March 15, 2015

Natural selection and ancient European DNA

A new preprint on the bioRxiv studies the same data as the recent Haak et al. paper, but focuses on natural selection in Europe. Until recently, selection could only be studied by looking at modern populations, but since selection is genetic change over time effected by the environment, it's possible that studies like this will be the norm in the future.

The new study seems to confirm the results of Wilde et al. on steppe groups, as the Yamnaya had a very low frequency of the HERC2 derived "blue eye" allele and a lower frequency of the SLC45A2 "light skin" allele than any modern Europeans. The Yamnaya seem to have been fixed for the other SLC24A5 "light skin" allele which seems to have been at high frequency in all ancient groups save the "Western Hunter Gatherers".

It seems that light pigmentation traits had already existed in pre-Indo-European Europeans (both farmers and hunter-gatherers) and so long-standing philological attempts to correlate them with the arrival of light-pigmented Indo-Europeans from the steppe (or indeed anywhere), and to contrast them with darker pre-Indo-European inhabitants of Europe were misguided. If anything, it seems that the "fairest of them all" were the Scandinavian hunter-gatherers, and a combination of light/dark pigmentation was also present in Neolithic farmers and Western Hunter Gatherers in various combinations.

It also seems that both the theory that lactose tolerance started with LBK farmers and the theory that it came to Europe from milk-drinking steppe Indo-Europeans were wrong, as this trait seems to be altogether absent in European hunter-gatherers, farmers, and Yamnaya, and make a very timid appearance in the Late neolithic/Bronze Age before shooting up in frequency to the present.

Another new development is the ability to predict "genetic height" from ancient DNA. I think this may be a little bit superfluous as you can predict "actual height" by measuring long bone lengths. On the other hand, actualized height depends not only on genetics but also on diet, disease, etc., so it's useful to look at genetic changes in such polygenic traits directly.

A big surprise was the presence of the derived EDAR allele in Swedish hunter-gatherers. This allele is very rare in modern Europeans and seems to have pleiotropic effects in East Asians. This raises the question why this allele (that was so successful in East Asians), never "took hold" in Europeans. One possibility is that it never provided an advantage to Europeans (I don't think anyone really knows what it's actually good for). Another is that Swedish hunter-gatherers simply didn't contribute much ancestry to modern Europeans and so the allele never got the chance to rise in frequency by much.

bioRxiv http://dx.doi.org/10.1101/016477

Eight thousand years of natural selection in Europe

Iain Mathieson et al.

The arrival of farming in Europe beginning around 8,500 years ago required adaptation to new environments, pathogens, diets, and social organizations. While evidence of natural selection can be revealed by studying patterns of genetic variation in present-day people, these pattern are only indirect echoes of past events, and provide little information about where and when selection occurred. Ancient DNA makes it possible to examine populations as they were before, during and after adaptation events, and thus to reveal the tempo and mode of selection. Here we report the first genome-wide scan for selection using ancient DNA, based on 83 human samples from Holocene Europe analyzed at over 300,000 positions. We find five genome-wide signals of selection, at loci associated with diet and pigmentation. Surprisingly in light of suggestions of selection on immune traits associated with the advent of agriculture and denser living conditions, we find no strong sweeps associated with immunological phenotypes. We also report a scan for selection for complex traits, and find two signals of selection on height: for short stature in Iberia after the arrival of agriculture, and for tall stature on the Pontic-Caspian steppe earlier than 5,000 years ago. A surprise is that in Scandinavian hunter-gatherers living around 8,000 years ago, there is a high frequency of the derived allele at the EDAR gene that is the strongest known signal of selection in East Asians and that is thought to have arisen in East Asia. These results document the power of ancient DNA to reveal features of past adaptation that could not be understood from analyses of present-day people.

Link (pdf)

39 comments:

Krefter said...

Both of the Mesolithic Russian foragers are inferred to have light skin like the Swedish ones. Also, there was hair color diversity in Russian and Swedish foragers. Add to that the first brown eyed forager was found in Karelia.

https://genetiker.wordpress.com/2015/03/07/phenotype-snps-from-prehistoric-europe/

Arch Hades said...

So the Yamna were not predominately light eyed and haired? Where does the data show that?

