So it seems that minimally a common ancestor links the several sampled blue-eyed individuals in both Turkey and Jordan with the blue-eyed population in Denmark. Could there be a different blue-eye mutation that occurred independently in other populations later, that have not been sampled? No one can say it is impossible but there is no evidence to suggest such at this time.
My question though is really the converse - this allele seems to be pretty wide spread across several populations: why did it become so widespread? I’m not buying the association with lighter skin and therefore vitamin D levels as Black Americans actually have as much of the active form of vitamin as white Americans do (source).
Or even that it occurred many times and the other instances didn’t die out. There could be modern blue-eyed people who have some other other mutation events among their ancestry as well as the ancestor they all share in common.
It’s highly likely that a massive chunk of us non-blue-eyed folks have that blue-eyed individual among our ancestors too, by the way. By the time you go back far enough that that many people have an ancestor in common, you’re probably not that far off from the point where everyone currently living has that ancestor in common.
This paper estimated the most recent common ancestor of all humans at 2,300 years ago; and the point of identical ancestry at 5,000 years ago. The latter point is where every human alive today had the exact same set of ancestors - everyone alive at that time was either an ancestor of every modern human, or their lineage died out completely.
But it’s possible that the blue-eye mutation was actually before that, because while every diploid human has two ancestors, an individual gene in a human only has one.
Yes, diploidy has this weird effect that we all have the same set of “people ancestors” quite recently, but as you say the ancestry of genes considered individually is quite different. The expected time to coalescence (common ancestry for all alleles) for a neutral locus is a function of population size, of the order of a million years in humans. There are polymorphic loci in the MHC (immune system) that are subject to balancing selection where coalescence is >20 million years ago, way back in the common ancestor of us plus all the other great apes.
This is wrong, if it’s just a few thousand years ago it couldn’t be drift. There is literature showing strong positive selection for blue eyes and other pigmentation genes over the last few thousand years.
Yes, that is a possible explanation, sexual trait selection. That said very little is arbitrarily “attractive”; likely more often it is our tastes that have been selected for. Why would blue eyes preference be selected for? And why with the apparent difference in impact by latitude?
Keep that as one hypothesis, but I’ll also point out I was careful to say that the allele seems to have been selected for, and spread fairly rapidly. The allele causes blue-eyes when homozygous, but that is not what the allele does, per se.
The variant allele alters “P protein” production which is a small molecule transporter, specifically tyrosine across membranes, in particular (?exclusively?) in melanocytes (there are pigmented cells in the brain that produce dopamine, in the substantia nigra).
The allele is tightly linked to a cluster of genes that produce GABA receptors.
The trait, blue-eyed, the homozygous condition, is associated with alcohol tolerance and with alcohol dependency.
Hypothesis: the allele, perhaps even in the heterozygous condition, either itself or by way of a GABA receptor variant it travels with, has some other impact that had “fitness” impact specifically greater in more northern conditions. One speculation could even include that stored foods over winters would sometimes ferment and that a greater ability to tolerate the alcohol that would result in increased fitness, with homozygous blue-eyed trait tagging along.
@Northern_Piper that thread was a long and interesting read. Possibly my speculation should go there and bump it but I had already asked here.
So, if I understand this correctly, OCA2 is strongly expressed in the brain, specifically in the choroid plexus.
Which does not prove anything but OTOH it does suggest that the protein does something of some significance in the brain, and that variant alleles may alter some how it does whatever it is it does.
An important piece that hasn’t been mentioned yet is that the mutation is in a highly conserved region. This is part of the genome that doesn’t change much—not just on a generational level, but on a species level.
Most of the paper is a long bit about how they narrowed down where the blue-eye gene is, from the q arm of chromosome 15, to a 166Kbp region. Then they did some functional stuff to show that a mutation in that region actually causes blue eyes, and narrowed it down more.
The “one ancestor” thing comes from the fact that a single haplotype, h1, with 3 other variations, h2-h4, carries the blue-eye gene. (h1 is the primary haplotype, and h2-4 started as h1, but have been shuffled around a bit since the original mutation.) If the mutation had occurred multiple times, we would expect to see it on multiple haplotype backgrounds.
A rough analogy: there are lots of small boxes. You open some of them and find the brown ones contain a few different numbers, but the blue ones always contain the same number. It is a reasonable conclusion that all of the blue boxes contain the same number.
They say there has been positive selection for blue eyes, but don’t really support it. There are other explanations for blue eyes being common among people with certain genetic ancestry. It could be selection, random, a founder effect, etc. There are ways to test those things to see which explanation best fits the data, but I’ve not bothered to investigate if anybody has done the work.
For completeness, this is a direct link to the paper in Human Genetics
I mentioned drift and population bottlenecks as general principles, but then realized that if this is a common (genetic) ancestor only about 6,000 years ago, those effects cannot plausibly account for how widespread it is. It requires strong selection:
Direct evidence for positive selection of skin, hair, and eye pigmentation in Europeans during the last 5,000 y
Here we present direct estimates of selection acting on functional alleles in three key genes known to be involved in human pigmentation pathways—HERC2 , SLC45A2 , and TYR —using allele frequency estimates from Eneolithic, Bronze Age, and modern Eastern European samples and forward simulations. Neutrality was overwhelmingly rejected for all alleles studied, with point estimates of selection ranging from around 2–10% per generation. Our results provide direct evidence that strong selection favoring lighter skin, hair, and eye pigmentation has been operating in European populations over the last 5,000 y.
Wait, none of this stuff is new? I usually check and make sure I’m not answering a zombie question, but I guessed the article talking about one blue-eyed ancestor must have been some new thing. Teach me to check the dates, yeah Elberg, et al (2008), the paper this thread is about, wouldn’t have cited Wilde, et al (2014).
You are not alone - I didn’t notice the date on the paper at first. I planned to check through the literature to see if there’s anything newer of relevance, but didn’t get around to it.
Investigating this line of thought a bit further … fermentation was in fact a definitely process used to preserve food by European Mesolithic Hunter-Gatherers (see here for example). And as the transition from HG to agriculture occurred across Europe, beer production did too.
It would be very interesting to track the timing of the increase in beer production across Europe with the probable timing of the rapid spread of this allele.
It’s not clear to me how they excuse it from reading the summary, but humans were scattered from Africa to Australia to Tierra del Fuego with minimal interaction. The split with the Australia population is likely around 40,000 years ago, though it’s possible there’s some cross-connection since then. The split with America, probably about 15,000 years ago or more from a Siberian group also fairly isolated before the split.
I fail to see that there would be sufficient interaction to make the complete common ancestor pool as recent as 5,000 years ago. I assume this is a theoretical calculation.
They take into account what is known about historical population structure. You need only minimal gene flow between subpopulations because of the exponential effect of having two parents. Suppose there is an isolated subpopulation, and just one outsider immigrates. If the immigrant has children and grandchildren who remain within that subpopulation, then before too long that immigrant is an ancestor of everyone within that subpopulation. And thus, tracing back through that immigrant ancestor, the subpopulation now has the same ancestors as everyone else in the world.
