January 26, 2014

Brown-skinned, blue-eyed, Y-haplogroup C-bearing European hunter-gatherer from Spain (Olalde et al. 2014)

There is nothing like a little ancient DNA weirdness to start off 2014, which promises to be as exciting as 2013 was.

The new study La Brana 1 identifies it as ancestral in the SLC24A5 locus in which virtually all Europeans are derived. This comes in the heels of the Loschbour preprint which identified that sample from Luxembourg as also being ancestral. Taken together, it's now clear that hunter-gatherers from Mesolithic Western Europe were brown.

Curiously, it now seems that both Europe and India were (in part) inhabited by brown people and became lighter by a process of admixture + selection. The process went "all the way" in Europe, but a cline of pigmentation was sustained in India.

The other finding (not mentioned in the abstract) is that La Brana 1 belonged to Y-haplogroup C6! This is a low-frequency European clade of haplogroup C. So now, we have evidence that haplogroup C is not eastern Eurasian (as the presence of its subclades in Australia, India, East Asia, and the Americas might suggest), but a pan-Eurasian entity. It remains to be seen whether this C-in-Europe can be pushed further back in time, but finding it in Mesolithic Iberia reduces the chance that it's some random eastern Eurasian who made it to the outskirts of Europe recently.

Finally, La Brana 1 has derived alleles at loci associated with pathogen resistance. This might be important, because a common hypothesis is that Europeans developed this type of resistance during the Neolithic as they started interacting with the pathogens of domesticated species and started living in less-hygienic higher-density settlements.


Nature (2014) doi:10.1038/nature12960

Derived immune and ancestral pigmentation alleles in a 7,000-year-old Mesolithic European

Iñigo Olalde et al.

Ancient genomic sequences have started to reveal the origin and the demographic impact of farmers from the Neolithic period spreading into Europe1, 2, 3. The adoption of farming, stock breeding and sedentary societies during the Neolithic may have resulted in adaptive changes in genes associated with immunity and diet4. However, the limited data available from earlier hunter-gatherers preclude an understanding of the selective processes associated with this crucial transition to agriculture in recent human evolution. Here we sequence an approximately 7,000-year-old Mesolithic skeleton discovered at the La Braña-Arintero site in León, Spain, to retrieve a complete pre-agricultural European human genome. Analysis of this genome in the context of other ancient samples suggests the existence of a common ancient genomic signature across western and central Eurasia from the Upper Paleolithic to the Mesolithic. The La Braña individual carries ancestral alleles in several skin pigmentation genes, suggesting that the light skin of modern Europeans was not yet ubiquitous in Mesolithic times. Moreover, we provide evidence that a significant number of derived, putatively adaptive variants associated with pathogen resistance in modern Europeans were already present in this hunter-gatherer.

Link

338 comments:

1 – 200 of 338   Newer›   Newest»
Glossy said...

I'm curious, did these brown-skinned, blue-eyed hunter-gatherers have modern European facial features? I see that the abstract talks about a "skeleton". I'm assuming that means that there's a skull.

Grey said...

Varieties of C as the result of numerous regional adaptations after a coastal expansion from a common source population?

(Khoisan-like imo)

mooreisbetter said...

Hmm... So some of the oldest NRY DNA we have discovered recently has been R* and C6. Both clades are found in the far reaches of west and east Eurasia.

If we apply the Wave Theory that is so in vogue among R1b apologists to the older clades, we come up with the simple explanation that both entered Eurasia through the middle of the continental mass, and spread out to the Far East and Far West.

Perhaps C came to dominate the east and R the west. Either way, this is more evidence for the likelihood that some kind of R1b precursor did exist in western Eurasia for a very long time, perhaps in a small and isolated population pocket.

Then, likely R1b expanded due to an innovative and beneficial mutation, like lactase digestion, in more recent times, which contributes to the complex and at times frustrating analysis.

In simple words: Wave theory seems to apply to both C and R. Both seemed to have entered Eurasia from the middle point of the continent, consistent with Out of Africa. They then both seem to have expanded and diversified in the far east and far west.

R is old in Eurasia. R1b or a precursor likely existed in far west Eurasia for a very long time, but expanded and further diversified much more recently (perhaps during Bell Beaker times) due to some other innovation.

Can we finally put to bed all the Eurocentric notions of R1b being the mark of the "Indo-European conqueror?"

Rokus said...

And now in Iberia, again no trace of R1b. I never believed so much in Iberian Bell Beaker expansions and rather adhere to Barry Cunliffe's view this culture was an oversea extension of the western Corded Ware complex. This find of rare YDNA C6 is consistent with a Late Neolithic advance of R1b to the expense of all previous Iberian YDNA.

Moreover, C6 being apparently much more ancient in Europe confirms the distribution of C(xC3) to be essentially congruent with YDNA MP and hence "north Eurasian" in origin, subject to similar migration paths towards the south. Maybe due to earlier expansions?

Another indication as well that "European Mesolithic" isn't really that close genetic unity as usually being though.

barakobama said...

Finally I have been waiting for this paper. So much has been learned about Mesolithic Europeans through Laz 2013 and this in just about one month!

What hair color did he have? Black like Loschbour? Did he have farmer and ANE ancestry? Based on admixtures him and La Brana-2 were put in I think they had some farmer ancestry and also ANE.

So it seems the three "light skin" mutations originated in the near east and spread to Europe in the Neolithic, this a huge discovery and proves Loschbour was not a fluke. All three of the light skin mutations(only the one in SNP SLC4A5 is around 100% the others can be under 50% in Europe) are around as popular in near easterns as Europeans so I think it was wrong for people to originally assume they are European mutations, because now we know they are near eastern. Since there is still a large amount of Mesolithic ancestry in Europe(possibly takes up majority of ancestry for most northern Europeans) these three light skin mutations may be passed down even if someone has a small amount of ancestry from the people the mutations come from.

Because near easterns obviously have brownish skin I doubt these mutations make a huge effect on skin color. Southern Arabians have basically the same skin color as Syrians even though southern Arabians almost never have these light skin mutations. I am sure there are other factors to pale skin in Europe.

Today in Europe Neolithic near eastern farmer ancestry or just near eastern ancestry period correlates with dark hair, dark eyes, and white-olive skin. While Mesolithic ancestry is the opposite it correlates with light hair, light eyes, and white skin. Since both Mesolithic west Europeans La Brana-1 and Loschbour had the "blue eye" mutation that they also probably had light skin and some light hair.

Loschbour ~8,000 year old hunter gatherer from Luxemburg probably had black hair or at least dark hair so that is not constant with Baltic people who have majority light hair and the highest amount of Mesolithic European ancestry.

Today blue eyes correlate with light hair and skin you never see any modern population with a high amount of blue eyes and also with dark skin. This is why I think there is also a good chance Mesolithic Europeans had light skin. Of course I think it is also a possibility they had dark skin based on La Brana-1 and Loschbour not having any of the three light skin mutations.

Loschbour(~8,000 year old hunter gatherer Luxemburg) and Motola12(~8,000BP hunter gatherer Sweden) probably had Y DNA pre-I2a1b or a close relative to I2a1b and now we know La Brana-1(~7,000 year old hunter gatherer from northern Spain) had Y DNA C-V20. This is evidence Mesolithic west and north Europeans paternal lineages have not survived well(I1, I2a2, and I2a1b3 L161.1 though may descend from them). I think most hunter gatherer ancestry in modern north-west Europeans comes from Indo Europeans so east European hunter gatherer ancestry. So I guess it is possible some Mesolithic Europeans were dark and some were light.



Daniel Szelkey said...

This proves that dna from modern populations has no direct association with ancient dna. C6 is very rare in Spain, rarer than Q. or O, or A. The fact that la brana had C- proves that no one knows where anatomically modern humans began. The out of africa theory is just a hypothesis, until we have mesolithic/paleolithic dna from a diverse area of Africa.

Annie Mouse said...

I think we should be cautious with these La Brana folk. They may not be typical of the area. I have long felt that our view of mesolithic Europeans is possibly being distorted by imported burial practices (as opposed to air, fire and water rituals as mentioned in our oldest oral European literature). Burials and entombment (as in La Brana) create more durable remains.

La Brana could represent nomadic folk following game back and forth across Europe and Asia.

http://en.wikipedia.org/wiki/File:Distribution_of_Haplogroup_C-M217_Y-DNA_-_worldwide.png

In my opinion that still makes them mesolithic Europeans or accurately Eurasians, but they may not have been typical, or even resident. Just part of the rich milieu.

Fanty said...

Reconstruction of that guy:
http://img.welt.de/img/evolution/origs124249541/3739723946-w900-h600/7000-Jahre-altes-Erbgut-entschluesselt.jpg

I would say, absolutely mesolithic hunter/gatherer stereotype ("Borreby" like skull). Except for the dark skin. And I think earlier, they would have made his hair blond.

Stenhög said...

Ok,

I think we have the reason behind the blue eyes in Europeans explained here also. The early Europeans did not possessed the (West Asian) light skin gene, but they still 'needed' lighter skin. So, they were selecting heavy for Blue eyes, as a side effect of this is lighter skin. If you don´t believe me on this matter, take this for an non refutable proof. I have one brown eyed sister and one blue eyed and there is a remarkable difference in their skin colour and the ability to tan.

Furthermore this can also explain the prevalence of red hair in the European population as this also is a trick to achieve lighter skin.

terryt said...

"La Brana 1 belonged to Y-haplogroup C6! This is a low-frequency European clade of haplogroup C. So now, we have evidence that haplogroup C is not eastern Eurasian (as the presence of its subclades in Australia, India, East Asia, and the Americas might suggest), but a pan-Eurasian entity".

Someone (I'm sorry, I forget who) had mentioned this a week or so back. Fortunately more work has been done on C recently and what was 'European' C6-V20 is now part of 'Japanese' C1. Specifically it is now C1a2-V20. Furthermore the old 'South Asian' C5-M356 is now C1b-M356. The three haplogroups are related to each other more than they are to the other Y-DNA Cs:

http://www.isogg.org/tree/ISOGG_HapgrpC.html

Quote from ISOGG:

"C-M356 has a significant presence in southern Asia. C-V20 is most common in Europe".

That means that just this single branch of C (C1) is 'a pan-Eurasian entity'. My guess is that, like R and Q, it originated near Pakistan/Iran and spread with them into northern Eurasia. In other words it formed part of the Ancient North Eurasian Upper Paleolithic spread. The other three C haplogroups are independent of that spread, unless the whole C spread was via Central Asia and not via South Asia as usually accepted.

Tobus said...

Typo: You say "SLC45A5" in the text but but I think you mean SLC24A5 judging by the link... The other "SLC" gene associated with European light skin is SLC45A2 and these results show La Brana had the ancestral allele at this site too.

Grey said...

mooreisbetter
"In simple words: Wave theory seems to apply to both C and R. Both seemed to have entered Eurasia from the middle point of the continent, consistent with Out of Africa. They then both seem to have expanded and diversified in the far east and far west."

Huh?

Why would R enter from Africa.

C entering from Africa and spreading along the coasts in all directions and developing into regional sub-populations makes sense.

Then one or more of those regional variations developing an advantage and spreading as part of its own "Out Of" event.

Then one or more of the daughter populations of that expansion developing an advantage and having an "Out Of" event

rinse and repeat

until finally this sequential "Out Of" process reaches R.

The later in the sequence the further from the start.

German Dziebel said...

Importantly, "the affinity of the Mal’ta individual and the La Braña 1 individual is not due to some ancient DNA attraction, and it is instead due to a shared genetic ancestry of the Mal’ta individual and the La Braña 1 individual." (Suppl Mat, p. 17-18).