Looks like this does go against all the old speculative assertions done by philologists and others that these steppe proto Indo-European people spread light eyes and hair into darker populations. I'm reminded of John V Day's book you reviewed so long ago.

So do you think Europeans just developed them on their own, gradually and selection favored significantly higher frequencies of light hair and eyes in Northern Europe? Sounds like the most plausible explanation.

Fiend of 9 worlds said...
This comment has been removed by the author.
matt said...

What happened to the supposed immune traits lacking in Y Haploid group I?

Crimson Guard said...

"The average height of the males in the ancient Yamnaya civilization was 172.4 cm, while that of the females 160.2 cm."

- István Ecsedy

So for males that would be around 5'7"/5'8".

Coon had these averages for the time periods:

"he Upper Paleolithic Europeans were modern Caucasoids. During their span of 20,000 years, their bodies changed physically very little if at all, for their adjustments to their environment left nothing to be desired. The famous "Old Man of Cro Magnon" was not a giant, as often depicted, but only 5 feet, 6 inches(168.4 cm) tall. The mean stature for twelve adult male skeletons was 5 feet, 8 inches 173 cm). The tallest 5 feet, 11 1/2 inchs(182cm), and the shortest was 5 feet, 3 inches(160cm). The five female skeletons recorded had a mean sature of 5 feet, 1 inch(155.5 cm) and a range from 5 feet, 1/2 inch(154 cm) to 5 feet, 2 inches(157.5 cm). This sex difference in stature varies closely with that in head size, as we shall presently see. "


http://i57.tinypic.com/2606gy1.png


So I don't really see what they're saying. Besides are like 50% Middle Eastern anyway.

Grey said...

"Looks like this does go against all the old speculative assertions done by philologists and others that these steppe proto Indo-European people spread light eyes and hair into darker populations."

If there were numerous depigmentation genes and the resulting phenotypes are the additive product of different combinations then they could have spread the genes that led to the combined phenotype.

example
say a phenotype needs the addition of gene x and gene y

population A has x and population B has y

then if population B spreads they add gene x and gene y together creating the phenotype

terryt said...

"If anything, it seems that the "fairest of them all" were the Scandinavian hunter-gatherers"

From what I have observed the 'fairest of them all' are still Scandinavians. Although perhaps even more concentrated along the eastern margin of the Baltic, as far as the Urals. Outside that region blonds are much more a minority.

"So do you think Europeans just developed them on their own, gradually and selection favored significantly higher frequencies of light hair and eyes in Northern Europe? Sounds like the most plausible explanation".

Agreed.

"A surprise is that in Scandinavian hunter-gatherers living around 8,000 years ago, there is a high frequency of the derived allele at the EDAR gene that is the strongest known signal of selection in East Asians and that is thought to have arisen in East Asia".

We do know that at some time there was a movement from the east into that region, hence the Y-DNA N. Presumably this carried the mutation as far as Scandinavia.

Arch Hades said...

I see the data.

Did the article have anything to say about hair color pigmentation?

eurologist said...

"It seems that light pigmentation traits had already existed in pre-Indo-European Europeans (both farmers and hunter-gatherers) and so long-standing philological attempts to correlate them with the arrival of light-pigmented Indo-Europeans from the steppe (or indeed anywhere), and to contrast them with darker pre-Indo-European inhabitants of Europe were misguided. If anything, it seems that the "fairest of them all" were the Scandinavian hunter-gatherers, and a combination of light/dark pigmentation was also present in Neolithic farmers and Western Hunter Gatherers in various combinations."

Of course, and I have said so many times. The selections on folic acid and vitamin D are too strong to make any other outcome plausible. Also, as I have pointed out many times, even a double allele of the Subcontinent/ West Asian "light skin tone" mutation can mean a very dark skin, e.g., by Mediterranean standards.

As to the derived EDAR allele, whether it is due to chance mutation or ancient human introgression, before agriculture it seems to thrive in dry-cold environments - not so much in wet-cold ones. Another Gravettian association in Scandinavia, or perhaps more realistically, more recent migrations (Ahrensburg and Hamburgian cultures?).

Wide-spread lactose tolerance may be related to the rise of standing armies and associated famine. Milk is a great source of protein, sugar, fat, vitamins, calcium, and antigens (e.g., smallpox), and is available when cheese production has ceased.

