Birds of a (Genetic) Feather Flock Together: How Genes Determine Human Friendship

That picture on the right is one of my favorite quotes about friendship, but a new study (and it's open access) points out that friendship might not actually be so random.  Friendship is a fundamental characteristic of being human, and genes are known to play a role in the formation and structure of friendship ties.  As it turns out, genes also play a role in deciding who we select as friends.  We share about 1% of our genes with our friends, and we don't find them at random.  Since we can't see their genes, we use our sense of smell to find them!

Human beings evolved as social animals; social groups provide more protection against predators and infections, provide information and security, and they provide the opportunity for cooperative exchanges.  Genes that help humans form friendships are obviously a huge evolutionary advantage. As humans, we also tend to like people who are like us - this is called homophily.  So how do you study the genetic similarity between friends?  The authors took data from 1 932 people who took part in the Framingham Heart Study, which included genetic information (including common genetic variations, or SNPs) as well as friendship ties.  The researchers underwent a genome-wide association study (GWAS) and identified over 460 000 SNPs, looking not at any particular SNPs, but at patterns across the whole genome. 

Genetic similarity is higher between friends (left), and
genetic dissimilarity is higher between strangers (right).

There were 1 367 friendship pairs (907 of which were not biologically related at all, even distantly) and almost 1.2 million stranger pairs (from which 907 non-related pairs were randomly selected) in the dataset.  By comparing genetic markers and looking at patterns between friendship and stranger pairs, the authors were able to show that friend pairs shared significantly more SNPs than stranger pairs, who shared about as many as were expected.  Which genes do we tend to share more among friends?  Specifically those involved in the olfactory transduction pathway, or our sense of smell.  On the other hand, genes involved in immune system processes are significantly different among friends, which supports the idea that complimentarity in social group immune systems has a protective effect against infection.  Ultimately, pairs of friends are as genetically similar to each other as fourth cousins.  This study provides a genetic basis for the evolution of friendship through the process of kin recognition (a process in which the sense of smell is very much involved in other animals as well).  You can imagine how forming friendships with other individuals who perceive and/or cope with the environment in a similar way to us can result in both individuals benefiting from the other's response.


These findings are still very preliminary.  The use of the Framingham Heart Study participants introduced a bias, as these people were predominantly white and of Italian descent.  Here, it is possible that genetic similarity and friendship may come from the simple preference for associating with ethnically similar others, especially since the first cohort was recruited in 1948.  The authors did control for population stratification in their data analysis, however, so this bias should be minimal.