Friends Are Genetically Similar (2014)
nature.com
nature.com
edit: corrected grammer
If that is the case, it would lead to the politically uncomfortable question of different genders or ethnic groups being more likely to have particular interests.
What is a problem is if you use that to justify discrimination. Just because members of a demographic are statistically different, it isn’t okay to treat individuals differently because they belong to a certain demographic. And that’s what people are complaining about!
People don’t complain when you say: “A lot of the fastest marathon runners are from Kenia”. People will complain if you use that fact to justify an imbalance in your workforce: “Well, it’s obvious that we hire only people from Kenia since marathons show they are much faster!”
So why do companies fire people who say stuff like that? Because discrimination is stupid from a business point of view: variance within demographics is much larger than variance of the mean between demographics. So if you hire people based on them being part of a “good” demographic, you will end up with worse hires than if you actually selected for individual aptitude.
People don’t complain when you say: “A lot of the fastest marathon runners are from Kenia”. People will complain if you use that fact to justify an imbalance in your workforce: “Well, it’s obvious that we hire only people from Kenia since marathons show they are much faster!”
This is precisely the point. If objectively they are better, why is it discrimination to only hire them? Definitely you give everyone the opportunity to work for you, but if a singular group always performs better, is it really your fault? And if society did determine that it was your fault, how long could that go on for? Forever? How long can that charade last? Hasn't a Darwinian economy (capitalism) done the best ever for man kind (massively raised standards of living over the last 200 years).
TL;DR - Don't hire only Kenyans, that's discrimination. You might end up with a lot of Kenyans if you hire only marathoners, but that's not the same.
Of course, a lot of the discussion is about faulty tests, that claim to test for individual aptitude, but end up testing for demographics instead (eg. interviewers inadvertently giving higher scores to people similar to themselves)
Discrimination is suspected if the test has little to do with actual job requirements.
Lack of diversity in tech is sad for me, but I don't think that there should be any additional incentives or bonus points on entry based on minority status. The problem for me is that large companies pretty much shy away from responsibility of training new employees for themselves and only hire people who got themselves fully educated already. This causes artificial barriers in entry to tech that let in only the people who in the past had the means and the will.
I'd put responsibility on the companies to hire for entry level positions people that have not previously worked in tech and attempt to train them. This could be proportional to company headcount and if company refuses to hire it should contibute significant amount of money to a fund that sponsors free tech education initiatives for anyone who's willing.
I think this approach would increase diversity and even help other groups disadvantaged on the job market.
It would put burden on companies but I haven't heard of company bankrupt because it spent too much on new employee training.
This kind of scheme is in place in my country to support employment of the disabled people. I think it should work on entry level employees too.
How is this even possible? People have to work full time jobs while interviewing.
> variance within demographics is much larger than variance of the mean between demographics. So if you hire people based on them being part of a “good” demographic, you will end up with worse hires than if you actually selected for individual aptitude.
did you not read the comment fully? that last part seems to pretty clearly address what you're asking here.
The uncomfortable questions are whether people are genetically predisposed to, or naturally have a greater aptitude for, some particular interest, particularly on racial or gender lines. For many physical activities this is a settled question for gender, but that's about all we know.
Even this alone is not necessarily controversial, the problem as we've seen time and again is that the reaults of such a study are taken as a basis for discrimination. If we could rigorously quantify and prove a statement like "people of East Asian descent show a 2.3% greater aptitude for mathematics," it doesn't mean that the Chinese guy sitting in front of you has any such aptitude. Historically the metrics by which we have attempted to measure such things have been deeply flawed and biased. I'm not sure it will ever be possible to make such assertions in any rigorous manner, the factors behind cognitive tasks are just too complex.
The answer is obvious here: yes. The only question is the size of the effect.
Let's say the statistical size of the effect is something around a millionth of a millionth of a percent, then I think most people would answer the question with "no" and be understood correctly by everyone. These kinds of questions almost always have a "big enough to have a perceptible impact"-clause implied. Pretending they don't feels needlessly pedantic.