Also, "Table S10 presents result for this analysis, indicating that the Mal’ta individual is closer to the La Braña 1 individual than it is to any European population, with the exception of Orcadian
and Russian populations. We also calculated D statistics using all the East Asian population from the HGDP, and found that the Mal’ta individual is significantly closer to the La Braña 1 individual than it is to the East Asians."

Since Mal'ta is the closest to Amerindians (Raghavan et al, Lazaridis et al.), this finding confirms the "easternmost" (Amerindian-like) affinities of European hunters-gatherers. Needless to say, La Brana's mtDNA allocation within hg C supports this, too.

mooreisbetter said...

Rokus,

Except...

This individual was autosomally very very close to modern populations bearing R-derived haplogroups, including your precious R1b.

So, unless he and his counterparts (magically) had time to intermix (substantially) with people with genetics like modern R1b populations (by using a time machine), then there was likely some R* or R1* in Iberia at the time.

Remember, Spain is a very big place. It is perhaps the soundest and most prudent to simply assume that this is one sample on a big land mass, but since he is so close to R populations and since R is so equally Eurasian, that we will find a small pocket of R1 in ancient DNA sometime soon, which could have led to the Bell Beaker expansion.

Average Joe said...

What are "R1b apologists?"

Average Joe said...

Can we finally put to bed all the Eurocentric notions of R1b being the mark of the "Indo-European conqueror?"

Wouldn't any discussion of the population of Europe be - by definition - Eurocentric?

RayBanks said...

This study is actually a wee bit obsolete. We changed the ISOGG C tree a few days ago to show that V20 and the existing C1 and C5 groups are actually within a new C1 group.
http://www.isogg.org/tree/ISOGG_HapgrpC.html
The curiosity of this new grouping is that V20 is parallel to the old C1 group which is overwhelmingly Japanese. Because Australian Aborigine samples are not available, it is unclear where they fall in the new C tree

terryt said...

"Varieties of C as the result of numerous regional adaptations after a coastal expansion from a common source population?"

Unlikely. Neither C1'5'6 nor C3 are particularly 'coastal'. That leaves just C2 and C4, but they are coastal only in the sense they had to cross Wallace's line to reach the region where they are now found.

"If we apply the Wave Theory that is so in vogue among R1b apologists to the older clades, we come up with the simple explanation that both entered Eurasia through the middle of the continental mass, and spread out to the Far East and Far West".

As well as C and R you need to add Q. Makes complete sense to me.

"Wave theory seems to apply to both C and R. Both seemed to have entered Eurasia from the middle point of the continent, consistent with Out of Africa".

R's history is far more complicated than a simple OoA. It descends from MNOPS, specifically the NP branch within that haplogroup (as does Q). MNOPS almost has to be Southeast Asian in origin.

"Moreover, C6 being apparently much more ancient in Europe confirms the distribution of C(xC3) to be essentially congruent with YDNA MP and hence 'north Eurasian' in origin, subject to similar migration paths towards the south. Maybe due to earlier expansions?"

The patterns for C and MP look to be completely different from each other. Just becasue two or three downstream mutations within the groups happen to have spread into North/Central Eurasia together in no way requires they had always been together.

"I think we should be cautious with these La Brana folk. They may not be typical of the area".

Good caution.

Nathan Paul said...

Terry:

"whole C spread was via Central Asia and not via South Asia as usually accepted"

You mean C reached Burma, Thailand, Vietnam via where?

Slumbery said...

Rokus

I do not think this finding has any effect on the theory of Iberian Bell Beaker source. A lot of time and more importantly a lot of known migrations separate this sample from BB. We cannot say that farmers changed the genetic landscape and that a single HG finding must reshape our view on later local farmer DNA in the same time.

Generally the entire uni-parental ancient DNA data is just some small fragment, we have no way to know if a Hg we find is typical or we just picked up something rare by blind luck.

Kristiina said...

In order to keep in mind also the other half of the story, both La Braña individuals have mtDNA U5b2c1!

According to an older mtDNA study of La Braña remains (http://www.ncbi.nlm.nih.gov/pubmed/22748318),”the mitochondria of both individuals are assigned to U5b2c1, a haplotype common among the small number of other previously studied Mesolithic individuals from Northern and Central Europe... Furthermore, analyses of 1.34% and 0.53% of their nuclear genomes, containing about 50,000 and 20,000 ancestry informative SNPs, respectively, show that these two Mesolithic individuals are not related to current populations from either the Iberian Peninsula or Southern Europe”.

According to FamilyTree, ”U5b2c has an age estimate of about 15,000 years based on 20 FMS samples. It has been found exclusively in western Europe. There is one U5b2c* person with ancestry in Ireland. U5b2c1 has 6 FMS samples including 2 from Spain, and one each from Ireland, England and Germany. One of the Spanish samples is from ancient human remains. Sanchez-Quinto et al. reported a FMS test result for the 7,000 year old remains of a Mesolithic hunter-gatherer at the La Brana-Arintero site which they identified as U5b2c1 … The presence of U5b2c1 in Ireland and northwest Spain might be indicative of early population exchange between those areas.”

Creative said...

I would also like to see more ancient DNA samples from the ME and especially the Levant.Tests on today’s populations always some up to drivel about Jewishness or Arabness."Frankly I don’t trust any side"

eurologist said...

I am sure there are other factors to pale skin in Europe

Barak (and also Stenhög),

Yes, of course - as I have mentioned in the past, this is a necessity since the Neolithic/ SE European/ West Asian variants don't really code for particularly light skin at all in the European context. And as you state, this causes a huge problem: if we don't know what those mutations/ locations are, we really have no clue what skin (or even hair) color Mesolithic Europeans had - but most likely it was not very dark or brown, at all.

eurologist said...

Let me repeat what I wrote in the other recent light skin pigmentation thread, just to keep the discussion somewhat grounded and scientific:

We should also keep in mind that this single mutation [SLC24A5] does not make the skin particularly fair, on its own (it occurs at a significant fraction of people who have Mediterranean, dark Mediterranean, dark W Asian, or even dark S Asian/ African skin color). As I have mentioned before, one also needs to distinguish between static skin color and adaptable (over the seasons), and perhaps take into account subtle photochemistry effects on the molecular level that don't express themselves that simply and naively as "skin tone."

Onur said...

Because near easterns obviously have brownish skin I doubt these mutations make a huge effect on skin color. Southern Arabians have basically the same skin color as Syrians even though southern Arabians almost never have these light skin mutations. I am sure there are other factors to pale skin in Europe.

The majority of northern West Asians have white-olive skin rather than brownish skin. Northern West Asians are pretty close to Neolithic migrants to Europe pigmentation-wise. Also the Syrian skin color is certainly not the same as the Arabian skin color, much less the southern Arabian skin color, as the latter are darker on average.

Simon_W said...

It's unlikely that R1b has been in far western Eurasia for a very long time. R1b in western Europe is mostly R1b-S116. This is descended from R1b-L11, which is found at appreciable frequency only in central England, eastern Denmark, northern Poland and eastern Switzerland. Obviously this older variant has a more eastern focus than the derived variant S116. And since R1b appears to be associated with ANE autosomal ancestry, it's unlikely that it has been in place for a very long time, since both Scandianvian and Western European hunter-gatherers had none or at most low levels of this ancestry, and it seems unrealistic that mobile hunter-gatherers in northern-central Europe didn't mix with their northern and western neighbours. Moreover the finding that a hunter-gatherer from western Europe was C6 isn't exactly evidence for a presence of R1b either...

By the way, that reconstruction image could be set well in a Spaghetti Western; with that heavy tan, the beard and the glaring blue eyes... :-D

Hector said...

That there are people who would like to believe MP has an origin in Central Asia reveals more about them than about the origin.

It is simply hilarious when you take a look at their next-of-kin and their distribution.

M526 obviously happened in southern Eurasia and then not to offend European sensibilities of some of their descendants it had to make an arduous journey to Central Asia and beget MP. And then M had to travel far to SE Asia or Oceania to beget dark skinned, frizzy haired people...
Isn't that funny?

The presence C6 whose next of kin is C1(Japanese) obviously reflects some degree of East Eurasian affinity of pre-Neolithic Europeans as proven already by various other means.

C6 has C1 C5 C2 C3 as "next of kin" in the order of closeness.
The position of C4 is yet unknown.
There is also a particular C* which seems the most "ancestral"(lacking even M216). And he is Southeast Asian.(Malay?)

Mark D said...

As the article is behind a pay wall, can someone verify what the authors mean by carrying "ancestral alleles" in what I assume are, as Dienekes mentions them, the SLC24A5 and SLC24A2 genes. SNPedia has this regarding the SNP rs1426654(A)on the SLC24A5 gene:

"It appears as if this SNP is a relatively new one in human evolution; one estimate [PMID 17182896] is that the rs1426654(A) allele, in other words, light skin pigmentation, spread through the European population around 6,000 - 12,000 years ago. Prior to that, "European ancestors" were most likely relatively brown-skinned. Another study ([PMID 24048645OA-icon.png]) has concluded that almost individuals carrying the A111T variant can trace ancestry back to a single person who most likely lived at least 10,000 years ago."

Is the "ancestral alleles" A,G or G,G rather than A,A?

SNpedia's chart on allele frequency shows Europeans as having practically no A,G or G,G.

The answer begs the question, did those with the A,G or G,G alleles simply die off and were replaced by, as "barackobama" suggests, Near Easterners, or did the SNP originate with this individual's descendants, or someone similar, who spread throughout Europe?


Eastern View said...
This comment has been removed by the author.
postneo said...

SLC4A5 is probably a persian gulf or ananatolian mutation. Otherwise its spread in south asia cannot be explained. the so called west eurasian lineages in south asia well differentiated from european branches. The only common links lie in west asia.

terryt said...

"This study is actually a wee bit obsolete. We changed the ISOGG C tree a few days ago to show that V20 and the existing C1 and C5 groups are actually within a new C1 group".

You've made a few other very interesting changes as well. A whole heap of new haplogroups. We have a new clade C3e1b, with subgroups C3e1b1 and C3e1b2. Where were they found? There is also a subgroup of the old C5 in the form of C1b2, with one hell of a tail (over 100 mutations). Again, where was that found? And 'Japanese' C1a1 has been subdivided into C1a1a1 and C1a1a2. Both still within Japan/Ryukyu?

"You mean C reached Burma, Thailand, Vietnam via where?"

C in those regions is mostly C3 and so it entered from the north, possibly as late as the Early Neolithic. Even the C* in the region is very unlikely to be at all closely related to 'South Asian' C1b-M356, as Hector said:

"C6 has C1 C5 C2 C3 as 'next of kin' in the order of closeness.
The position of C4 is yet unknown.
There is also a particular C* which seems the most 'ancestral' (lacking even M216). And he is Southeast Asian.(Malay?)"

I don't know that you could safely call C3 'more closely related to C1'5'6' than it is to any of the other C haplogroups, but we have: C1'5'6 in Pakistan/Iran, C3 in East Asia, C* in the Malay peninsula/South China, C2 in Wallacea and C4 in Australia. That hardly demonstrates a South Asian route to Australia.

"M526 obviously happened in southern Eurasia and then not to offend European sensibilities of some of their descendants it had to make an arduous journey to Central Asia and beget MP. And then M had to travel far to SE Asia or Oceania to beget dark skinned, frizzy haired people...
Isn't that funny?"

And obviously wrong.

"The presence C6 whose next of kin is C1(Japanese) obviously reflects some degree of East Eurasian affinity of pre-Neolithic Europeans as proven already by various other means".

But C6 is almost certainly not part of that connection at all. Both C6 (now C1a2) and C1 (now C1a1) split as a combined haplogroup from 'South Asian' C5 (now C1b). The old C6 therefore need not have spent any time at all anywhere near East Asia. C1 and C6 represent opposite ends of the spread of a C5-derived haplogroup. That means this is wrong:

"The C6 may be the ultimate reason why pre-Neolithic Europeans are Amerind-shifted".