Grey said...

If different depigmentation genes developed in different regions for whatever reason then the population that would get all the different ones would be a population in the middle.

(effective middle in this case which includes East Asia via the shortest route i.e. over the top of the globe)

.

The Motala HGs are critical because they show a) the major skin lightening gene probably wasn't a farmer diet adaptation and b) that Scandinavia, maybe surprisingly for some, was the closest entry point for East Asian pigmentation genes at that time.

.

Lactose tolerance is highest in the Atlantic bio-region which has a much wetter climate - perfect for cattle but not so good for crops developed in the med. i.e. the ideal place for LP to spread dramatically.

Atlantic cowboys.

Unknown said...

terryt,

We do know that at some time there was a movement from the east into that region, hence the Y-DNA N. Presumably this carried the mutation as far as Scandinavia.

I agree, it's not strange for the region at all, but what puzzles me is that none of the Motala samples show up with East Asian admixture in any of the many ADMIXTURE analyses I've seen of them. They just have that same baseline Amerindian affinity we see in all European and Siberian HG and HG-admixed populations. The only thing that distinguishes them from other European HG populations is their ANE admixture.

It's unclear to me at this point whether real East Asian admixture sneaks into the Motala population in ADMIXTURE analyses under the cloak of ANE-affinity and all of the 'Amerindian' components that go with that, which would unravel with a greater number of ANE samples and maybe some ancient East Asian genomes besides Tianyuan. But Motala appear as solid European blue at the highest K's in the recent Haak et al. paper (Extended Data Figure1), which is more than pretty much any other ancient sample.

Mathieson et al. report a significant f4 relation between Scandinavian HGs and the Han Chinese, but they don't compare this relationship side-by-side with other HG populations. I'd prefer to see a more rigorous demonstration that Scandinavian EDAR-positive HG populations have a special relationship with the Han Chinese that EDAR-negative HG populations lack, which would then lead back to the question why ADMIXTURE doesn't pick it up. Surely with EDAR at 4/7, East Asian admixture in Motala should be significant, unless selection from very minor input was hyper-aggressive -- moreso than in modern groups, whose EDAR is often proportionate to their East Asian admixture levels (with some key exceptions, but even in those the East Asian element is far from invisible). The effect of ice age selection?

But we also see EDAR at 12.5% in modern Somalians here:

http://alfred.med.yale.edu/alfred/SiteTable1A_working.asp?siteuid=SI663326A

I've only ever seen noise-level East Asian in Somalians, so this could be additional evidence that hyper-selection of EDAR can occur in populations essentially unadmixed with East Asians, or, even more strangely, that EDAR isn't strictly East Asian.

You know a lot more about the origin of EDAR (and about the reliability of previous estimates of the time and geography of its mutation/introgression), so I'm wanting your take rather than actually proposing anything. Also, it would be interesting if someone who's good at this kind of thing could try to tease out an East Asian signal in the Motala genomes.

Pneumatikon said...

"It also seems that both the theory that lactose tolerance started with LBK farmers and the theory that it came to Europe from milk-drinking steppe Indo-Europeans were wrong, as this trait seems to be altogether absent in European hunter-gatherers, farmers, and Yamnaya, and make a very timid appearance in the Late neolithic/Bronze Age before shooting up in frequency to the present."

That's kind of a shocker. Is in possible West Asians like me spread that? (J2a-M92) Because we're the ones who started pouring in right after the Bronze Age Collapse. It's also ironic because it looks like the R1b's like the Yamnaya are the ones who first domesticated cattle. We got that from them.

Rokus said...

I am not sure anybody already noticed the allele frequencies of figure 2 show the Yamnaya profile to be most similar to the Early Neolithic profile. Parallel evolution or rather Early Neolithic contacts?

eurologist said...

I'd like to see a similar analysis as Extended Data Fig. 4 on all Neanderthal samples and on Denisova. This could clarify whether the derived EDAR allele is due to introgression or a spontaneous mutation. A chance mutation with (presumably) strong local selection effects would have been about 10 times more likely for ancient humans, living there more than 10 times as long as modern humans.

Nirjhar007 said...