There are going to be definite real world measurable differences.
OP leaves open the possibility, for instance, of cultural differences that have a effect an order or two of magnitude larger though.
Personally I doubt this, but OP seems open to the idea the size might be small.
I'm not saying they are wrong, I just disagree with the tone.
In small groups that are already interested in something, finding genetic similarity would mean that their nature is similar, and that nurture has also had an opportunity to take effect.
So small groups of people interested in something, can't necessarily be extrapolated to a larger group.
I'm sorry, but genes clearly determine appearance.
Genes, along with other factors, contribute to appearance, but it is possible to be (for a pair of humans) relatively close in appearance and generically dissimilar, or vice versa.
Genetic similarity and appearance similarity aren't the same thing.
If your talking about the often vaunted genetic diversity on the continent of Africa, that has no bearing on the existence of clusters of relative genetic similarity around the world we have come to refer to variously as ethnicity, race, population group, etc. In fact that African genetic diversity supports the reality of distinct population groups and ethnicities because it is a bunch of genetic material that other groups don't have.
The very basis for ethnicity/race/population groups is a genomic "alphabet"/string within certain bounds. In considering the global level distinctions between populations it makes no difference the level of genetic diversity in any particular group.
The defining factor is that various groups share traits that no other group does.
Asians and Europeans have stretches of DNA that we've determined through the magic of modern science come from an extinct species of human called Neanderthals. None of these stretches of DNA are found in Africans.
Likewise, Africans have stretches of DNA that come from as yet identified extinct species, these stretches of DNA aren't found in Asians or Europeans.
Then of course there are the myriad other genetic structures that while having origins in the ancestors of all living humans have nonetheless diverged through the process of evolution over the millennia into reliable genetic markers for the various population groups.
And it is true that ethnic groups are defined as well by visual appearance, for example cranium shape and size, nostril diameter, length of nasal bridge, hair color/texture etc. but these things aren't the result of members of your culture gathering around at your birth and molding these things like clay, these things are encoded in genes and what molded their shape into your genes are evolutionary pressures in your ancestral environment.
Also, for many human traits (i.e height), the intra-population variation is much greater than the inter-population variance, especially when controlling for the effect of environment.
We have a tendency to anchor our biases on the extreme examples of any population when we form our biases, often through mediums like elite/pro sports which select those at the extremes.
How did they control it? Even amongst Europeans, I imagine French people are just more likely to have French friends.
I might counter-hypothesize that the 1% they found is just the amount of variance in ethnic group that wasn't controlled for perfectly.
But within any major metro area you're right, there's typically little meaningful distinction anyone would pick up on between two Americans of different European ancestry.
It's kinda neat, actually. I am also a citizen of Canada, but don't speak French. It's fun to learn and be exposed to the culture, but I digress.
We don't know. At least I don't. But it would hugely surprise me if we are not constantly picking up on all sorts of subtle gene expressions. And this needs not in any way be a conscious process. Sophisticated ability to suss out degrees of relatedness would be a far too useful feature not to have been selected for again and again and again.
1. Surely very high intermarriage rates between different European ethnicities complicate that.
2. In that case, wouldn't the opposite behavior be expected? If the goal were to have as little genetic variation as possible humans would be better off reproducing asexually.
Very high intermarriage rates? Wherever did you pick up such an idea?
> In that case, wouldn't the opposite behavior be expected? If the goal were to have as little genetic variation as possible humans would be better off reproducing asexually
There is no 'goal'. And too little variation seems to be a universally bad idea. It is a simple, observablke fact that most of us stick to partners reminiscent of ourselves. We may argue about the reasons, but the thing itself is clear and obvious.
To my second point, the whole benefit of sexual reproduction is genetic diversity, so it's odd to have some evolutionarily dictated behavior pulling the opposite way.
Statistics to the contrary are out-of-date or highly-misleading/inaccurate (counting Brazilians as "non-Hispanic white" and/or, for obvious reasons, undercounting illegal and non-English-speaking immigrants).