European C6 has nothing at all to do with anyone anywhere near America. You are still viewing Y-DNA C as a single haplogroup, which is not the case at all. That is the same as saying all other non-African Y-DNAs are a single haplogroup because they all derive from F.

German Dziebel said...

@Eastern View

"Central Asian wave 40-35K ago:
Aurignacian: (NRY: C6; mtDNA ?)

Near Eastern wave 30K ago:
Gravettian (NRY I; mtDNA ?)

The C6 may be the ultimate reason why pre-Neolithic Europeans are Amerind-shifted."

Agree. The Amerindian component in Mal'ta, albeit smaller, is an older one and goes all the way down to UP, while the "Western Eurasian" component (BLUE) represents Gravettian influence.

Grey said...

Anyone know which of the (seven apparently) skin lightening genes Native Americans have and how they compare with La Brana?

#

"In order to keep in mind also the other half of the story, both La Braña individuals have mtDNA U5b2c1!"

heh, too busy fighting wars over y dna

Grey said...

"The C6 may be the ultimate reason why pre-Neolithic Europeans are Amerind-shifted."

Two ways to Siberia: north first, east second or east first and north second. The first is shorter.

Tobus said...

@German: "Since Mal'ta is the closest to Amerindians (Raghavan et al, Lazaridis et al.), this finding confirms the "easternmost" (Amerindian-like) affinities of European hunters-gatherers".

Nice psuedo-logic there, but you failed to note that only 30% of Mal'ta is shared with modern Amerindians, the other 70% is shared with modern Europeans. Contrary to your conclusion, Extended Data Figure 5 shows no increased affinity between La Brana and Amerindians - affinity with Mal'ta doesn't necessarily mean affinity with Karitiana.


@Terry: It's not just SLC24A5, La Brana also has the ancestral alleles of SLC45A2 and TYR as well. There may well be other undiscovered factors to European light skin, but based on what we know about the genetics of skin pigmentation at this point in time we'd expect him to have dark skin. He has the same genetic skin pigmentation profile that we see in Australian Aboriginals, Papuans and Sri Lankans.

AWood said...

C6 has 18 descendants living today that we know of. I2-M423 has several dozen in western Europe today, but he was alive and well in Belgium 7,000 years ago. The traditional "Neolithic" burials were overwhelmingly G2a3 and we know they were violent, expansionists with new technology from the east, and still account for less than 10% of the YDNA lines in central Europe, and less further west and north. R1b and R1a don't fit into this picture at all, and could have only entered Europe after the Neolithic period in Europe.

mooreisbetter said...

@averagejoe

"Eurocentric" was a bad choice of words. I should have coined the term "R1bcentric."

Forums (fora) are filled with armchair scientists and armchair genealogists, who tested themselves and their families and dabble in theories on boards like this.

DNA testing is largely an American and British phenomenon. So many populations (from French to Chinese) are grossly underrepresented in our data.

Because of the prevalence of R1b in the UK, and because so many Americans are descended from Brits and Irish, huge numbers of Americans who pay to get tested and then grace the world with their theories are of R1b heritage.

Thus, we see a whole lot of overt or covert "pride" let's just say.

We hear nutty stuff about R1b:

--Everyone who is R1b is the descendant of a rugged, manly, horse-riding Indo-European. No scientist would ever say that. On all of these boards though, you find lots of people willing to imply it or state something similar, in shorthand, outright.

--You see people stretching to make R1b the answer to every linguistic and historical phenomenon. I recall a few years back some guy who was German descended was kind of bummed to test R1b, but he quickly theorized that his particular clade just *must* be a signal of a Germanic tribe, and not just any Germanic tribe, but the Teutons and Cimbri who threatened the Roman Republic in the first century BCE.

It's just horse pucky. And it creeps into so much of the online discourse.

Here, I only pointed out that R1b/IE adherents claim that R1 entered Europe through the "middle", then with wave theory, spread east and west, with the "purest" and most prevalent strains being farthest away from the center.

I pointed out that the same applies to C.

But:

-if C is Paleolithic or Mesolithic

-and if La Brana is closely related to Mal'ta

-and if the Wave Theory people like that theory.

Well then, it could mean that the same applies to R1, right? Sorry to destroy anyone's fantasies of horseriding conquerors.

Fanty said...

@Simon:
"By the way, that reconstruction image could be set well in a Spaghetti Western; with that heavy tan, the beard and the glaring blue eyes... :-D"

I almost think I know what you are thinking about.

A character called "Keoma", who is suposed to be a half blood (half European, half Amerindian) played by a blue eyed Italian....

Ok, the handband and the heavy tan are meant to ...uhm... let him apear Amerindian admixed, right? :-D

http://images3.cinema.de/imedia/1591/2451591,1xQNTpdIzXgGblAWsL0gEAUWseGmAcQ0S6oHWr_Ngcdh_f%2BTjbZV41hWsbub%2BKmqj92KlnytUMFuYdVnwD7_Wg%3D%3D.jpg

or

http://gralharocka.com.br/wp-content/uploads/2013/03/keoma.jpg

Lathdrinor said...

I have the feeling a lot is going to change in the coming days with respect to how we trace haplogroup movements and expansions. I remember a recent presentation from Mike Hammer in which he endorsed the view that R ancestors migrated Out of Africa, into Southeast Asia, then backtracked to South Asia, migrated to the Near East, spread into Central Asia and Siberia, before finally settling in Europe. This bizarre trajectory does explain why R is on one side of the world, NO is on the other, Q and P in between, and the ancestral haplotypes of K and M all the way in Southeast Asia and Oceania, but what sort of factors could have possibly driven these people on such a sojourn? The choice of Southeast Asia is specifically mind boggling, given that we have no evidence that Southeast Asia was uniquely formative / conducive to any sort of world changing life style. Why did the ancestors of the main body of European and East Asian ancestors choose there and what, after they got there, propelled them to expand west and take over the world?

The current model requires a lot of detail sketching and explaining, and till these are provided, little is provable. Presently speaking, and given how fast this field moves, I rather believe we got it all wrong than that we got it all correct. That's not because an Out of Southeast Asia theory is fundamentally repugnant. On the contrary, it ought to have a humbling and unifying effect on modern peoples. But there is no way that this genetics model is going to be accepted by archaeological, anthropological, and other circles till the supplementary work is done.

eurologist said...

"Near Eastern wave 30K ago:
Gravettian (NRY I; mtDNA ?)"


Eastern View,

That can't be correct by itself, since early Gravettian is by archaeological finds clearly associated with the Ukraine, Russia, and Siberia (among others, Venus statuettes). So I'd say y-DNA haplogroups P --> Q and R - another "Central Asian" wave (or, more correctly, couple of waves), i.e., originating from a northward flow out of the NW subcontinent and west of the Himalayas.

However, the expansion of P and its subclades in the NW subcontinent might have driven there-existing F-sons (G, IJ) out and toward W Asia, with G(2a?) confined to extreme SE Europe, Anatolia, and immediate surroundings, while IJ made it further into Europe proper (both perhaps bringing mtDNA H with them to S Europe).

Your thought about proto-Aurignacian is interesting, though.

Simon_W said...

I think some commenters here are seriously conflating something. Judging from the PCA in the Lazaridis et al study, La Brana1 and 2 have nothing to do with the ANE shift of modern Europeans! Modern Europeans are placed inbetween ANE and western European hunter-gatherers, to which the La Brana individuals clearly belong. The ANE shift in modern Europeans was therefore caused by some other influence, presumably a later one and more from the east. That the western European hunter-gatherers are nonetheless relatively close relatives of Mal'ta was also mentioned by the Lazaridis study, and explained with the idea that ancient West Eurasians and ancient North Eurasians stem from a common root.

So speculation about some y-chr R1 being present in mesolithic Iberians is absolutely unfounded. And anyway the idea that the mutation from R1 to R1b arose in Iberia is fantasy, given the absence of old R1b variants in Iberia and western Europe and their presence in southeastern Europe and Asia. Iberian R1b is mostly in S116, a highly derived variant, so you'd have to speculate about S116 or its direct ancestor being present in mesolithic Iberia, which is unrealistic with regards to the current age estimates and the presence of R1b-L11 in central-northern Europe rather than in the west.

Simon_W said...

The reason why western European hunter-gatherers are closer to Mal'ta than to modern Europeans is the substantial Early Farmer admixture in modern Europeans.

bmdriver said...

He looks Indian. But that wont sit well with the ''white people'' who came from Europe, who are now trying to define everything in terms of color as a way of differentiating themselves from the logical truth, that Caucasians come from migrating Indian tribes.

Grey said...

What would be a quick and easy genetic way to lighten skin in an environment where it was beneficial - various forms of albinism? But maybe not the one with the most negative side-effects?

So question then for early euros: dark or light or both at once?

http://media-cache-ec0.pinimg.com/236x/57/b2/70/57b2708d0d40295de096599768578969.jpg


http://i.imgur.com/Ew5G7OL.jpg

gradually replaced later by genes without UV side-effects?

Rokus said...

Glad the truth about one single mutation once again reveals the nonsense of Kurganist- and Orientalist- claims alike on Indo-European origins:
'The mutations responsible for the blue eye color most likely originate from the neareast area or northwest part of the Black Sea region, where the great agriculture migration to the northern part of Europe took place in the Neolithic periods about 6–10,000 years ago (Cavalli-Sforza et al. 1994).'

Annie Mouse said...

La Brana is 7kya

The Japanese branch of C1 is dated
"11,650 (95% CI 8,460–18,690) years" from Wikipedia

I imagine V20 and V222 C1/C6 are similar in age. This divergence (Japanese/Europe) should give us the approximate date of arrival of C in Europe.

Rokus said...

'the affinity of the Mal’ta individual and the La Braña 1 individual is not due to some ancient DNA attraction'
Don't the authors refer to a lack of admixture here? As such, the Amerindian affinity of La Brana is essentially different from the (admixed) Scandinavian Motala samples.

'this finding confirms the "easternmost" (Amerindian-like) affinities of European hunters-gatherers. Needless to say, La Brana's mtDNA allocation within hg C supports this, too.'
This finding suggests that Mal'ta was preceded by a closely related group higher up in the YDNA CF tree, but whose influence may be rather related to 'Japanese' Jomon and the Upper Cave remains in China. Unfortunately, C1 was never attested among Native Americans. Native American YDNA C3 apparently represents the East Asian component of NAtive Americans.

'The patterns for C and MP look to be completely different from each other.'
At least one more ancient YDNA lineage but MP achieved a wider Eurasian West-East diffusion combined with a clear cut North-South differentiation, ie. YDNA C1 encompassing the C-K29 lineages that link (southern) European C6 and Japanese C-M8 on the northern hemisphere, against Indian C5 on the southern hemisphere - all recently unified in a single C1-branch by OSIGG (Aboriginal C4 may be part of this family?). Since YDNA C-K29 has a closer affinity with 'southern Europe', this may confirm 'basal Eurasian' as an older expansion marker than 'basal West Eurasian', and still both basal lineages appear related and partially congruent in their expansion. Some dual polarity in time of modern non-African populations? The other YDNA C lineages center further east, hence the hypothesized origin of C* and C3 somewhere between South Asia and Southern China is strongly suggestive for an ever deeper, or ultimate, origin of basal Eurasian in a more eastern direction.

Rokus said...

'This individual was autosomally very very close to modern populations bearing R-derived haplogroups, including your precious R1b.'
We still don't know from where R1a and R1b entered (the rest of) Europe, except that it was present in Late Neolithic northern Europe. Its expansion caused most other YDNA lineage to become extinct or rare. The YDNA lineages of Iberia where no exception to this phenomenon.
What affinity? It's a pity the study doesn't mention the affinity with Loschbour, that certainly predated the success of R1.