@Krefter
Judging by your Pic you look exactly like one of the foragers!:D

Rokus said...

'As I said before the Motala group are dead-ends. Their pigmentation is irrelevant as is their stiff coarse Asian hair and their shovel shaped teeth due to their mixing with proto East Asians in Siberia. They are not your ancestors and their pigmentation was probably the result of their mixed origins.'
Nonsense! In the vein of this same article 'eight thousand years of natural selection' it is obvious that natural selection in Europe was directed against the expression of the EDAR gene. Extended data Figure 4 shows that EDAR-related but EDAR-less haplotypes were already available in Scandinavia and still flourish in CEU, that derive of NW Europe, ie outside Scandinavia. This means evolution rather than extinction.

Actually, I wonder the European "EDAR-related" haplotype features the same sweep as Eastern Asia for this evidence we need to decide where it ultimately originated.

terryt said...

"This could clarify whether the derived EDAR allele is due to introgression or a spontaneous mutation. A chance mutation with (presumably) strong local selection effects would have been about 10 times more likely for ancient humans, living there more than 10 times as long as modern humans".

Here is Razib's take on the original research. The authors place the mutation's oring in northern China/northeast Tibet, although they leave open the option of introgression or mutation within the modern human geneome:

http://blogs.discovermagazine.com/gnxp/2013/02/is-girls-generation-the-outcome-of-the-pleistocene-mind/

"It's unclear to me at this point whether real East Asian admixture sneaks into the Motala population in ADMIXTURE analyses under the cloak of ANE-affinity and all of the 'Amerindian' components that go with that"

It's spread is certainly not an open and shut case. Different groups appear to have kept a different proportion of incoming genes, which is what we'd expect really. As you suggest, selection is almost certainly involved in many cases. Thanks for the breakdown of distribution.

Grey said...

@Ponto

The depigmentation gene was thought to be either
1) farmer diet adaptation
2) environmental adaptation
and as it seemed to have been spread by the farmers that made it look like it was the farmer diet option.

Now it looks more likely it was an environmental adaption of some kind which implies the farmers spread it because they already had it.

This speaks to the possible source region as you'd need somewhere that was within or close to the range of the depigmentation zone and within the range of at least one set of early farmers.

As to that range if HGs around the west European coasts didn't have it then maybe the environment that selected for it was the Eurasian interior.

Also if it was an environmental adaptation then it makes it more likely to be neanderthal.

.

"Lactase retention originated in Europe on the steppes of Hungary"

It doesn't matter where it originated if it was still very rare when it arrived at the Atlantic coast.

LP is highest in the Atlantic coast bio-region and so far only a very small number of early dna samples have LP so the question is did LP expand dramatically in the Atlantic coast region from a few individuals with LP who arrived there from wherever.

This would explain a lot imo.

eurologist said...

Hamar Fox,

If the mutation or introgression first occurred in Siberia, then there is no need for a combined East Asian admixture with the derived EDAR allel in Europeans.

If it is introgression, it could be a very old mutation that simply did not have any advantage outside dry-cold areas. So it randomly could occur in Africans. Again, a similar analysis as that of extended data Fig. 4 would likely answer these questions.

terryt said...

"You know a lot more about the origin of EDAR (and about the reliability of previous estimates of the time and geography of its mutation/introgression), so I'm wanting your take rather than actually proposing anything".

I am interested in the gene because over the years I have become fascinated by the Polynesian expansion into the Pacific and their origin in Asia. It has long been obvious to many scientists studying the subject that there was a major expansion of 'Mongoloid' genes into SE Asia from further north starting around 10,000 years ago. This has been consistently overlooked by others who are obsessed with the idea that the Mongoloid phenotype developed gradually as humans, including Y-DNA O, moved north. That idea doesn't make sense. The study on the origin of the EDAR370A mutation and its spread provides evidence for the southward expansion idea.

The EDAR370A mutation is the only one so far discovered that can be associated with a fairly well-defined regional characteristic. Until the discovery of more such genes for the particular regional characteristic EDAR370A has to act as a proxy for the Mongoloid phenotype's expansion. The authors' conclusions regarding the matter make complete sense to me. But the mutation's development can tell us much more. It is a perfect example of a gene arising in a particular region and then expanding from that region for some distance through the human species. All genes, including haploid ones, have a similar history, although originating at different times and places.