Further the non-Hispanic white population is itself ethnically diverse (Italian, Portuguese, Irish etc) with "English/Anglo-Saxons" a small minority.
This study is in the US. By "people of european descent", they are referring to white americans rather than actual europeans.
Also french is a linguist/national/ethnic classification rather than a genetic one.
So if the European population they used consisted of half English and half French participants, then the first PC would likely divide the two. So if British or French ethnicity affected the analysis, this effect would be reduced or eliminated using just the first PC.
How many PCs should be used? Well, ten is a common number chosen. Is it enough? Probably for most studies including this one.
So ancestry and ethnicity in this type of research emerges purely from a PC analysis on the genotypes. They do not use self-reported ethnicity or any other type of common ethnic grouping. To control for population structure, self-reported ethnicity is rarely useful. Its far too unreliable. But these groupings do sometimes loosely correspond to the PCs found in the analysis, supporting the use of PC as a method to control for ethnicity.
(original article: http://www.pnas.org/content/111/Supplement_3/10796.full )
To me there are a few things worth highlighting, that are at least tangentially relevant:
1. This article was a PNAS Direct Submission. PNAS Direct Submission, a privilege typically reserved for members of the National Academy of Sciences, is like using trump-card to basically skip peer review. Since NAS members are usually great scientists with a track-record of publishing in high-impact peer reviewed journals, PNAS Direct Submissions are usually saved for pet-projects where the underlying science is... mmmm... of dubious quality. This is not always the case, but when all-star scientists opt for PNAS DS, it's certainly a reason to be speculative.
2. Nick Christakis and James Fowler have teamed up for a number of fancy-ass studies (the type that ensure people will want to write wikipedia pages about them, and lead to TED talk invites); some suggesting personal attributes like obesity, smoking, and happiness arise via 'social contagion', that is, behavioral propensities are transmitted over social networks. And studies like this one, where they find friendship affinities embedded in the genome.
3. I recently read a genome-wide association study (GWAS) with over 400 authors (https://www.ncbi.nlm.nih.gov/pubmed/28714976). I think 400 authors is way above median, but even relatively tame GWAS typically require a group effort just for data QC. Here, somehow these two guys did it all themselves.
4. Not only did they find that some SNPs were positively correlated, they also found some that were negatively correlated... "Across the whole genome, friends’ genotypes at the single nucleotide polymorphism level tend to be positively correlated (homophilic). In fact, the increase in similarity relative to strangers is at the level of fourth cousins. However, certain genotypes are also negatively correlated (heterophilic) in friends. And the degree of correlation in genotypes can be used to create a “friendship score” that predicts the existence of friendship ties in a hold-out sample." So the HN/Nature title might as well be: "Friends are Genetically similar, and Genetically Different".
2. Yes, these scientists are looking to make their research popular, highlighting an unfortunate conflict of interest present in academia. They do this to get big grants. Important observation, but insufficient to dismiss a finding out of hand.
3. So many authors, such a joke! These people mostly just help collect data, or sometimes simply share data they already have and then give the manuscript a quick read before publishing. But that is actually beside the point. If a lab already had data ready, and isn't pressured to add authors for political reasons, then two authors is not necessarily a red-flag. In fact, these analyses are relatively easy to do with available software. If they have the data, they just plug and chug.
4. The findings of the paper suggest strongly that friends are more genetically similar, on average, akin to fourth cousins. The results have yet to be rebutted to my knowledge. The fact they found some regions were negatively correlated does not conflict with their overall results. I have some doubts that their regional analyses should carry much weight (e.g. olfactory and immune regions). Furthermore, their explanations using natural selection appear to be a stretch. These are possible overstatements of results, allowed because of your points 1 and 2.
My overall impression of the paper is that main results are somewhat interesting and entirely novel. But there is a ton of stretching for just-so stories and "interesting" findings that will be unlikely to hold up. Typical PNAS.
I am not a genomics expert, so my comment above was intended to be more of a question directed to Real_S and those versed in genomics analysis (i.e. given this list of concerns, are these findings still legit?).