'Remember, Spain is a very big place. It is perhaps the soundest and most prudent to simply assume that this is one sample on a big land mass'
One sample of YDNA that became virtually extinct because of Bell Beaker-related expansions. Bell-Beaker also expanded considerably outside Iberia, without indication this favoured the distribution of YDNA C6 in any way. The logical conclusion remains that R1b does not originate in Iberia, it must have arrived with immigrant groups, probably strongly related with northern Beaker people such as Swifterbant, TRB or Corded Ware, whose culture evolved in situ into Iberian Bell Beaker.

Grey said...

1) La Brana has IRF4

http://en.wikipedia.org/wiki/IRF4

"This gene is strongly associated with pigmentation: sensitivity of skin to sun exposure, freckles, blue eyes, and brown hair color."

De-pigmented Indian?

http://media-cache-ec0.pinimg.com/236x/57/b2/70/57b2708d0d40295de096599768578969.jpg

2) More than one route to light eyes imo.

3) "explain why R is on one side of the world, NO is on the other, Q and P in between"

In and out of America chasing mammoth.

Hamar Fox said...

bmdriver,

He looks Indian.

He looks like the modern descendants of mesolithic Europeans, but with dark skin. If Gerhard Schroeder had dark skin, he'd look like La Brana 1. In fact, he looks like quite a few Germans I know of. I suppose Faelid is the type I'm thinking of.

Eastern View,

The C6 may be the ultimate reason why pre-Neolithic Europeans are Amerind-shifted.

But it is MA-1 (R)and AG (Q) who are markedly 'Amerind-shifted', not La Brana 1. Nor does La Brana 1 have any particular relationship (relative to modern Europeans) with Eastern populations, even those high in C, according to the paper. Finally, there's nothing in the distribution of C or C6 that would explain why Amerindians in particular would be glued to pre-Neolithic Europeans.

At K3, where there is only African, Western, and Eastern poles, whatever limits ADMIXTURE has to infer population history (i.e. Kalash) does not exist, so populations should break down into their basic ingredients. If Mal'ta and pre-Neolithic Europeans were fully Western, like Near Eastern (i.e. Semites), they shouldn't be intermediate in component analysis.

I've no desire to reignite the same tired discussion, but it's important for uninitiated readers to know that the above statement is spurious.

Hector said...

The problem with West Asia - Central Asia dispersal of C1 C5 C6 is that their next of kin is C2 and C3 forms an entirely different clade of its own.

(C3(C2(C5(C1,C6)))) is the most current phylogeny. C2's relationship has not been accepted by ISOGG because only a single mutation has been found for this cladistics but I am pretty confident that it is genuine(ie. not a laboratory error).

Anyway I think C6 is a good candidate for the mysterious C*(xC3) that represented a single non R1a remain found in Bronz age Siberia.

terryt said...

"Two ways to Siberia: north first, east second or east first and north second. The first is shorter".

Yes.

"@Terry: It's not just SLC24A5, La Brana also has the ancestral alleles of SLC45A2 and TYR as well".

I'm staying clear of the 'whiteness gene' discussion and so your comment must be directed at someone else. I agree with Eurologist that it is not as simple as a single gene.

"I remember a recent presentation from Mike Hammer in which he endorsed the view that R ancestors migrated Out of Africa, into Southeast Asia, then backtracked to South Asia, migrated to the Near East, spread into Central Asia and Siberia, before finally settling in Europe".

I came to that conclusion when I first saw the McDonald maps way back in 2006. Needless to say I agree with Mike Hammer.

"This bizarre trajectory does explain why R is on one side of the world, NO is on the other, Q and P in between, and the ancestral haplotypes of K and M all the way in Southeast Asia and Oceania, but what sort of factors could have possibly driven these people on such a sojourn?"

My guess at the time was the development of the (improved?) boating technology necessary to cross Wallace's Line. I had endless arguments with Maju on the subject. So much so that he has now banned me.

"That's not because an Out of Southeast Asia theory is fundamentally repugnant".

I easily accepted the idea because I came to it through my study of Polynesian origins. They certainly were accomplished sailors.

"YDNA C1 encompassing the C-K29 lineages that link (southern) European C6 and Japanese C-M8 on the northern hemisphere, against Indian C5 on the southern hemisphere"

I don't think C5 could be called specifically 'southern hemisphere', certainly not south of the Equator. I understand it is confined to just the northwest of South Asia and even spills over into Arabia. Ebizur will be able to enlighten us on that score.

"all recently unified in a single C1-branch by OSIGG (Aboriginal C4 may be part of this family?)".

I'd actually be surprised if C4 was part of that new C1 collection although I'd assume C4 will be shown to be closely connected to Wallacean C2.

"Since YDNA C-K29 has a closer affinity with 'southern Europe', this may confirm 'basal Eurasian' as an older expansion marker than 'basal West Eurasian', and still both basal lineages appear related and partially congruent in their expansion. Some dual polarity in time of modern non-African populations?"

But a deeper connection to MP? I think not although C1-K29 may have emerged from northeast South Asia together with P. But P's deeper ancestry probably lies in SE Asia as per the Mike Hammer idea mentioned above. I doubt very much that C1-K29 was ever anywhere near SE Asia even though it has distant relations there, as well as further north in East Asia.

Average Joe said...

He looks Indian. But that wont sit well with the ''white people'' who came from Europe, who are now trying to define everything in terms of color as a way of differentiating themselves from the logical truth, that Caucasians come from migrating Indian tribes

But he is not similar to Indians genetically which is the main point.

Average Joe said...

But what evidence is there that R1b is Paleolithic or Mesolithic? As far as I can see there is no evidence that R1b was in Europe before Bell Beaker time.

German Dziebel said...

@Tobus

"Nice psuedo-logic there, but you failed to note that only 30% of Mal'ta is shared with modern Amerindians, the other 70% is shared with modern Europeans. Contrary to your conclusion, Extended Data Figure 5 shows no increased affinity between La Brana and Amerindians - affinity with Mal'ta doesn't necessarily mean affinity with Karitiana."

You are again trying to use any loophole to push your agenda. It's not gonna fly. I understand logic is not your forte. Neither is data mastery. Fig. 1 in Raghavan clearly established that Amerindians are closer to Mal'ta than any other Eurasian population. Now that la Brana has turned out to be closer to Mal'ta than to any other West Eurasian population, this means it's closer to Amerindians. There's nothing special about Mal'ta's "west eurasianness" that La Brana wouldn't have "found" in Europe. The only reason it's closer to Mal'ta, therefore, is due to Mal'ta's Amerindianness. The fact that PCA don't readily reflect this means that the earlier Amerindian component in La Brana has been diluted by subsequent admixtures that made La Brana look more West Eurasian and pulled it away from Amerindians. The same happened to Mal'ta but just to a lesser degree because Mal'ta is older than La Brana and closer to Amerindians geographically. So the 30% of Amerindian "blood" in Mal'ta represents common descent of which a portion is also shared by La Brana, while 70% of West Eurasian "blood" - later admixture.

@Rokus

"This finding suggests that Mal'ta was preceded by a closely related group higher up in the YDNA CF tree, but whose influence may be rather related to 'Japanese' Jomon and the Upper Cave remains in China. Unfortunately, C1 was never attested among Native Americans. Native American YDNA C3 apparently represents the East Asian component of NAtive Americans."

I doubt that the phylogeny of CF is accurate. We know from ancient DNA (Mal'ta) that there were no East Asians in Siberia when Amerindians were there. hence, it's likely that Siberian C3 represents a later backflow from America. The absence of East Asian hgs N and O in America supports this.

terryt said...

"(C3(C2(C5(C1,C6)))) is the most current phylogeny. C2's relationship has not been accepted by ISOGG because only a single mutation has been found for this cladistics but I am pretty confident that it is genuine(ie. not a laboratory error)".

It's possible but you're missing C4. Obviously the five haplogroups (C1, C3, C*, C2 and C4) have some sort of relationship. It is surely unlikely that C3 lies at the base though. In fact the haplogroups may have separated in the order I listed.

"The problem with West Asia - Central Asia dispersal of C1 C5 C6 is that their next of kin is C2 and C3 forms an entirely different clade of its own".

But the order above places C1's origin firmly in West Asia - Central Asia while the other C ancestors spread on east to to Eastern Asia (C3) then down Eastern Eurasia to the Malay Peninsula (C*) and to Wallacea (C2) and Australia (C4).

"explain why R is on one side of the world, NO is on the other, Q and P in between"

I don't think the correct explanation is, 'In and out of America chasing mammoth'. The original question ignores MNOPS. As a result we have M and S in between P and NO. That most likely means NO moved north and P moved west and then north, separately, while M and S remained behind. Although R and Q probably enetered the steppe together there is no reason at all why the two should remain evenly spread across it.

"As far as I can see there is no evidence that R1b was in Europe before Bell Beaker time".

That is looking more and more likely.

Annie Mouse said...

From the new phylogeny:

C1a1 (P122 aka M8 etc)= Japanese
C1a2 (V20 aka C6 etc) = Europe

C1a (C-CTS11043) No men known with this exact haplogroup. Central Asia seems most probable however given the progeny.

C1b (M356 aka C5 which appears to be synonymous with C-F1370 in the ISOGG tree)= Middle East/India/Nepal/China

https://sites.google.com/site/haplogroupcproject/c-cts11043

https://sites.google.com/site/haplogroupcproject/c5


C1, is the parent group (C-F3393)

In the description of the group there is a strong hint (and that is all it is) that they think C4 is also a subgroup of C1.

https://sites.google.com/site/haplogroupcproject/c-f3393

At this stage the C1 parent looks Central Asian to me, with the east/west flow becoming C1a and the a southern flow becoming C1b.

How the Australian aboriginal C1s fit in however could change everything. My guess is they will be a branch of C1b.

Daniel Szelkey said...

Where was R1b1b2/or Pre R1b1b2 before the bell beaker time? We need an ancient sample with at least R1b to have a good idea. Modern Distributions are poor indicators, or haplogroups in ancient times at times C6 in ancient Spain!), and are useful in other times, but I am not even going to imagine where R1b was 7000kya until we have 7kya old samples of R1b,

Rokus said...

'But a deeper connection to MP?'
It would be indeed in line with my thoughts if 'greater C1' turns out to share more SNP with F than eg. C3 does - if this is what you mean. with northern Eurasian populations spawning F subclades over a longer period, including YDNA M and, why not, L, H, T before that. Instead, northen Eurasian contact, low effective population size and rapid replacements would have boosted the currency of more derived branches that now happen to be most numerous.

But no, I don't believe a derived branch like YDNA M has an ancient origin within the limits of their ancient habitat. Eg. the relative genetic proximity of Papuan with Sardinian indicates there is no need for such an assumption.

Kurti said...

If a Haplogroup C individual has Caucasian facial features, what does this say about the whole "Asian" Haplogroup range?

Kurti said...

Also what does this say about Scythian samples from Central Asia which showed some individual cases of C and was associated with East Asian admixture? What if the Scythian brought this Haplogroup to East Asia?

Kurti said...

If I am not wrong, even before this study most scientist put the origin of C in the Middle East. To be more specific Southeast Iran.

http://upload.wikimedia.org/wikipedia/commons/c/c0/C%3DM130-Migration.jpg

Doesn't this once again tell us a "Gedrosia" origin of H&G?

Tobus said...

@Grey: Anyone know which of the (seven apparently) skin lightening genes Native Americans have and how they compare with La Brana?

The are 6 genes used for skin colour prediction in the 8plex test and only two of these are common in Amerindian (and East Asian) populations - rs885479 in MC1R and rs1545397 in OCA2. La Brana has the ancestral versions at both locations, as we'd expect for any European population.


