A new study that analyzed more than 400 million people has revealed that genetics has a smaller impact on how long a person will live than scientists had previously thought. Researchers from Calico Life Sciences in collaboration with Ancestry (an online genealogy company), have established that inherited life span is much below what was previously estimated as earlier beliefs did not take into account people’s tendency to select partners who had similar traits to their own.
The goal of the study was to assess the heritability of lifespan…whether a person’s parents who lived long lives could predict whether that person lived a long life. Heritability measures to what extent specific genes can explain differences in a person’s traits. In the case of this study, life span. This is different than non genetic differences such as sociocultural factors, lifestyle and accidents. Previous human life span heritability estimates ranged from about 15 to 30 percent.
The researchers looked at a carefully chosen set of family trees and relevant information from over 400 million people who were surveyed by Ancestry. They began with 54 million subscribers to Ancestry which represented six billion ancestors. They then removed entries that were redundant and people who were still living thereby stitching all remaining pedigrees together.
Ancestry then stripped away any identifiable information from pedigrees leaving just the birth year, death year, birth place and familial connections that are part of the family tree structure. Most of the people were based in the U.S. and of European descent and connected to another by either spouse/spouse or parent/child relationship. The researchers then examined the similarity of life span between relatives so they could estimate heritability from each family tree.
They combined statistical and mathematical modeling to analyze data of relatives who were born during the 19th and early 20th centuries. They observed that first cousins and siblings showed same heritability estimates that were observed in previous studies. They also noted the life spans of spouses were likely to be correlated and were actually more similar than in siblings of opposite gender.
The correlation between spouses could be attributed to a variety of non genetic factors such as living in the same household or shared environment. The results then really started to surprise the researchers when they compared different types of first cousins in laws and siblings in law though there was no blood relatives and typically did not share households.
The researchers were able to focus in on correlations for other even more remote relationships which included uncles in laws and aunts, first cousins that are once removed in law and also a variety of configurations of co siblings in law. The findings which indicated that a person’s siblings, spouse siblings or their spouses siblings spouse had similar life spans, made it obvious that something besides a person’s genes was in place.
If people did not share genetic backgrounds and also did not share households, then the question became what best accounts for the life span similarity between individuals that had these types of relationships. The team went back to the data set and proceeded to perform analyses that would detect assortative mating which means the factors which are important to longevity tend to be very similar between mates. People tend to choose partners that have traits similar to their own and in this case how long they live.
Because people marry obviously before one is deceased, then assortative mating must be based on other characteristics. The mate choice could be sociocultural or genetic or both. In regards to non-genetic characteristics, wealthy people tend to marry other wealthy people for example. Or related to genetics, tall people might prefer to marry another tall person and height is somehow correlated to how long a person will live which would also inflate estimates of heritability life span.
The new analysis found that by correcting for the effects of assortative mating, life span heritability is likely no more than seven percent and perhaps even lower. The findings in this study certainly point to how low heritability of lifespan is. There are many things to learn about the biology of aging from human genes, however the recent findings temper expectations about what types of things can be learned and how easy it will be.
To view the original scientific study click here: Estimates of the Heritability of Human Longevity Are Substantially Inflated due to Assortative Mating