You will have heard the statement, 'each gene has its own evolutionary history'. It has been said often enough. Even as far back as in this 1999 paper:

http://www.sciencedirect.com/science/article/pii/S0002929707617200

It is important to remember that, although we individually receive genes from only two people, not all our genes come from only two people. I have just given up on an argument over the connections between early Y- and mt-DNA haplogroups in Africa with a person who seemed unable to understand that simple fact.

I'll continue, I'm afraid.

terryt said...

Some more ideas arising from consideration of the EDAR370A gene.

The suite of genes the human species is left with today come from a variety of sources. We are therefore, in a sense, wasting our time looking for a single time and place where 'the modern human species originated'. It is possible to see the human species as being a single, slowly changing, regionally diverse collective from the time of H. erectus, in fact obviously before that.

Most mammal 'species' are perfectly capable of forming fully-fertile hybrid offspring until at least half a million years of genetic isolation. That is why it is so difficult to define where 'subspecies' ends and 'species' begins. But that is the period when Neanderthals and modern humans (or their ancestors) became genetically isolated from each other. But Neanderthals were later able to leave the genetic equivalent of one Neanderthal great-great grandparent in all individual non-African humans.

By one million years a of genetic isolation breeding between mammal species becomes more haphazard. By then usually only female hybrid offspring are fertile. That is the period when Denisovans and Neanderthals/modern humans became genetically isolated. Yet specifically Denisovan genes survive in SE Asia, Australia and New Guinea, mostly at about the same ratio as do Neanderthal genes in all non-African humans. Each individual has the equivalent of one great-great grandparent who was a Denisovan.

I am sure that more examples of genes from 'archaic' populations will be discovered. In fact I'd place the origin of the EDAR370A mutation in modern humans to just such an archaic population.

terryt said...

"for this evidence we need to decide where it ultimately originated".

The original paper on the subject placed the EDAR370A mutation's origin squarely in Northern China/Northeast Tibet. Whether the 'European "EDAR-related" haplotype' is EDAR370A or not seems debatable.

"A chance mutation with (presumably) strong local selection effects would have been about 10 times more likely for ancient humans, living there more than 10 times as long as modern humans".

That is the main reason I suspect introgression from an 'archaic' species.

eurologist said...

"Here is Razib's take on the original research..."

Thanks, Terry (I believe the link is to his old blog, though).

I am glad Razib stated: "Another question is whether we know the genetic architecture of pigmentation well enough to actually infer that these ancient populations are easily predicted in their trait character by modern models which map genotype to phenotype. In other words, were Paleolithic Europeans light skinned because of different alleles?"

- a point I have made numerous times, in the past.

And, how often do I need to state that the derived allele of SLC24A5 does not particularly express light skin? It is much more likely that it allows for higher seasonal skin color variation (tanning). This is particularly important for agriculturalists, who spend a lot of time outdoors and without tree shade and in the middle of the day in the summer, as opposed to HGs, but very little time outdoors at all in the winter.

Fiend of 9 worlds said...
This comment has been removed by the author.
Rokus said...

'The EDAR370A mutation is the only one so far discovered that can be associated with a fairly well-defined regional characteristic.

Until the discovery of more such genes for the particular regional characteristic EDAR370A has to act as a proxy for the Mongoloid

phenotype's expansion.'

I seriously doubt this is necessarily the case. Extended data Figure 4 shows that also the EDAR-related 'ancestral' haplotype, almost identical except for lacking the derived rs3827760 allele of in the EDAR370 mutation, was already available in Scandinavia. This

haplotype still flourishes in CEU, that also reflects the genetic composition of NW Europe, ie outside Scandinavia. Kamberov et al.