@terry: My bad, I meant @eurologist.

SB said...

@terryt
C5 is the most prominent haplogroup among south Indian tribals, who are considered one of the oldest populations in India. Hence it is spread all over India geberously among all groups.

Eastern View said...
This comment has been removed by the author.
Hamar Fox said...

Eastern View,

Sigh. Even if you take out Mal'ta, Native Americans, Mesolithic Europeans would still show Asian affinities while Near Easterners do not.

Fallacy number one: Conflation of points. You specifically referred to the affinity of pre-Neolithic Europeans and Amerindians, not pre-Neolithic Europeans and East Eurasians. C does not explain why Amerindians are markedly closer to Western Eurasians than are other E. Eurasians. In short, you changed your point.

Fallacy number two: Blatant disregard for the findings of the paper(s) discussed. It's apparent from the f3 statistics in the extended figures that La Brana 1 essentially plots along the same vector as modern Europeans (HG and Neolithic admixed) and Middle Eastern populations relative to Sardinians on one axis and Han on the other (extended data 5, figure b). The other two recent papers you're referring to did not make any definitive mention of modern Middle Eastern or Near Eastern populations (although the latter certainly have ANE, so you're already caught out there). You're obviously referring to 'basal Eurasian' here, but you're stating your own interpretation of 'basal Eurasian' and its global distribution as fact, which, considering you're not very bright, is not a great idea.

If you take out Asians, Mesolithic Europeans were still show ASI and Oceanian affinities while Near Easterners do not. Again, my "lies" are straight-forwardly stated in the supplementary material.

Fallacies number three and number four: Cherry picking and selective assent (along with a repeat of fallacies number one and number two). If you paid any attention to this paper, they comprehensively ruled out any particular affinity of LB1 with populations rich in C, while these populations (Australasians, East Asians etc.) had a strong affinity with each other. This strongly suggests, as Dienekes notes, that while most variations of C are East Eurasian-specific, at its very root it seems to be pan-Eurasian and, based on the statistics included in this paper, it does not, as it appears in La Brana, seem to correlate with heightened East Eurasian affinity.

Again, I know you're referring again to basal vs. West Eurasian, and saying C is the root of the distinction. But 1) this is irrelevant to the Amerindian issue, and you therefore irrelevantly introduced this new point, and 2) this current paper does not support your notions of the distribution of 'basal European' within West Eurasia.

Also, as for the selective assent bit. If you're so bowled over, as you seem to be, by these recent researchers' discovery of basal Eurasian and their absolutely stellar ADMIXTURE analyses, then you'll of course also accept the conclusion of these geniuses (as I'm sure tyou must find them) that Amerindians are an admixed population.

Eastern View said...
This comment has been removed by the author.
Tobus said...

@German:Now that la Brana has turned out to be closer to Mal'ta than to any other West Eurasian population, this means it's closer to Amerindians.

Firstly, La Brana is not "closer to Mal'ta than to any other West Eurasian population"... what made you think that? Perhaps you meant "closer to Mal'ta than any other West Eurasian population is"? That statement would be true, but expected since La Brana is 7,000 years closer to Mal'ta to begin with.

Look at the PCA plots, La Brana is smack bang in the middle of the Europeans when compared to Karitiana (a). On the plots with a La Brana axis(c, e) Amerindians show the same affinity to La Brana as South Asian and Middle Eastern groups do. There is simply no evidence that supports the idea that La Brana is closer to Amerindians than modern Europeans.

It seems your logic is flawed, I'm guessing because you've assumed "A is like B, B is a bit like C therefore A must be a bit like C too". This just doesn't hold up in a situations as variable as the human genome - populations can show affinity with each other in different parts of the genome, so X and Y can both share DNA with sample A, yet not share any with each other.

Truthteller said...

Yes, C5 or C1b is found in South Indian tribals, however, its frequency is the higest in West India (Gujarat area). 14 Hapmap Gujaratis were found to have the P92 mutation (meaning C5a). How many Hapmap Gujaratis were lower castes? I am willing to bet none. I am willing to bet that it is found at a high rate in Rajasthan as well.
Now, 15% of Paniyas were found to be M356 as well. The thing though is, it is found at a very low frequncy in other tribals of South India.

Kurti said...

@Eastern View. If you mean me. I didn't said a word about C1 but C* and it's not something sudden but was speculated by scientist even earlier. somewhere between Southeast Iran and West Pakistan.

And the closest to the subclade Cv20 to which this individual belongs is the Jomon C m8. Aren't the Jomon the people who were once thought to have Caucasian like facial features? So what if these Caucasian facial features of Jomon people are not coincidence but real Caucasian "admixture"!

However C is one of the oldest Haplogroups of this world so it's hard to connect one specific phenotype to such an old and diverse Haplogroup. Still the Jomon connection is very suspicious.

German Dziebel said...

@Tobus

"Firstly, La Brana is not "closer to Mal'ta than to any other West Eurasian population"... what made you think that? Perhaps you meant "closer to Mal'ta than any other West Eurasian population is"?"

I think the data bears out both statements to the effect that LB and MA-1 share an affinity with each other, which is stronger than either of them has toward any other West Eurasian population (save for 2 - Orcadian and Russian). See pp. 17-18 of Suppl Mat plus the tables referenced on those pages.

As we discussed before, PCA is not as good of a method as D statistics. You've ignored the explanation of PCA plots I gave above. Please review it again.

"It seems your logic is flawed, I'm guessing because you've assumed "A is like B, B is a bit like C therefore A must be a bit like C too". This just doesn't hold up in a situations as variable as the human genome - populations can show affinity with each other in different parts of the genome, so X and Y can both share DNA with sample A, yet not share any with each other."

This is just absurd. If genomewide, MA-1 is close to Amerindians and LB is close to MA-1, it means LB is close to Amerindians. They are close to each other to a varying degree depending on time and geographic proximity to the source but they are close to each other due to common descent. On top of that primary relationship (supported by Y-DNA C6) there's a layer of later waves of admixture.

German Dziebel said...

@Hamar Fox

"Also, as for the selective assent bit. If you're so bowled over, as you seem to be, by these recent researchers' discovery of basal Eurasian and their absolutely stellar ADMIXTURE analyses, then you'll of course also accept the conclusion of these geniuses (as I'm sure tyou must find them) that Amerindians are an admixed population."

What in the world is this? All facts point to Amerindians being an unadmixed population. Your "geniuses" haven't provided a single bit of evidence to the contrary.

"Amerindians are markedly closer to Western Eurasians than are other E. Eurasians."

You just provided evidence that Amerindians are an unadmixed population. If they were admixed, only fringe and derived Western Eurasian populations showed affinity to all of Amerindians, not all of West Eurasians.

mooreisbetter said...

Who besides me thinks it's time for the self-proclaimed powers that be to come up with a more workable NRY tree?

When groups split by as much time and geography as the new C1a1 and C1a2 get such deceptively close classifications, you realize just how arbitrary the whole thing was. Especially at which downstream SNP to assign a new letter.

If redoing the tree, it would be better served to do it based on major lineages by raw numbers of males bearing the lineage, or by TMRCA.

terryt said...

@German: "I doubt that the phylogeny of CF is accurate".

Of course you doubt it. Accepting it would mean abandoning your whole belief system.

@Eastern View: "And C4 (putative C1b2) is found in Melanesia, with untested C2 in Australia".

The other way round actually.

"Since there are no indigenous C3 in South Asia or further east in southeast Asia, C3 likely entered Siberia via Central Asia".

I agree with that. In fact I strongly suspect C2 and C4 developed from a C who had moved through Siberia before moving south to Australia.

"How the Australian aboriginal C1s fit in however could change everything. My guess is they will be a branch of C1b".

The Australian Aborigines are C4-M347, not C1-F3393. My guess is that C4 will be a branch of C2. And I doubt the two will be part of C1, but will be a separate group, just as C3 is.

"At this stage the C1 parent looks Central Asian to me, with the east/west flow becoming C1a and the a southern flow becoming C1b".

There is no reason why C1b should not be the one that remained behind when C1a took off north though. Does anyone know where the new clade C1b2-Z-to-infinity was found? Its lineage looks to have been reduced to virtually a single family more than once. It may be the lone survivor from the home region. The other offspring all wandered off when ecology allowed it.

"Where was R1b1b2/or Pre R1b1b2 before the bell beaker time?"

According to ISOGG there is no such thing as R1b1b2. Just R1b1b-M335. I thing European R1b is all R1b1a, in fact R1b1a2a-M269. Is that correct? Hasn't R1b1a2a been found in Bashkirs at the southern end of the Urals?

"with northern Eurasian populations spawning F subclades over a longer period, including YDNA M and, why not, L, H, T before that".

Those lineages have a more South Asian/Middle East distribution, don't they?

"I don't believe a derived branch like YDNA M has an ancient origin within the limits of their ancient habitat. Eg. the relative genetic proximity of Papuan with Sardinian indicates there is no need for such an assumption".

Don't we know that haplogroups and aDNA are not always intimately connected? We can be fairly sure that both M and S lived for some time in Sundaland before they were able to venture across Wallace's Line. The same goes for the Australian/New Guinea M haplogroups (M14, M15, M29'Q, M27 and M28). Therefore I think it extremely likely that both Y-DNAs M and N originated in Sundaland. We also know that a couple of mt-DNA haplogroups are shared between Palawan and South Asia (M19'53 and M24'41) and one even between Australia and South Asia (M42). The direction of movement is as likely to have been from east to west as it is from west to east. If MP originated in Sundaland and P's descendants spread through much of Eurasia that would explain the Sardinian/Papuan connection, along with many other connections.

"C5 is the most prominent haplogroup among south Indian tribals, who are considered one of the oldest populations in India. Hence it is spread all over India geberously among all groups".

Have you a reference on that? As far as I'm aware it is primarily northwestern but with a presence in Nepal. Its presence in parts of Southern India is the result of farming's spread from the north.

"Yes, C5 or C1b is found in South Indian tribals, however, its frequency is the higest in West India (Gujarat area). 14 Hapmap Gujaratis were found to have the P92 mutation (meaning C5a). How many Hapmap Gujaratis were lower castes? I am willing to bet none. I am willing to bet that it is found at a high rate in Rajasthan as well.
Now, 15% of Paniyas were found to be M356 as well. The thing though is, it is found at a very low frequncy in other tribals of South India".

Thanks Truthteller.

Hector said...

After reviewing some of the comments here I came to the conclusion that most of them are political agenda not scientific speculations.

Quite funny that there are people like Klyosov who claimed that F derived Y haplogroup as a whole represents "Caucasoid" elements and then now it is claimed that C is also "not necessarily East Eurasian".

This is about territorial ambition and attempts to justify conquests and colonizations.(ie. Amerindians were Caucasoid in the male line and that the land was rightly taken over by "brothers".
Siberia all the way to Japan was originally Caucasoid taken over by overly fecund sallow little people of Neo East Asia so on so forth.)

I don't know whether this post will get through the mods but it should be mentioned that a C2 individual was recently tested positive for K29, which Terry seems to have difficulty accepting.

Grey said...

Random theories

.

Sequence

1) out of tropics

adaptive pause

(Khoisan looking)

2) out of Africa

(coastal)

adaptive pause (multiple regions)

3) out of India

pause

4) etc

.

IRF4: associated with dark hair, light eyes, freckles

I don't think the 8plex system used for La Brana and Louschbour takes into account the combinations only commonly found in Ireland and Scotland i.e. the places furthest from the spread of SLC24A5 et al.

I think both La Brana and Louschbour were likely dark haired and light eyed with pale but very freckled skin as a fast adaptation for northern latitudes of out of India looking dudes.

