Friends of a feather

Do our genes influence who we choose as comrades?

Jan 17, 2011
Vanessa Schipani
The age-old idioms "birds of a feather flock together" and "opposites attract" might have some truth in them -- along with laughter and secrets, friends may share genetic similarities, as well as some differences, according to a new linkurl:study; published in the January 17th issue of Proceedings of the National Academy of Sciences (PNAS).
Wikimedia Commons, Kevin Jaako
"This paper is an invitation for people to think about the genetics of human behavior," an area of research that has long eluded scientists, said linkurl:Ting Wu,; a geneticist at Harvard University, who was not involved in the study."Our ability to make friends and keep friends is a part of humans that makes us unique," added linkurl:James Fowler,; professor of medical genetics at the University of California, San Diego and first author on the paper, yet "the biology of social networks is relatively unstudied."Drawing on data collected from two long-term health studies, Fowler and his team found that social networks may form from a genetic predisposition to surround ourselves with people who are like us in some ways and unlike us in others. Specifically, the researchers found that friends tend to share the DRD2 gene, which is associated with a propensity for alcoholism. Because drinkers are obviously more likely to surround themselves with people who are "slightly more likely to have a well-stocked liquor cabinet or meet you in a bar," the finding is not all that surprising in and of itself, Fowler said. But the fact that the genetic landscape of human populations may be affected by these friendships is a new and interesting concept, he added. As a result, studies that make associations between an individual's genes and behaviors could be inherently biased if they don't take into account the genotype of their subjects' social groups. For example, if a person with the DRD2 gene hangs around people who also have DRD2 gene, that individual may be more likely to drink alcohol than if he or she associated with people who lack the gene. Thus, a study linking alcoholism and DRD2 may be skewed if it doesn't take into account the genotypes of the subjects' friends. The group also found that people with the gene CYP2A6, which may be tied to the trait for open-mindedness, tended to make friends who lacked the gene. While it doesn't make sense that open-minded people would befriend close-minded individuals, and vice versa, said Fowler, such a negative correlation could simply reflect an unconscious choice to associate with people who are genetically different from oneself. When searching for mates, for example, people are known to select for individuals with different immune systems to increase the health and survival of one's offspring. The same idea might also apply to making friends, Fowler said. "It makes sense to be surrounded by people that get different diseases," he said, because it may limit transmission. But it's just speculation at this point, he added. Though "there's something about Mary that draws you to her," whether for reproduction or friendship, he said, "researchers haven't figured out what that something is."The group controlled for population stratification, or the likelihood that people of similar ethnicity or ancestry congregate in the same geographic area, but sociobiologist linkurl:Dalton Conley; of New York University said he would have liked to see controls for alcoholism as well. Being a drinker is an obvious trait that can be identified by a potential friend relatively quickly, said Conley, so it's possible that it's not the DRD2 gene itself, but the behavior that DRD2-positive folks are selecting. Without controlling for alcoholism, it would be impossible to know, he said. It is the complexity of these relationships that "makes studying the genetics of human behavior so difficult," said Conley. While studies such as these might reveal tantalizing correlations, there are many subtle factors influencing the patterns that might be overlooked. J.H. Fowler et al., "Correlated genotypes in friendship networks," PNAS, AOP, doi:10.1073/pnas.1011687108, 2011.
**__Related stories:__***linkurl:Bee social activity tied to gene;
[6th March 2007] *linkurl:Battling Bad Behavior;
[1st February 2006]