This is from Framingham. The QC was all done well before, so not having 400 authors is no more suspicious than a one or two-author paper using UKBB data from 500k people - the point of having a long-running systematic study such as Framingham or UKBB or the twin registries is, among other things, to amortize the cost of data collection and preparation, instead of expecting every paper to collect an entire new sample and build a pipeline and analyze it.
Don't know about happiness, but regarding smoking and obesity this seems so obvious to not even need a study (excluding pathological causes for obesity of course which are fewer and far between in people compared to the epidemic rates of obesity we see).
And perhaps happiness is quite impacted by living among other "positive looking/happy/chilled out" people as well -- it sure doesn't help if you're into a social network of depressed people, grumpiness and misery.
Seated meetings/meetups are my friend.
I've generally found that in a professional setting, my height is not generally an issue.
However, if, at a bar, my height is considerably (>= 6-8") below the median of the people i'm hanging out with, interjection does require more effort than it does amongst those who are closer to my own height.
It may just be pure acoustics :P
[Edit: in case it isn't clear, this is in situations where everyone is standing]
* People who are statistical outliers tend to seek out similar outliers to form a support network.
* People who are involved in activities in which their physical size is pertinent (e.g. dancers in the Rockettes) tend to befriend others in those activities.
* If height affects success (whether or not that is warranted) and if friendships are stratified along success-based lines, then you might expect groups of friends to have similarities in height.
and then studies happen and less normies take part in them.
It would have been more accurate to title this article "Friends Are Genetically Similar in Some Ways and Different in Others" but I guess that would not have been as edgy.
EDIT: The overall trend reported is that friends are more genetically similar than would be expected, as pointed out by user Real_S. Good title, after all.
If I were to volunteer for this study, all of my 'enemies' (more just people I don't care for) would either be people who I haven't seen in forever because I don't want to associate with them or people that I see too often that I wouldn't want to bluntly declare as my enemy and deal with the consequences.
People share exactly 50% of their genes with their parents and children, but only probably 50% of genes with siblings -- it could be 51%, or could be 48%. To preserve their own genes, people will feel altruistic to give their lives for three of their siblings (about 150% of their own genes saved), but not for one sibling (only about 50% saved). People share between 90% to 110% of their genes with two siblings (I'm guessing the variance here), so will give their lives for two siblings they share 110% of their genes with, but not two they only share 90% with. The feeling of "liking a sibling as a friend" is the same as being willing to give one's live for them, so in today's age of large urban groups where friendship is a proxy for one's siblings in small tribes during the hunter-gatherer era, it follows people choose their friends because they share more than the average genetic material with them.
https://en.wikipedia.org/wiki/Sociobiology:_The_New_Synthesi...
The two brothers/eight cousins idea goes back as far as Haldane regarding kin selection:
> No, but I would to save two brothers or eight cousins.
https://en.wikiquote.org/wiki/J._B._S._Haldane
Wikiquote has that as quoted in 2007, but I'm not sure how far back he originally stated it.
I don't see why the percent shared could not be higher than 50% with your parents or children. You can share a gene with someone that you did not get it from or give it to. For example, suppose some particular gene has three variants, V1, V2, and V3. If your mother has V1/V2, and your father has V2/V3, the possible cases for you are V1/V2, V1/V3, V2/V2, and V2/V3.
In two of those cases you share 50% of that gene with each parent. In the other two cases you share 50% with one parent and 100% with the other parent.
What you wrote might explain why parents have "favorite" children they're more likely to give their life for.
This makes me think of the thesis of "Sex At Dawn" by Christopher Ryan and Cacilda Jetha. If humans evolved as polyamorous foragers for whom communal bonds were more important than paternal certainty, then it makes sense that you want your friends to like the same environments and add immunological diversity to your shared gene pool.
"Third Cousins Have Greatest Number Of Offspring, Data From Iceland Shows"
https://www.sciencedaily.com/releases/2008/02/080207140855.h...
Not exactly what you'd expect given how often we see "hybrid vigor" touted as a purported benefit of interracial marriage.
Humans are still amazingly tribal creatures.