.

mammoth steppe: conveyor belt between Europe and Siberia (and back again).

.

Into America chasing mammoth
P -> bottleneck -> Q

Out of America chasing mammoth
Q -> bottleneck -> R

Grey said...

@Glossy
"I'm curious, did these brown-skinned, blue-eyed hunter-gatherers have modern European facial features?"

Narrow forehead
High cheekbones
Brow ridge

For example

http://www.imdb.com/media/rm3089345536/nm0614165?ref_=nm_ov_ph

(but with a ton more freckles)

Hamar Fox said...

Hector,

After reviewing some of the comments here I came to the conclusion that most of them are political agenda not scientific speculations.

Yes, but you are obviously one of those people.

German Dziebel,

What in the world is this? All facts point to Amerindians being an unadmixed population. Your "geniuses" haven't provided a single bit of evidence to the contrary.

While I have absolutely no animosity toward you, I simply don't have the energy to talk in circles about this anymore. We have three key pieces of research in the last couple of months alone that have each rigorously tested the relationships of modern and ancient samples. In a thread from about June or July last year, terryt and I both argued a position that strikingly foreshadowed the publication of these three papers, and I'm fairly sure that neither of us knew they were in the pipeline. Accurate predictions of future scientific findings are the best indication that someone is analysing data as objectively as possible.

This is a majorly important topic to you, but for me, it's the psychological dimension of the fact that people simply can't accept they were wrong that keeps me active in the debate. I've little tolerance for ideology -trolls like Eastern View, but you'll note that I leave you alone, because I don't believe you're trolling at all. But I do believe that you are wrong.



German Dziebel said...

@Grey

"1) out of tropics

adaptive pause

(Khoisan looking)"

FYI: Khoisan looking skulls appear in the craniological record only in the Holocene. The Hofmeyr skull at 36,000 YBP clustered with UP Eurasians.

"Out of America chasing mammoth
Q -> bottleneck -> R"

Makes sense. And Y-DNA hg C was subject to similar "extinction" in both America and Europe.

@terryT

"Of course you doubt it. Accepting it would mean abandoning your whole belief system."

I don't have a belief system. Out-of-America has started from the pretty systematic patterns observed in linguistics, kinship systems, comparative folklore and ethnology. Now with more genomewide and ancient DNA data this perspective receives greater and greater support. mtDNA and Y-DNA phylogenies have been built having untested assumptions about "African roots" and the "New World" in mind. And these assumptions were heavily fertilized with Adam and Eve imagery.

Annie Mouse said...

"I'm curious, did these brown-skinned, blue-eyed hunter-gatherers have modern European facial features?"

We the bald guy in the middle looks just like my local Butcher, so yeah.

Tobus said...

@German: EDF 5 isn't PCA plots, it's outgroup f3 statistics, and I think you agree that there is no increased La Brana/Amerindian affinity shown there. I understand and agree with the reasoning behind your theory of expecting such affinity due to the increased Mal'ta affinity, but since it's not the only possibility and is not supported by the data, it's a theory that can be safely abandoned as untrue. La Brana does not have increased affinity to Americans, even if there's a good reason to expect he might.

@Grey:
Loschbour has pretty much the same pigmentation genetics as La Brana, dark skin, blue/light eyes. Stuttgart has one and a half of the light skin alleles so was probably a bit lighter. SLC24A5 is fixed in Ireland and Scotland, but I agree the 8 plex system isn't particularly comprehensive.

Chad Rohlfsen said...

Why can people not accept a European that has not always been pale with blonde or red hair? Do people not understand anything about mutations, positive/negative, and sexual selection? I do not understand the big deal.

The change to light skin was sped up by the mixing with EEF, who was already lightened. Do people not understand what a dominant and recessive trait is either?

Let's accept things for what they are, and find links to the reasons for extra lightening and the genes in Europeans. Let's not make ridiculous claims based on modern phenotypes to explain the past. We see how well this crackpot science worked for the R1b ice age refuge. We have actual physical data here and some refuse to accept it. If you don't like what science has to say, then go enjoy things like religion.

Chad Rohlfsen said...

Sorry to double post. When I say that mixing with EEF sped it up, I mean the lightening that was probably due to farming. Just incase some do not understand that everyone was mixed with EEF in Europe, and we don't have remnants of these people.

What would be a good date to have this mixing complete, as in no full-blood Mesolithic? Around 2500BCE in Northern Europe? Maybe sooner?

One interesting thing that I have noticed is that NW Europeans(Brits and Scandinavians) share the same reflective quality of skin and lactose tolerance percentages. Whether or not the two are related, I am not sure. However Vitamin D does help absorb calcium. Perhaps lactose tolerance required even lighter skin. Perhaps it's just coincidence.

Chad Rohlfsen said...

Sorry for a third post here. I am wondering if they checked the hair texture alleles. If they have a complexion in the Aborigine / Papuan range, do they have hair more of a straight to wavy or wavy to kinky hair?

Tobus said...

@German: I took some time to go over p17 and associated tables of the supplement again with your arguments in mind, but I can't see anything saying that La Brana is closer to Mal'ta than he is to Europeans. You appear to have fallen victim to a false assumption (self-proclaimed superior logistical mastery notwithstanding) that if A is closer to B than to C, then B must be closer to A than C. Consider that your head is closer to your knee than to your foot, does that mean that your knee is closer to your head than to your foot?

I understand how you could be confused because of the affinity of La Brana and Mal'ta, but your conclusion was wrong - La Brana is closer to modern Europeans than he is to Mal'ta.

terryt said...

"However C is one of the oldest Haplogroups of this world so it's hard to connect one specific phenotype to such an old and diverse Haplogroup. Still the Jomon connection is very suspicious".

Agree absolutely. In fact many used to see 'Caucasian' features in Australian Aborigines, another population with a high proportion of Y-DNA C. Very suspicious.

"If genomewide, MA-1 is close to Amerindians and LB is close to MA-1, it means LB is close to Amerindians".

As Tobus keeps trying to explain your logic is flawed. If Amerindians and LB share a different range of similarity with MA-1 then Amerindians and LB can be very different from each other. Virtually all human geographic variations are a product of a mix of long periods of isolation and admixture. That is so even for your beloved Amerindians.

"On top of that primary relationship (supported by Y-DNA C6)"

What close relationship does European C6 have to Amerindian C3? Answer: almost none. Just marginally more than do any F Y-DNAs.

"When groups split by as much time and geography as the new C1a1 and C1a2 get such deceptively close classifications, you realize just how arbitrary the whole thing was".

Hardly 'arbitrary'. From the classification we know that C1a1 and C1a2 are more closely related to each other than either is to C1b. And these are more closely related to each other than they are to any other C Y-DNA, and so on. I have just been informed that Y-DNA C is almost twice as old as Y-DNA G, which is an ancient derivative of F. That means that although the C1a1/C1a2 split was ancient it is still much more recent than the C/F split. The current nomenclature allows us at a glance to see relationships. The problem with C is that it lacks the research that Y-DNAs R, Q and O have had and so historically has been given just one letter to cover all, whereas F-Derived haplogroups have been split into a multitude of letter names. Accident of history, or simple twist of fate.

"it would be better served to do it based on major lineages by raw numbers of males bearing the lineage, or by TMRCA".

Except that the 'raw numbers of males bearing the lineage' is also the product of a simple twist of fate, depending on what has been sampled, and the estimated TMRCA is very far from being an accurate figure.

"it should be mentioned that a C2 individual was recently tested positive for K29, which Terry seems to have difficulty accepting".

I understand the test is from a single individual and is unreliable. In fact Ray Banks (associated with ISOGG) has told me, 'Within the old major C groups, there has been no sequencing of C2 (Pacific islanders) and C4 (Aborigines)'. Although he adds, 'We have been able to get a few individual tests for a rare C2 sample'. We await further testing on C2 individuals.

terryt said...

"it should be mentioned that a C2 individual was recently tested positive for K29, which Terry seems to have difficulty accepting".

Well. Have a look at this:

https://sites.google.com/site/haplogroupcproject/experimental-c-tree

Seems C-M38 is as likely to be a clade within C-M217 as it is to be related to C-K29.

terryt said...

"it should be mentioned that a C2 individual was recently tested positive for K29, which Terry seems to have difficulty accepting".

I've rechecked and I see C-M38 is in no way connected to C-M217.

Annie Mouse said...

"There is no reason why C1b should not be the one that remained behind when C1a took off north though. "

Also very possible.

Annie Mouse said...

So far as I can tell from the "hint" no one has tested M347 men for F3393.

According to ISOGGs SNP index"
F3393=23023974 C->A
M347 = 2877479 A->G

So at this stage anything is possible.
Oooh the excitement!

Kristiina said...

It is true that we-are-not-them and they-are-not-us attitudes abound. To me, the discovery of C in Neolithic Europe confirms again that most y lines are global. The big expansive branches, C, DE and K, all extend well beyond the racial boundaries from east to west. It is a pleasure to discover that people have been moving here and there. In my opinion, it is somehow absurd to link an individual ancient y line with the looks of an individual modern group of people. The modern looks must be a result of a long line of ancient and more recent admixtures.

However, I find it interesting that mtDNA lines do not show the same expansion patterns as y lines. MtDNA lines seem to conform better to our racial boundaries. On mitochondrial side, only N and R levels seem to be Pan-Eurasian.

eurologist said...

@Eurologist: The earliest Gravettian is in Romania at 32,000 years close to the Straight of Bosporus facing Turkey, with Balkan precedents. By the time Gravettian gets to Russia at 30,000, the Russian Plain already had 15,000 years of Upper Paleolithic. By the time Gravettian influence shows in Mal'ta, Siberia already had 20,000 years of Upper Paleolithic. So no, Mal'ta is not a good representative of ancient Siberians.

Eastern View,

Seems like you are chasing ghosts.

Mal'ta evidently is a late representative of Gravettian, and clearly not a representative of early UP Siberians, nor Aurignacian Siberians. As is evidenced by him being neither C nor P nor Q.

AP said...

Re: "As far as I'm aware it is primarily northwestern"

In the complete Sengupta sample set of "176 samples representing 8 Pakistani populations" it was found in "one Dravidian Brahui in Pakistan" or 0.57% http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1380230/table/TB5/

C-M356 is completely missing from Haber's "total of 204 Afghan samples."

SB said...

@terryt
@truthteller

Y-DNA C is found in the Paniya,Paliyan, Kadar, Irula (who are all south indian tribals) and also with low frequency in Thoda and Kattunaickan

http://www.plosone.org/article/fetchSingleRepresentation.action?uri=info:doi/10.1371/journal.pone.0050269.s005

German Dziebel said...

@Tobus

"I can't see anything saying that La Brana is closer to Mal'ta than he is to Europeans."

I don't think we can interpret the D statistics provided on those pages in any other way. D statistics measures the mutual affinity between two populations compared to a third one. I can't wrap my mind around the logic whereby MA-1 is closer to LB than to other Europeans but LB is closer to other Europeans than to MA-1. The only possible way in which you and I can be both right is if the affinity between MA-1 and LB is older than the affinity between LB and other Europeans. This is what I suggested above and you better accept it if you want to maintain some semblance of intellectual integrity in my eyes.

For comparison, see Razib's observation (http://www.unz.com/gnxp/the-palearctic-conveyor-belt/) that LB is closer to Asians than any other European population. Although Razib is a science freak who I banned from anthropology and then Discover fired him, his interpretation that LB shows "paleo-Siberian ancestry which can be found in many West Eurasians and Native Americans" is stronger than yours. And it comes close to my interpretation.

@Hamar Fox

"But I do believe that you are wrong."

