Friendships down to your genes

Shared interests and hobbies might bring people together, but new research has shown that we may be picking friends with similar genes to our own.

By comparing the genes of friends and strangers, James Fowler, from the University of California San Diego, has found that we share more than just our interests with our close allies, as he explained to Chris Smith...

James - Well we looked all across, the human genome at different markers and found that friends tend to share genes in common. In fact, the level of similarity between friends is so strong,  it's as if that friend who we don't share common ancestor with is our fourth cousin. Even more interesting than that, one of the things that we found was that those genes that we share in common with our friends are also the ones have been evolving the fastest. So as a result, these things together suggest that friendship has been turbo charging human evolution.

Chris - Wow! So why should people logically if they're friends with each other have more genes in common than you would expect by chance. Why should that happen?

James - Well, we know a wide variety of traits are similar between friends. In this case, we find that friends have some more genes is because they chose each other or they were drawn to similar environments where they met other people who were like them. So just for example, imagine that you have genes that make you really, really love coffee. Like a zombie, you're just drawn to the coffee house and you might find other people who are similarly drawn to the coffee house and that's where you tend to make friends.

Chris - And how did you actually discover this?

James - We took information from the Framingham Heart Study on about 2,000 people and looked at their genes and also looked at their friendships. We compared how similar how the genes were between friends to how similar the genes were between strangers drawn from the same group.

Chris - The thing is that if you look at who we tend to marry, have sex with and have children with, we're always being told that the human race is evolving to have the biggest genetic diversity. We go for the biggest differences. So, if you were to do your study on people who were choosing mates rather than choosing friends, would you see a different relationship?

James - So, it's interesting you should ask that question. Some colleagues of ours just recently published a paper that showed that spouses also tend to be genetically similar. But that hides some parts of the genome where we tend to be different.  nd so, one long standing result, when you look at spouses is that they tend to have opposite genotypes for the immune system. And the reason there is, you don't want your spouse to be susceptible to something that you're also susceptible to. You want them to be an extra line of defence against disease. We found actually the same thing in our friendship study. Your friends also tend to have genotypes that are opposite to you for the immune system. So, it's really interesting that we get these similar results both for spouses and for friends.

Chris - So, what are the implications of this then that literally, you could take my genes and you could make predictions from a bunch of people in my environment and say who I'm likely to get on with or not? Could you do that?

James - Yes, we could. This is one way to validate what we found. We take our model where we look at all the genes that were similar and all the genes that were different and with our friends. And we can use that then just to take the genes of people for whom we don't know whether or not they're friends and make a prediction. We can create what's called a friendship score. In going from a low value of this friendship score to a very high value on this friendship score actually increased by 24%, the probability that 2 people were friends. I was really surprised actually that it was this strong because we know that social environment is also really important for determining who's friends with whom. But here, it looks like genes are also playing an important role.

Chris - So, why do you think that evolution is opted to make us be drawn to people who have a similar genetic makeup? Do you think it is just as simple as, "I have a chemical preponderance to like coffee and so do you therefore, we've become friends"? Or do you think actually there are more subtle mechanisms at work here that make people who are genetically similar be drawn to each other?

James - I think that at least some of this is active choice and the reason why I think that is because another thing that we found in the study was that those genes that we share in common with our friends, those genes are actually evolving the fastest which suggested those are also the genes that are giving us a survival advantage. So, think about language for example. If you're the first person to get a mutation to speak language, is that going to do you any good whatsoever unless you are connected socially to other people who have that same mutation. This is a genetic variant that only gives you an advantage if someone else has it. And that's exactly what we'd expect to find across the whole genome if it was a case that we had this relationship between how much you share genes in common with friends and how quickly those genes are evolving. In other words, one of the things that it seems like we found is that this is a way that friendship might actually be increasing the speed of human evolution.