Although people in the US are increasingly choosing similar mates, this does not seem to be driving changes in the American gene pool.
The finding comes from a team of researchers who tracked traits with genetic cues, such as height.
“We’re asking how spouses are alike, how this is affecting the number of children they have, and then asking how both of these are changing over time,” says Ben Domingue, an assistant professor in the Graduate School of Education at Stanford University and one of the authors of the study. “We see an increasing stratification across society in terms of mating and fertility, but it’s not corresponding to changes in the underlying genetic signature.”
Consider, for example, a scenario in which tall people have a propensity to marry each other and have many children, while short people likewise married but had fewer children. If we assume that the height of these people reflects the influence of their DNA, one might expect to see more genes for tallness in the population over time.
“While there is a tendency for people who are genetically similar along key dimensions to marry each other, there does not seem to be any increase in this tendency, despite what some people may fear about inequality getting baked into our genes,” says first author Dalton Conley, a professor of sociology, medicine, and public policy at New York University. “Ditto for how many kids folks are having—while there are genetic associations, they appear to be stable over time.”
Four traits in the DNA
To test this general hypothesis, the researchers analyzed genetic data involving 4,686 non-Hispanic white adults and their spouses (the information comes from a longitudinal dataset, and is considered significant such that similar conclusions can be inferred across other groups in the United States). For each participant, they tested for genetic patterns that have been associated with four different traits: educational attainment, height, body mass index, and depression.
Genetic aspects of the four traits were defined based on measured DNA. In the twisted ladder structure of DNA, base pairs form the rungs, about 3 billion in total. The vast majority of these pairs are the same across all people, but roughly 1 in every 300 base pairs is a point at which humans differ. The research team used 1.5 million of these spots of variation, along with previous research linking them to the relevant traits, to examine patterns in mating and fertility.
The researchers found that spousal genetics influencing education and height are associated. They also saw that pairs with more education and those with taller statures tended to produce fewer children. However, while education was an increasingly strong predictor of the number of offspring and the overall education of that entire family in more recent birth cohorts, these characteristics were not associated with changes in how the genetics of education were associated with these traits.
“If you just look at how people select for similar spouses, you might think there are changes in how genetics are related to decisions regarding marriage and fertility,” Domingue says.
“Our point with this study is that none of the trends in the observed traits—height, for example—seem to be associated with changes in the relevant genetics. If we just had the phenotypic information and we tried to use that to infer what was happening genetically, we may get the wrong answer.”
Their findings show that assumptions about DNA are not always correct. Despite a large amount of attention to genetics, scientists are still growing the understanding of how the human gene pool interacts and changes over time.
“As we get better and better measures of genotypes with bigger and bigger subject pools, it will be important to reproduce this study to keep tabs on how social forces—like rising income inequality or changing marriage norms—may be affecting us at the population genetics level,” Conley says.
This paper was a collaboration among researchers at New York University, Duke University School of Medicine, the University of Wisconsin, the University of Colorado Boulder, and Stanford. Their study appears in the Proceedings of the National Academy of Sciences.