Hamar, what you believe in is irrelevant. Just show me the data that Amerindians are an admixed population, while MA-1 is not. Then we can talk. Your hifalutin prose is getting on my nerves. Sorry. And affiliation with TerryT who believes that humans came from Antarctica doesn't do you any good. ;)

Grey said...

@Chad
"Why can people not accept a European that has not always been pale with blonde or red hair?"

Straw man. It's not a question of whether de-pigmentation happened but when.

.

@Tobus
"Stuttgart has one and a half of the light skin alleles so was probably a bit lighter. SLC24A5 is fixed in Ireland and Scotland"

If SLC24A5 is 100% fixed in Europeans then you're just assuming that the underlying color was brown and these alleles lightened it.

If the underlying color was de-pigmented with freckles then those "lightening" alleles wouldn't have done anything except cover the freckles.

Given the light eyes the assumption that they were de-pigmented with freckles is more valid than the assumption they were brown.

cos

1) That's how the Neanderthals did it.

2) It's a quicker fix to the UV problem.

3) All the later records talk of red hair, light eyes (grey or green not blue) and pale skin i.e.

de-pigmentation

not lightening of dark skin but browning of de-pigmented skin with freckles.

(I doubt if SLC24A5 is 100% fixed. People are just assuming again on the basis they think people without it would be brown when in fact they'd be 50% white with 50% brown freckles. I expect someone like that will be found soon enough as apart from anything else i expect they have a lot of skin problems so there's probably a study somewhere related to skin problems and people with a lot of freckles.)

.

@Kristina
"To me, the discovery of C in Neolithic Europe confirms again that most y lines are global. The big expansive branches, C, DE and K, all extend well beyond the racial boundaries from east to west. It is a pleasure to discover that people have been moving here and there"

It doesn't show that most y lines are global. It shows that the early expansions of HG groups from various "out of..." events mostly displaced the male population of previous expansions.

The y lines get less global as time goes on as populations get more adapted to their niche and/or increased population density makes it harder to completely displace a population.

.

"However, I find it interesting that mtDNA lines do not show the same expansion patterns as y lines."

The obvious explanation for that is the females weren't killed.

Hector said...

"Yes, but you are obviously one of those people"

Even if it were true, that would not mean you are not one of them.
In fact you are a quintessential example.

The test on the C2 man was funded by Korean amateurs and they have the first hand info. Ray Banks is very slow to accept more recent results. In fact he still has M188 as a downstream of M7 even after hundreds already tested the other way around. Even in his "experimental tree".(This is O3 tree)

The update on the C2 man is that he tested negative for CTS11043.
This means that so far the relationship between C2 C5 and C1+C6 is unknown inside K29(or F3393) But Fudan university thinks along the phylogeny I already presented. This was back in August last year before we had results from the C2 man at FTDNA.

So I trust most of what Fudan University says. They also complained about ISOGG and its relative tardiness in accepting Chinese results. I am no fan of Chinese but I think they have a point.

What makes me trust Fudan University more is that they too don't know about the status of C4 as they have no sample.

terryt said...

"In a thread from about June or July last year, terryt and I both argued a position that strikingly foreshadowed the publication of these three papers, and I'm fairly sure that neither of us knew they were in the pipeline. Accurate predictions of future scientific findings are the best indication that someone is analysing data as objectively as possible".

Correct. An here's another one in response to:

"it should be mentioned that a C2 individual was recently tested positive for K29, which Terry seems to have difficulty accepting".

I realised last night that it would make perfect sense that C2-M38 would have the K29 mutation. In fact it is amazing that such a small amount of information would reveal so much about prehistory, and here's why. Basal C is obviously C-M130(xK29). C2 and C4 having K29 would leave just C3 (East Asia) and C* (centred on the Malay Peninsula) with that basal form. We can be virtually certain that C2-M38 formed in the triangle of islands of Sumba, Timor and Flores/Alor. And any population reaching those islands must have come via the Malay Peninsula. Humans were open-water fishing in Timor by 42,000 years ago and had reached Australia by 46,000 years ago. It wouild be most parsimonious to assume the K29 mutation formed within a C* population who had entered Southern Wallacea from the nearby Malay Peninsula. The limited geographic extent of the triangle would mean drift would lead rapidly to a single haplogroup's survival ther. The presence or absence of K29 in Australia would therefore depend on wether K29 had formed before or after that continet was reached.

So what about Eurasian K29? Did the mutation occur twice? We know that a smattering of C* reached eastern South Asia. And it turns out that C1b is not mainly confined to Gujarat. The mewly identified C1b2 is primarily from Bangladesh, the Ganges/Brahmaputra Delta. Exactly the region where a people pre-adapted to an island habitat would feel most at home. Gujarat is not exactly at the Ganges headwaters but perhaps near enough.

Did someone mention MNOPS? No? Well I will. S and M in New Guinea Melanesia, but not in Australia. A sort of MNOPS in the form of K1-P60 is though. Anf K2-P79 is found in Melanesia. And K3-P261 is from Bali.

We await with extreme interest the resolving of K29 and the C2 haplogroup.

Truthteller said...

@SB,
And C5 is found in 14 Gujarati samples as well:
http://biorxiv.org/content/biorxiv/early/2013/11/22/000802.1.full.pdf

How many individuals did they analyze? Let's assume 80. So 14/80 or 17.5% of Gujaratis have this haplogroup. This is probably the highest frequency in all of India.
C5 in South Indian tribals can easily be explained by the fact that the Dravidian language migrated from the Northwest to the South. The Dravidian language is the language of migrating Northwest or West Indians.

Truthteller said...

By the way, C-M356 has also been found in Uygur samples, and some other central asian populations.
C-M356 in Nepal can easily be explained by the fact that the rumor has always been that some of the populations there came from Rajasthan. It is found in Tharus and Newars as well.

The fact that it has been found in a Brahui (the Harappa database shows that Brahuis are even more Western Eurasian than the Baloch) tells me that this is a Dravidian (or neolithic farmer) haplogroup. It is found in South Indian tribals due to the fact that dravidian speakers were forced to migrate South (where they intermingled with hunter-gatherer populations).

Gary Moore said...

Eurologist wrote:

"Mal'ta evidently is a late representative of Gravettian, and clearly not a representative of early UP Siberians, nor Aurignacian Siberians. As is evidenced by him being neither C nor P nor Q."

There is no evidence associating the Gravettian culture with YHG R1a. Because R1a and R1b are both closely related to Q, there is nothing to exclude R1a from being a core component of early Siberian populations.

In a related note - some remarks have been made in this thread regarding "Eurocentric bias" and this is fully evident in the discussion of the Mal'ta findings. The popular press made much of the MA-1's YHG R1a, loudly crowing that it was proof that Native Americans are derived from Europeans - for example, one headline in the popular press proclaimed: "First Americans were Europeans". Technically, it's not really true: MA-1 finding only indicate that Europeans and Native Americans share common ancestry from a population that lived in Siberia 20,000 years ago. There is no proof from ancient DNA that any contemporary Europeans had R1a or R1b at all. The truth is that R1a and R1b are recent arrivals to Europe from the east - probably as a result of migrations out of central Asia or Siberia, and the languages these populations spoke are related to modern North American languages, albeit with a lot of borrowed words and features.

Hamar Fox said...

Hector,

Even if it were true, that would not mean you are not one of them.
In fact you are a quintessential example.


Which explains why I've made numerous accurate predictions and raised questions about data that ultimately was revised in precisely the way I anticipated it would be. And I'm not entirely sure how supporting the obviously correct conclusions of these three papers makes me 'politically biased'. Did these researchers deliberately develop statistical measures to generate conclusions they felt would justify the course of history? Perhaps if I entered every thread like a bitter little man with an axe to grind, as you do, your case would have more clout.

German Dziebel,

Hamar, what you believe in is irrelevant.

I knew my politeness would come back to bite me. I was saying, in the least frictious way possible, that I don't want to discuss the same issues over and over again with you.

Your hifalutin prose is getting on my nerves.

So don't read it. Anyway, it's how I roll. I'd comment on your own prose, but it's hard, because I don't read your posts.

terryt,

In fact many used to see 'Caucasian' features in Australian Aborigines, another population with a high proportion of Y-DNA C. Very suspicious.

In fact, I saw Australoids referred to as 'archaic Caucasoids' a few weeks back (only on Wikipedia, mind). It's interesting to consider if their phenotype contributed to the formation of Caucasoid features. C is likely 'in' every Eurasian population, even if it's been phased out/become invisible in some modern Eurasian populations. La Brana had it, but he isn't closer to Australasians than modern European groups, which suggests to me that their ancestors, Neolithic and/or Mesolithic, had C also at some point. The main snag with the idea of Australasians, as they exist now, being ancestral to Western Eurasians is that there's simply no autosomal sign that this is the case. They're much more related to East Asians and not significantly closer to W. Eurasians (inclusing La Brana) than is any other East Eurasian population.

It's something to keep our eyes on, though. Perhaps Australasians represent in phenotype and Y-DNA a population from which W. Eurasians derived, even though they are not equivalent to this population.



Hector said...

First of all it was an error when I said there was a C* man who is M130+ M216-. It is M216+ M130-. That it is the most "ancestral" form known so far is not changed.

As for the MP clade, if by luck ordained by DNA God, we found one of the SNPs determining MP clade earlier than 92R7 or M45 we would have called the MP clade "P". And it would have been noted that there are a lot of mysterious P* in Melanesia and Papua NG. The general vincinity of South East Asia would have been an unchallenged home of this clade.

No one would have speculated that the whole P* clade mysteriously and comically family-transfered from Central Asia to Southeast Asia whole sale.

As for Hamar Fox et.al. they don't seem to be able to read certain statistical graphs correctly. f3 statitstics is not a PCA. Yet they talk about "right in the middle of" etc. I am sure they will deny it in some fashion but I am certain that they are mathematical ingenues.

Anyway, Finns for instance are nearly 100 percent European autosomally. Yet a minute part of their autosomes along with their Y chromosome profile faintly but definitively illustrates their distant ancestry. La Brana will of course be close to Europeans. It is obvious and there is nothing very interesting there. It is other anomalies that are interesting and informative.

Grey said...

"The popular press made much of the MA-1's YHG R1a, loudly crowing that it was proof that Native Americans are derived from Europeans"

I may have got it wrong but isn't MA1 autosomally closest to Europeans as well?

If so the interesting question then is how did the y dna get changed somewhere along the line.

AP said...

"...17.5% of Gujaratis have this haplogroup. This is probably the highest frequency in all of India."

That would be for that particular sample-set not for all of Gujarat. The sampling was done from Gujaratis living in Houston: NA20845 to NA21145.

Within this sample set there is close cluster and these have some of the highest %age of components variously labelled South Asian or South Indian in the subcontinent.

In Magoon's analysis:
"M356 subgroups, concentrated in the 14 P92+ Gujarati samples" I think includes 12 GIH samples (NA20889 to NA21133) and one Telugu HG03867 and HG03850.

All of these M356+ Gujarati samples are part of that 'South Indian' type close clustering group. None of other GIH samples outside this close cluster has M356.

So it does not look like a coincidence that one of the most "South Indian" groups in India, such as Gujaratis-A and Tharu have M356.

HAP K=12
Gujaratis-A 82 13 1 0 0 3 0 0 0 0 0 0
Gujaratis-B 62 24 4 1 0 7 1 1 1 0 0 0

Tobus said...

@German:
I don't think we can interpret the D statistics provided on those pages in any other way. D statistics measures the mutual affinity between two populations compared to a third one

...where the two populations differ!. You made this same mistake in our last extended discussion: D(Out, A, B, C) can tell us absolutely nothing about how close B and C are to each other because it only looks at the SNPs that are different between them. What this D-stat tells us is whether A/B affinity is greater than A/C affinity - it's intentionally excluding the B/C affinity and makes no measurement of it. The B/C affinity could be 99% yet still show a significant D-statistic because the 1% that is different is all that is considered.