(2013)'s figure 1A shows the 'dark blue' ancestral haplotype is most common among Caucasoids (on average over 75%), although his grid of 24 SNP's was probably much too loose. Instead, the ancestral haplotype of Motala and CEU can hardly be distinguished at all from the EDAR370 haplotype, using all sites within 100kb of rs3827760! This implies either the Motala ancestral haplotype being truly the closest ancestral predecessor of the EDAR370 mutation, or otherwise evolution within Europe back to an ancestral state. Extinction of the people during 'eight thousand years of natural selection', as the title of Mathieson et al. article goes, seems to me out of the question. The article favors the arrival of EDAR370 by migration, though the continuation of the closely related 'ancestral' haplotype in northern Europe should at least imply evolutionary adaptation directed against the expression of the EDAR gene rather than

extinction. Not so hard to imagine, with derived rs3827760 resulting in a massive valine to alanine substitution at position 370 of the amino acid sequence (V370A) whose effects on tooth morphology and hair thickness are noticeble enough to be subject to sexual selection at will. However, selection is not at odds with that other explanation, of EDAR370 actually deriving of northern Eurasian populations. Especially when we consider the dearth of the ancestral 'dark blue' haplotype in the East Asian region of origin according to Kamberov et al.'s figure 1A.

terryt said...

"I believe the link is to his old blog, though".

It is, but it is the only link I have found that covers the region the authors considered to be the mutation's origin.

Ingo Bading said...

By the way, here is a new aDNA-study - Linearbandkeramik seems to be genetically very similar to Starcevo:

http://rspb.royalsocietypublishing.org/content/282/1805/20150339?etoc

Tobus said...

@eruologist:
I am glad Razib stated:..."were Paleolithic Europeans light skinned because of different alleles?"

And I'm glad he concludes: "but that’s not a particularly parsimonious solution from where I stand"

And, how often do I need to state that the derived allele of SLC24A5 does not particularly express light skin? It is much more likely that it allows for higher seasonal skin color variation (tanning).

The studies identifying the depigmentation effects of SLC24A5 were done on unexposed skin - typically under the arm - so tanning effect was controlled for. It really does make skin lighter, no matter how many times you "need" to state otherwise.

I suspect part of the problem you have with this is your frame of reference - white people in white societies tend to call people "dark" who non-whites is non-white societies would call "light". The reflectance data in Jablonski 2000 make it clear that many people considered "dark" by European standards are in fact lighter than the average human in an objective measurement. For example you have previously used Razib as an example of a double-derived individual at this locus who has dark skin. What you probably don't realise is that the average South Asian is at roughly the median point of skin tone variation worldwide, and Razib, being lighter than the average South Asian, is probably a standard deviation or more towards the "light" end of the scale... in a global context he does indeed have light skin. He recently described himself as "notably brown, though obviously not extremely dark". There is a whole lot of variation between "brown" and "extremely dark" and SL24A5 is a significant part of that variance.

eurologist said...

"There is a whole lot of variation between "brown" and "extremely dark" and SL24A5 is a significant part of that variance."

Tobus,

I totally agree with that - but please let us leave out racial connotations or arguments that do not have any place here, nor in any scientific debate.

My point is that the difference SL24A5 makes is completely insufficient to regulate the vitamin D <-> folic acid balance required at high latitudes, and that it is much more likely to have been introduced as a new requirement due to agriculture, which suddenly set extremes of annual UV exposure differences.

Tobus said...

@Eurologist:
but please let us leave out racial connotations or arguments

Huh? I don't think I used any racial connotations or arguments, I just pointed out that if you grow up in a culture where a phenotype at the extreme of human variation is considered "the norm", then your subjective interpretation of that variation is likely to be skewed away from an objective one. If an objective measurement of sample X falls on one side of the global spectrum and our cultural norms say the opposite, then our natural judgement is out of whack with reality and needs some calibration before we can trust it in a scientific context... this applies to skin colour, height, running speed, spiciness of food, whatever.

My apologies if it came across like I was calling you a racist - I wasn't.

My point is that the difference SL24A5 makes is completely insufficient to regulate the vitamin D <-> folic acid balance required at high latitudes

But SLC24A5 is only one of three genes known to affect European pigmentation (and there are probably more that aren't yet known), no one that I know if is saying that it is completely sufficient to regulate anything at high latitudes, rather that it is part of a combination that does so (if indeed that is the selective factor, AFAIK the jury is still out on exactly why light skin was so strongly selected for).

it is much more likely to have been introduced as a new requirement due to agriculture

Makes sense, and yet both SHG and EHG have it in addition to the SLC45A2 mutation that also increases depigmentation and which EEF don't have... go figure! It seems that Western HGs were quite dark while Northern/Eastern HG's were almost as light as modern Europeans and Southern farmers were somewhere in the middle - I don't know the reasons why and I'm not about to guess, but it appears to be more complicated than a simple HG vs farming scenario.