I am pretty sceptic about this study and it seems kind of filled with opportunity for interpretation and bias plus have no idea of the data set.
But as far the relation of genetics and genetic similarities to race, all genetics are related to race.
In terms of the fact that a single Adenine molecule alone is fairly useless, it is only when that molecule is combined into the structure of a genome that it takes on meaning[0], and genomes have a race, so I don't think in that sense there are any genes or genetic similarities that aren't related to race.
Every member of a race is more genetically related to every other member of that race than they are to any member of any other race.
What "genetic similarities" could there be that are more similar across races than within races?
[0]"Meaning" at the level under discussion in terms of human relationships and whatnot.
The population used in this study was the population in the Framingham Heart Study, presumably because they had the DNA and friendship graph already lying around. But starting from a small town's population, you've got pre-existing social networks which will both influence the future genetics and the future social networks of the town. You could get a lot of confounding correlations like "you're more likely to be friends with someone you live nearby to and you're more likely to live nearby to people you are related to".
If you repeated this with college students you'd have friendships formed between people with few if any previous connections.
https://iwsmt-content-ok2nbdvvyp8jbrhdp.stackpathdns.com/525...
So in effect for these similarities to present, they have to be cousins on some level right? Whether that is sharing N slots at the 2048 level, or some other combination?
In a sense all of humanity are cousins, and races are clusters with a higher coefficient of cousin-ness than the global average.
If you haven't read it yet, I highly recommend it!
We have intellects, a sense of morality, and a will. Those allow us to drastically curtail what might otherwise be instinctive behaviors.
I know there are some determinists who say that free will is illusory and we're as much slaves to our genetics and instincts as lobsters are.
But it certainly seems as if, in practice, tribalism (at least the clear-cut type you're describing where we exclude people who don't look like us) is recognizable and avoidable.
In a social setting, with people around having conversations, your brain cortex fires off. It's spinning like crazy. And the cortex is all about logic. It's all about intellect, a sense of morality, about a will, and group hierarchy, and talking, and relaxation. You get the picture. When the phobia kicks in... guess what... the first thing that will happen is that cortisol will turn off your cortex. In the stressful situation, in fight or run situation, your cortex is switched off brother. That's why this spider is dead. That's why after 30 years of perfect marriage you might do stupid things when 23 year old hot blonde is sexually provoking you. That's why live isn't white and black. Even in 2017.
Just because some people react with their emotion doesn't mean everyone does and it doesn't mean we have to.
Now, I don't fear spiders and also don't mind them as long as they don't wreak the place with their filaments. I was surprised when a cousin of mine was freaked out by one of them and wanted to insta-kill it.
Now if there is a bug... that's a whole different story. Even though, logically thinking, these beasts are the same.
Given how many spiders are only inside or only outside spiders, you basically just said you seen him sending the spiders to their death. The person smashing the spider isn't much different than the person putting it outside.
Not the best source, but a good starting point: https://www.livescience.com/48479-spider-myths-busted.html
>Just because some people react with their emotion doesn't mean everyone does and it doesn't mean we have to.
And yet consider the example you gave in light of the current body of evidence.
One can easily espouse something like that while not wanting a big spider scurrying around in your house. Also, plenty of people (myself included) escort such spiders outside alive instead of smashing them, not the least because I don't want to smear their guts everywhere, creating more work.
How is that not logical? Spreading your genes to healthy young women seems like a straightforward choice.
Actually you have to make a number of conscious decisions to have an affair. "It just happened" is an excuse.
An excerpt from Julian Jaynes's 'Origin of Consciousness' comes to mind here:
> Consciousness is a much smaller part of our mental life than we are conscious of, because we cannot be conscious of what we are not conscious of. How simple that is to say; how difficult to appreciate! It is like asking a flashlight in a dark room to search around for something that does not have any light shining upon it. The flashlight, since there is light in whatever direction it turns, would have to conclude that there is light everywhere. And so consciousness can seem to pervade all mentality when actually it does not.
