they would not survive. Today you go to the doctor and get antibiotics. Until not that long ago, this simply was not the case. In historical terms, this is very new.”
5 View gallery
Prof. Uri Alon and doctoral student Ben Shenhav in research that is reshaping how scientists understand life expectancy
(Photo: Weizmann Institute of Science)
Behind these numbers lies a question that has accompanied science for decades: What matters more for how long we live, genetics or the environment in which we grow up and live? To try to answer it, researchers have long relied on one of the most classic and established methods in genetic research, twin studies. “This is the classic way in science to understand how important genetics is for anything,” Shenhav says. “Identical twins share all of their genetics, so they are essentially copy and paste of one another. Non-identical twins are like siblings. They share genetics that are similar but also differ by about 50%.”
Comparing the two groups allows researchers to isolate the genetic contribution. “If you look at the similarity between identical twins and non-identical twins and combine those two things, you can isolate the effect of genetics from the effect of a shared environment, since both identical and non-identical twins grow up in the same household,” he explains. “For the past 50 to 60 years, twin studies have been used to measure the genetic influence on traits like height, intelligence and more.”
But when scientists tried to apply the same approach to life expectancy, they ran into a fundamental limitation. “The first studies on longevity using twins were published about 30 years ago,” Shenhav says. “To do this kind of research, you need to look at a population that has already died. That means you have to go back at least 100 years. As a result, the first studies focused on twins born at the end of the 19th century.” When the data were analyzed, the picture seemed clear. “Genetics was found to account for about 20% to 25%.”
'It is intuitively clear that external mortality distorted earlier estimates. Classical statistical analysis can lead to the mistaken conclusion that genetics is not important. Our world today is safer, and people die of old age. This was not a mistake by the researchers, who did their work properly, but there was a variable that was not taken into account'
In the past decade, researchers tried to refine these estimates using even larger datasets. “New studies came out that analyzed data not only on twins but on family trees, millions of people over hundreds of years,” Shenhav says. “The idea was to put as much data as possible into the analysis and, based on similarities in age at death among family members, get a more precise number.” Here too, the results were surprising. “These studies attributed an even lower influence to genetics, about 10%.”
In hindsight, Shenhav says, it is possible to understand why. “It is intuitively clear that external mortality distorted the earlier estimates,” he explains. In the old datasets, extreme gaps could be found between twins. “You might see one twin who lived to 100 and another who died at 30, and you ask what is going on. In such cases, classical statistical analysis might lead you to the mistaken conclusion that genetics is not important.” Today’s world, he notes, is much safer. “People die of old age. This was not a mistake by the researchers, they did their work properly, but there was a variable that was not taken into account. The biggest problem is that for those twins, and also in family tree studies, there was no information about cause of death.”
Virtual twins instead of missing data
To overcome this problem, the researchers adopted an unconventional approach for aging research. Instead of relying only on incomplete historical data, they turned to tools more familiar from physics and mathematics. “We are mainly physicists and mathematicians in the lab, not classical biologists, so the tools we use to approach scientific questions are different,” Shenhav says. “What physicists like and know how to do is build mathematical models that describe the laws behind quantitative phenomena.”
In Prof. Alon’s lab, which has been studying aging for nearly a decade, the researchers developed exactly such a model. “We already had a model we used in the past that successfully captures the dynamics of human aging,” Shenhav explains. “It is a mathematical equation that describes the balance between the accumulation and removal of damage in the body. When the damage crosses a certain threshold, that is the point at which a person dies. Using this model, I can create a kind of simulation of different people.”
5 View gallery
Your genetics account for about 50% of your expected life expectancy
(Photo: Shutterstock)
This led to the study’s central question: Could the model be used to isolate the contribution of genetics even without knowing the actual cause of death? Here the key idea emerged. “We realized that we could try, using the model, to create a simulation of twins,” Shenhav says. In other words, in the simulated world created by the researchers, each person has different biological parameters. In the case of identical twins, those parameters are identical as well.
Unlike the real world, however, the simulation allowed the researchers to control every variable. They could therefore remove external mortality in order to isolate the genetic factor. “As the person controlling the simulation, I can turn off external mortality and say, no more infections or violence,” Shenhav says. When they did this, the result was unequivocal. The genetic influence jumped sharply. “We saw that genetics accounted for about 50%. At first we were surprised, but this gave us a prediction we could test to see if it was indeed correct.”
Beyond simulations, the researchers also sought confirmation in the real world. “We had a very clear prediction that if you reduce external mortality, you should see the genetic influence increase,” Shenhav says. As part of the study, they used a unique dataset made possible through collaboration with researchers in Sweden. “The dataset included twins born in later years, between 1920 and 1930, meaning in a world where external mortality was already lower,” he explains. “We were able to see in the data that our prediction was indeed realized. Heritability increases as external mortality decreases.”
'The conclusion is not that lifestyle and exercise are unimportant, absolutely not. Even if genetics accounts for 50%, that leaves another 50% that is not genetic, which is substantial. Lifestyle, nutrition and physical activity likely become more important as people age'
Still, Shenhav stresses that the findings do not negate the importance of lifestyle. “First of all, the conclusion is not that lifestyle and exercise are unimportant, absolutely not,” he says. “Even if genetics accounts for 50%, that leaves another 50% that is not genetic, which is huge.” That remaining half, he explains, includes all the familiar factors. “Lifestyle, nutrition, physical activity and many other things,” and their importance may even increase with age. “They likely become more important as you get older.”
Beyond personal implications, the study also has broad significance for future research on aging. For years, attempts to study the genetic basis of longevity were met with skepticism. “Previous studies cast doubt on the importance of genetics,” Shenhav says. In such a reality, a researcher seeking to focus on the genetics of longevity could encounter resistance.
One of the main achievements of the new study, he says, is that it “lays new and solid groundwork for follow-up research” and now makes it possible to move forward with dedicated genetic studies of lifespan. The motivation, he adds, also comes from everyday observation. “We all know people from daily life who can reach 100 without illnesses or health problems, even if they smoked and ate poorly. It is clear they have some genes that protect them.”
From this, he argues, follows the importance of deepening the research. “It is worth investing money and effort in research to understand what those genes are that protect people who reach 100 in good health. There have been studies in the past, but we need more of them,” he says, adding that over time the data pool will only grow. “There are secrets in genetics that can hopefully one day help us develop drugs to fight aging,” Shenhav says. If and when those mechanisms are understood, he concludes, it may be possible to translate them into practical medicine.