Grey said...

@eurologist

"it is much more likely to have been introduced as a new requirement due to agriculture"

I'd agree it seemed more likely but it doesn't appear to be so.


@Tobus

"It seems that Western HGs were quite dark while Northern/Eastern HG's were almost as light as modern Europeans and Southern farmers were somewhere in the middle - I don't know the reasons why and I'm not about to guess"

As an amateur imma guess it's something to do with distance from the coast combined with latitude i.e. something that *either* lighter skin or seafood/water provides hence the triangular distribution in Eurasia with the base of the triangle across the top and the point of the triangle in the Himalayas.

eurologist said...

"It seems that Western HGs were quite dark while Northern/Eastern HG's were almost as light as modern Europeans and Southern farmers were somewhere in the middle..."

Tobus

Yes, it may seem that way based on a very limited subset of now-known alleles - but it makes no sense, given what we know about current world-wide distributions and a few ancient ones. A more parsimonious argument is that we just haven't figured out what alleles made the WHGs at least moderately light skinned (similar to extant Europeans of the same latitude).

Grey said...

@myself

"triangular distribution in Eurasia with the base of the triangle across the top and the point of the triangle in the Himalayas"

in case not clear:

Baltic - Siberia - Himalayas as the points of the triangle

Tobus said...

@eurologist:A more parsimonious argument is that we just haven't figured out what alleles made the WHGs at least moderately light skinned

I guess I don't find that parsimonious because I don't think that everyone who ever lived in Europe automatically had to have light skin. I see an introduction of light skin into Europe at ~7,000 years ago from both the South (SLC24A5) and the North/East (SLC45A2) more than enough to explain current worldwide distributions. If a theory of *why* light skin evolved doesn't make sense with that then I'd be looking at changing the theory, not trying to change what the facts suggest.

While your idea is certainly plausible, it's ultimately non-falsifiable and we could argue about it forever. Although I find it unnecessary, improbable and thus *less* parsimonious to suggest there is a now extinct allele that conferred a trait that was highly selected for in the populations that wiped it out, I'm happy to let you believe it if you wish. I will however continue to correct if you make statements that aren't factually correct (such as "the derived allele of SLC24A5 does not particularly express light skin").

eurologist said...

"Makes sense, and yet both SHG and EHG have it,..."

Tobus,

EHG? What paper is that from? Perhaps I missed something, thanks.

As to SHG, four in five of the Motala samples also have the derived EDAR gene, proving they are more "cosmopolitan" than other European HGs. And by 5,700 BCE, there is a good chance the gene could have spread north for quite a while, if it was so beneficial. Note that there was a recent paper that suggested spread of mtDNA from the Levant northward just after ~ the younger Dryas, before the spread of agriculture.

It just appears that agriculture truly made it wide-spread and eventually essentially fixed it in Europeans. The authors say:

"In contrast, the derived allele of SLC24A5 increases rapidly in frequency to around 0.9 in the Early Neolithic, suggesting that most of the increase in frequency of this allele is due to its high frequency in the early farmers who migrated to Europe from the southeast at this time, although there is still strong evidence of ongoing selection after the arrival of farming."

Tobus said...

@eurologist:
EHG? What paper is that from? Perhaps I missed something, thanks.

No paper (yet!). It originally came from a blogpost at Genetiker, but to be sure I downloaded the BAM's and checked for myself. Krefter on the Eurogenes blog also posted a summary which confirms it. Both are fixed derived for SLC24A5, Samara is fixed derived for SLC45A2 while Karelia is heterozygous. I'm surprised they didn't include this info in the Mathieson paper, but there are only 2 EHG samples and one of them (Samara) is only 2-3x coverage so perhaps that's why.

SLC24A5 seems to have been widespread all around Europe before the Neolithic, and was eventually introduced from the South by the farmers (I note that it's still widespread from North Africa to India).