There exist a great deal of subconscious processes that effect behavior in ways that an 'individual' cannot necessarily perceive. Often, a person does something, then post-hoc uses narratization to fit those actions within their preexisting belief system and story of themselves. The important thing to note is that your perception of yourself is not you, and the world you perceive around you is not the world itself. Both act as very helpful mental models that help you navigate your world effectively. As with all models, however, they have holes, and the really damaging ones are the ones you're not aware you're not aware of (as you cannot even adjust for those weaknesses).
Non-consciously accessible processes play an important role in our ability to reason, predict and control behavior in a manner Evolution could not anticipate. They are no less us.
[0] - Jonathan Haidt's "The Righteous Mind"
Further, while there may be a correlation, it's far from deterministic. Experiences count for a lot in developing values.
Examples:
- I had a friendship which originally started over shared nostalgia for a cantonese children show
- I was excluded from a group because I didn't know enough american pop culture references to get their jokes
- I stopped eating dinner with another group because I felt awkward being the only one not praying before the meal
- In another group, we only ever talked about programming puzzles and non-programmer friends (such as their girlfriends) learned quickly to leave us alone
All of these groups came together and/or left each other due to unwritten emergent social dynamics. You can never stop these clusters from forming. As long as you have groups/tribes, there will be people excluded based on the that tribe's values. Is that really so bad?
No, it doesn't. I don't know anyone that has come even close to escaping tribalism.
So instead of doing hard work, we decided to make tools to circumvent the bugs, for example by avoiding buggy routine executing personal communication - and instead discuss that paper online.
Work with the smart, no matter what skin colour, instead of doing the PC-Dance.
In case you think I'm arguing against a strawman, the blank slate view is _extremely_ popular, and likely even more so in the circles that this commentariat likely run in.
Thank you for saying this.
As history has shown, it's only too easy to take a scientific research paper and draw some godawful conclusions from it.
We can - and must - do better.
Sounds like something a plantation owner from the antebellum south would agree with.
Now obviously anti-racism isn't "the real racism", but I think it's kind of funny that many commentators here seem to think that by defining certain sects of people based on cherry picked attributes will somehow prove that we've moved beyond the biological impulses of discrimination.
There are no moral imperatives to be found in genes or epigenetics, only moral factors. Hume's observation about is -> ought is as pressing as ever, despite what Sam Harris might have you think. It's particularly dangerous to believe the former is true, since we make such a habit of expanding and distorting the scope of empirical results to ideological fantasy-land near instantly.
Whether we tightly and uniformly prefer genetically similar friends, whether we conversely happen to like genetically diverse mates, or whether all of these are more generally attendant to environmental factors like diet and disease -- it remains that we can collaboratively use reason to decide on our 'oughts', not just pick the interpretation and scope of 'is' that we prefer at the time. Issues of justice and responsibility seem much more interesting than endlessly turning the wheel of sciencey-tribalism. We struggle enough to figure out how best to treat each other already.
More to the point, though, racism is an obvious manifestation of the kind of tribalism I think we are talking about and I don't think it's really controversial to argue that it is harmful at least to the marginalized groups. I'd personally even go further and say it harms all parties involved.
It's a factor way too often ignored with this issue: If "hurting people's feelings" will become a crime, then we will see no end of regulating "undesired behaviors" because people get their feelings hurt in the most subjective and unintended ways [0].
So much about "obvious examples of racism".
[0] http://thedmonline.com/greek-life-retreat-ends-abruptly-bias...
Urbanization isn't unnatural and it agrees with our tribal instincts.
Look at Tokyo: extremely high density with extremely small social unrest. Paris, now a collection of different tribes, can't say the same.
On another note, I think Japan ends up with kind of a distorted picture since so much of the crime is associated with organized criminal groups with ties to legitimate business and government operating with a degree of openness that I think would surprise most Western observers. Dubro's Yakuza book is a great primer on the topic.
I'm not claiming that. I am claiming that multiculturalism plays a role, which many people refuse to acknowledge.