SLC45A2 seems to have originated in SHG/EHG to the North and East and didn't become common in Western Europe until the Bronze Age (introduced from the Steppe?). It also seems to have been common in the Bronze Age Scytho-Siberians as far East as the Altai Mountains and Tarim Basin (see here) but seems to have pretty much disappeared from there (due to Mongol/Han expansions?)

eurologist said...

"Although I find it unnecessary, improbable and thus *less* parsimonious to suggest there is a now extinct allele that conferred a trait that was highly selected for in the populations that wiped it out, I'm happy to let you believe it if you wish. I will however continue to correct if you make statements that aren't factually correct (such as "the derived allele of SLC24A5 does not particularly express light skin")."

Tobus,

I think you still don't understand what I am saying. People with the derived allele, even double-carriers, do not necessarily have "light" skin by European standards - and we are talking about Europe, here. In fact, they can be darker-skinned than about pretty much any person with largely European ancestry (including the Mediterranean) and even from large parts of West Asia. Thus, logically, the other alleles that give Europeans light skin have not died out, at all - they are alive and kicking.

This also fits with the well-known theory of skin color selection by sunlight (mainly UV) exposure, in which there must be a balance between vitamin D production and folic acid destruction (which can be shifted toward darker tones if the diet consists of a lot of seafood, and not just seasonally).

Tobus said...

@eurogenes:
even double-carriers, do not necessarily have "light" skin by European standards

But Europeans standards are an *extreme* and aren't the same as evolutionary standards. You don't have to be completely white to get an evolutionary benefit from depigmentation - Razib would get more vitamin D from sunlight than Samuel L Jackson does for instance.

and we are talking about Europe, here

Ancient Europe though, not modern Europe - WHG is only one of the three main contributing populations to modern Europeans, and the other two (assuming ANE came from SHG/EHG or later pops) both carry depigmentation alleles... so no need for WHG to be a source. The latest data also suggests that pre-Neolithic Europeans were "on the brink of extinction", so perhaps they weren't very well adapted to the environment in the first place.

the other alleles that give Europeans light skin have not died out, at all - they are alive and kicking.

I agree with that. Although the effect of SLC24A5 has been quantitatively estimated at about 1/3 of the difference between modern Europeans and modern Africans, which means that if we assume Loschbour and La Brana have *ALL* possible skin lightening alleles *except* SLC24A5 then the very *lightest* we'd expect them to be is about 2/3 of the way between modern African and modern European - and according to Jablonski's reflectance data this puts them at roughly the same skin tone as the average modern South Asian. On top of this, they both lack SLC45A2 and TYR, the other two known contributors to modern European "whiteness" and so we'd expect them to be noticeably darker than this (with the unknowns still making them "light" compared to modern Africans).

This also fits with the well-known theory of skin color selection by sunlight (mainly UV) exposure

Of course it "fits", paleo-Europeans being white was one of the assumptions that help create the theory 15 years ago. Using that theory as a reason to insist paleo-Europeans *had to* be white is a circular argument though.

As I said in a post above, if theory and data don't align, you change the theory, not the data.

eurologist said...

Coming back to Extended data Figure 4 (100kb of SNPs around the EDAR gene). A few notes:

- CEU is much more diverse in this region than CHB.
- several CEU share some of the CHB SNPs that appear linked to EDAR, but only a subset. This indicates that CEU harbors several older lineages related to the one the derived EDAR one occurred in or introgressed, into. See also the 9th column from the right, where two of the best matches in CEU carry the non-CHB/ non-derived EDAR-linked SNP.
- some Motalas have SNPs (e.g., columns 11, 13, 29, 41) that are common in CEU but non-existent or rare in CHB, indicating that they were close to the origin of CEU, as expected.
- some Motalas carry a CHB SNPs (column 20) that doesn't exist in the CEU samples, indicating that their input is indeed C or E Asian, but very shortly after the EDAR mutation or introgression.
- the CHB who carry single, specific CEU SNPs are as expected heterogyzotic in EDAR, except if far away from the EDAR position.

Given that CEU appears to be ancestral in several lineage where EDAR first occurred, it now looks to me like mutation - not an introgression. There is no identifiable block of SNPs that would suggest introgression, unless it took place several times with and without the derived EDAR allele.