And for murder, that can be a bit of a tautology as socially unacceptable killing is murder and anything else isn't considered on the same level as murder. But if we consider killing that isn't considered murder and if you consider forms of killing that, if not legal, are punished lighter than murder, it becomes clear that our society does allow some forms of killing others.
It does if there are enough nitpicks showing that natural behaviors are becoming more accepted because they are natural. That isn't to say they are acceptable, but they are more accepted. And this doesn't have to be true of all behaviors, because there are many factors at play. It is even possible there could be a trend where behaviors are becoming more acceptable because they are natural even if all behaviors were becoming less acceptable, if there was another stronger force acting to make all behaviors less acceptable.
But if we want to get to the actual argument first presented, we would have to realize that comparing it to murder, adultery, or sexual assault is a strawman to begin with. The OP wasn't defending going around and murdering people who didn't look like you. They were saying that millions of years of evolution will influence our behaviors in small ways. It would be compared to other preferences, such as preferences in intimate partners. So, is wanting to have a specific gender, race, etc. in an intimate partner wrong? Is it socially unacceptable?
If you read the article, it's actually not at all about people looking the same.
In the study, people had friends with similar smell genetics, and different immunity system genetics.
Additionally:
> formal analysis has shown that selection for group adaptations requires special circumstances, with negligible within group selection (Fig. 4), such as when (a) the group is composed of genetically identical individuals (clonal groups, r=1), or (b) there is complete repression of competition between groups (i.e., no conflict within groups; Gardner & Grafen, 2009). It is useful here to distinguish adaptation and design from dynamics of how selection leads to design. The dynamics of selection can be examined with either an individual (inclusive fitness or kin selection) or group selection approach. However, only the individual level approach provides a general model of adaptation. The idea that individuals strive to maximise their inclusive fitness holds irrespective of the intensity of selection operating within and between groups (Section 2; Fig. 4). In contrast, as discussed above, group adaptations or maximization of fitness at the group level are only expected in the extreme case where there is no within group selection.
I really doubt your view is consistent with the conditions for selection in the group. If non-tribe-member competition caused significant competition between groups, then it far outweighs the pressure of in-group adaptation. If "friends are genetically similar" is a genetic adaptation from tribalism, then it has to be an enhancement from within the group when it lacks competition - you benefit from the adaptation relative to people within your tribe that don't have it. FYI the paper disagrees with the notion that friend similarity is likely a result of decision.
Personally I think that 1% relative similarity could easily be explained in the ambiguous language they use to explain why it was "mostly" homogenously European - why would it surprise me that the non-Europeans would tend to have less DNA in common with the friends they have while also likely having fewer friends per capita?
Though, possibly, because the most likely to be similar SNPs were related to smell (and therefore taste), and to linoleic acid, maybe it reaffirms that people who break you bread are your companions - and that there is a real difference between butter cultures and olive oil cultures :P
[0] https://www.zoo.ox.ac.uk/group/west/pdf/WestElMoudenGardner_...
That's exactly what it means. Knowing what we know about sexual reproduction and trait heredity what else could it mean? It's not like half your genome spontaneously arises in a zygote in some random location half way around the world.
I can imagine that whether or not cilantro tastes like soap, whether or not you love the smell of books, similar allergies, and if you can stand to eat spicy foods goes a long way toward the likelihood of friendship. There are bound to be social factors, geographic factors, and other factors. This study is just pointing out that friends are likely to be more similar in certain ways than people who aren't friends. Don't jump to too many conclusions on just the summary article about the study.
> They examined [polymorphisms from] 2,000 people and correlated their similarity within the roughly 1,400 pairs of friends within the group.
>the Framingham data is relatively homogeneous in terms of ethnicity, consisting largely of people of European descent.
The population they studied seems very close knit and ethnically homogeneous, and the study's inferences are conditioned on this sort of population.
Presumably the consistent background genetics of the population allowed them to control for larger genetic differences and detect the subtle differences correlated with friendships in the group.
I would be surprised if the observed effect held up in a more cosmopolitan population where other stronger factors, both genetic and sociocultural, have more of influence in the formation of friendships. And, as others have noted, the observed effect was real, but very small.