If we had a D(Out, La Brana, Mal'ta, European) this tell us who LB is closer to, but we don't have one, so these D-stats are no good for answering the question we're asking. The f3 stats are more than sufficient though, please take a look at them again and you'll see that they show La Brana is closer to Europeans than to Mal'ta.


can't wrap my mind around the logic whereby MA-1 is closer to LB than to other Europeans but LB is closer to other Europeans than to MA-1.

Seriously? It's incredibly simple.

Try this:

1. On a piece of paper draw a line that is 24cm long. Label one end "E" and the other "MA-1"
2. At 7cm in from the "E" label, make a mark and label it "LB"
3. Confirm the following statements are both true:
3a) "MA-1" is closer to "LB" than it is to "E" (17cm/24cm)
3b) "LB" is closer to "E" than it is to "MA-1" (7cm/17cm)
4. Face-palm yourself for failing to see something so obvious!

You will no doubt have noticed that I used 24cm and 7cm as these are the ages of the samples and so the line you've just drawn accurately represents distance between the samples in time? In terms of genetic makeup, if there were no admixture events this is roughly what we'd expect to see due to drift, but this is not part of my argument so please don't bother contesting it. The point I'm making is that your basic logic is flawed - "A is closer to B than to C" does not necessarily mean "B is closer to A than C". Since the f3 statistics show the opposite of what your false logic lead you to assume, the only logical conclusion is that your assumption is wrong.... La Brana is in fact closer to Europeans than he is to Mal'ta.


@Grey: If SLC24A5 is 100% fixed in Europeans then you're just assuming that the underlying color was brown and these alleles lightened it.

I suggest you read some scientific papers on the topic - SLC24A5 (as well as SLC45A2 and TYR) has been conclusively associated with light skin multiple times in multiple admixed populations (African-Americans, South Asians and Melanesians). There is simply no doubt that each ancestral allele you have will lighten your skin to a noticable degree - it's a fact, not an assumption. The assumption is to what degree, not if it has an effect. Given what we know at this point in time, La Brana and Loschbour share the same skin pigmentation alleles as Australian Aboriginals, Sri Lankans and Papuans. We may undercover futherer contributors in the future, but at present we have to accept the possibility that Mesolithic inhabitants in Europe had very dark skin.

Freckle/tanning genes like EFR1 and MCR1 variants are also found in dark-skin populations - freckling is not a "white" person phenomenon and neither is tanning, although both are obviously more visible on depigmented skin.

terryt said...

"MA-1 finding only indicate that Europeans and Native Americans share common ancestry from a population that lived in Siberia 20,000 years ago".

That's correct. MA-1 shares 'part' of his ancestry with Native Americans, not all of it. That is what German cannot get his head around.

"And affiliation with TerryT who believes that humans came from Antarctica doesn't do you any good. ;)"

German. I did not claim humans came from Antarctic. I used that idea to show how simple it was to prove anything you wished to if were were prepared to adopt your usual method of ignoring generally-accpeted data. In fact I was surprised how easy it was to 'prove' humans originated in Antarctica if one was prepared to ignore generally-accepted dates for the breakup of Gondwanaland and for the deforestation of Antarctica. For those interested in seeing how German's method of 'proving' beliefs works here it is again. Most of the phrases are only slightly adapted from ones German has used:

"It is often claimed human expansion through America was from south to north. We know that Australia/New Guinea and South America were the last continents to break contact with Antarctica, and Gondwanaland. Taken at face value that easily explains both the increased linguistic diversity and the similar kinship systems found in the two regions. The absolute chronology is a big issue, no doubt. I know the currently accepted dates for Gondwanaland's breakup do not support this scenario but the ancient dating is a mess, and has been promoted by scientists whose prejudice against Antarctica as a source for at least part of Old World diversity stems from a 500-year-old cultural stereotype. The African haplotypes that many mistake as ancestral to modern humans are likely the product of archaic admixture in Africa. Africa had broken away from Gondwanaland some time before Australia and South America had, carrying only 'primitive' and 'archaic' humans. These human types were able to move towards modernity only once the Antarctic humans had entered Laurasia and so were eventually able to reach Africa. Scientists have failed to find evidence for this Antarctic origin because most of them are especially prone to holding uncritically accepted beliefs. They have closed minds, and are ignoring the possibility and so do not look for the evidence that must be there. As for the intense cold: we know that beech forest existed on Antarctica in the past. The dating usually given is 2 million years ago, but this date too is probably wrong".

The above actually fits with almost all of German's claims.

eurologist said...

Gary,

Late Gravettian Mal'ta boy's y-DNA was a derived R* (R3, so to speak) - not R1a. So, with the deduction of the related P and Q previously in the general region from their current Siberian and American distribution, it is reasonable to associate R with the Gravettian at least in Siberia, if not also in NE Europe.

Hamar Fox said...

Dziebel, I actually bothered to waste some time reviewing your prose (which I found unimpressive) and stumbed upon this comedy gem:

I can't wrap my mind around the logic whereby MA-1 is closer to LB than to other Europeans but LB is closer to other Europeans than to MA-1.

You honestly can't work it out?

X>Y>>>>>Z

Y (La Brana) is closer to X (modern Europeans) than to Z (MA-1). Z (MA-1) is closer to Y (La Brana) than to X (modern Europeans).

Grey said...

@Tobus
"I suggest you read some scientific papers on the topic - SLC24A5 (as well as SLC45A2 and TYR) has been conclusively associated with light skin multiple times in multiple admixed populations"

Yes, I understand that adding skin-lightening alleles to brown-skinned people does exactly what you say.

But what about people like this - would it make them lighter?

http://realhistoryww.com/world_history/ancient/Misc/Common/India/Indian_Albinos/The_Bhatti.jpg



terryt said...

"MA-1 finding only indicate that Europeans and Native Americans share common ancestry from a population that lived in Siberia 20,000 years ago".

That is correct. And both Europeans and Native Americans have become admixed with other populations since.

"The main snag with the idea of Australasians, as they exist now, being ancestral to Western Eurasians is that there's simply no autosomal sign that this is the case".

Your use of the term 'Australasians' actually includes three quite distinct populations: Australian Aborigines (mainly Y-DNA C4 and mt-DNA N including R), Papuans (Mainly Y-DNA MNOPS and mt-DNA M) and Austronesians (Mainly Y-DNA C2 and mt-DNA B4a). And to say 'there's simply no autosomal sign' is wrong. Western Eurasians are almost completely Y-DNA MNOPS and mt-DNA R.

"The general vincinity of South East Asia would have been an unchallenged home of this clade".

Yes. Take note German.

"the interesting question then is how did the y dna get changed somewhere along the line".

The Y-DNA didn't change from one haplogroup to another. The most obvious explanation is that Q, R and C all made it onto the steppe and moved around, different haplogroups coming to dominate different regions as time passed.

terryt said...

"First of all it was an error when I said there was a C* man who is M130+ M216-. It is M216+ M130-. That it is the most 'ancestral' form known so far is not changed".

Interesting, but it doesn't really alter anything, just refines things. We already knew that SE Asian C was very varied, probably because selection was less in that region than it was in the north (C3) and across Wallacea (C2,C4). The finding shows the M130 mutation is more recent than the M216 one though. More work on the Malay/SE Asian C* would tell us a great deal.

Hamar Fox said...

Hector,

As for Hamar Fox et.al. they don't seem to be able to read certain statistical graphs correctly. f3 statitstics is not a PCA. Yet they talk about "right in the middle of" etc. I am sure they will deny it in some fashion but I am certain that they are mathematical ingenues.

So you make an inaccurate statement and then predict that I'll point out that it is inaccurate? I'm certain that you are several things beginning with C, but let's stick to the facts please.

Anyway, Finns for instance are nearly 100 percent European autosomally. Yet a minute part of their autosomes along with their Y chromosome profile faintly but definitively illustrates their distant ancestry. La Brana will of course be close to Europeans. It is obvious and there is nothing very interesting there. It is other anomalies that are interesting and informative.

Finns are around 7-10% East Asian. Lithuanians would obviously have been a smarter choice to make your case with. It's possible that a haplogroup can be unrepresentative of a population's autosomal make-up, but I know of no case of it not corresponding at all. The authors of this study obviously thought the same, hence the statistical tests they included. Since you've not had a discussion with me on the subject, and clearly haven't tracked my position on the subject, your strawmen are about as worthless as can be.

German Dziebel said...


@Tobus

"Seriously? It's incredibly simple."

I don't understand when our wires got crossed (I must have misspelled something again or you misdirected me and I didn't catch it early on), but the issue at hand is the special proximity between Amerindians and MA-1 (Lazaridis, Raghavan) and then between MA-1 and LB (Olalde). IMO, this of necessity means proximity between LB and Amerindians. We should not be talking about LB, MA-1 and modern Europeans.

@Hamar

"Y (La Brana) is closer to X (modern Europeans) than to Z (MA-1). Z (MA-1) is closer to Y (La Brana) than to X (modern Europeans)."

See above. There was a mistake upstream that derailed the whole conversation. Next time instead of seeking comedic fun, read the whole string and help people by identifying mistakes.

"I knew my politeness would come back to bite me. "

If you continue to claim, like a broken record, that Amerindians is an admixed population, while MA-1 is a pure one without providing any evidence for this claim, I'm going to call you on this. This is because by doing so repeatedly you are being impolite. And don't congratulate yourself on being verbally "polite" if you can't live up to an ethical standard when it comes to a scholarly discussion.

"So don't read it."

That's my "second nature" as an anthropologist. I read people's stuff. And it's not because it's always great. If you don't want to be read, don't post. It's just that simple.

Annie Mouse said...

"In fact many used to see 'Caucasian' features in Australian Aborigines, another population with a high proportion of Y-DNA C. Very suspicious."

I think this is the impact of a later coastal wave carrying south indian phenotypes (caucasoid), perhaps sweeping future Australian C1b/C4 along with the flow.

But it could simply be that the combination of east asian neoteny and Australoid strong facial features roughly simulates European facial features. That is to say, what makes a European stand out from an African is bigger brow ridges, a higher nose bridge and other prominant facial features. The effect of our archaic admixture perhaps? I favour the first explanation personally but it is possible that both effects are contributing to the caucasoid-like appearance.

eurologist said...

at present we have to accept the possibility that Mesolithic inhabitants in Europe had very dark skin.

Tobus,

Nothing could be further from the truth. We know that extant Northern Europeans are much, much lighter in skin color than N Africans, SW Asians, or S Asians that share their SLC24A5 allele.

Ergo, we know that the SLC24A5 allele is not responsible for the light pigmentation of N Europeans.

Central and N Europe receive way too little sunshine for Mesolithic inhabitants to have been of dark or "brown" or even Mediterranean color. As I alluded to elsewhere, this SLC24A5 allele might be particularly effective in seasonally adaptive protection against the sun, as necessary for farmers (field workers), but less so for HGs, who usually stay out of intense sun because they have no business being in it.

Chad Rohlfsen said...

I would bet that once the diet shifted to grains, vitamin d absorption from the sun is what causes change from the Mediterranean to Britain.

We can't use modern North Africans and near easterners as comparisons here. They have different admixture, and I believe a couple fewer skin alleles.

Meat and fish eaters up north don't need light skin. There's plenty of examples out there.

Chad Rohlfsen said...

@eurologist

You think hunter gatherers would be inside more than farmers? You must have never done either. Farming doesn't require many days in the sun to tend crops. Hunters and gatherers are always on the move looking for herds, berries and such.

Trust me, I grew up on a farm and hunt primitive archery 4 to 6 months a year. You can't kill a deer or mammoth from a hut.

I'm not trying to be an ass, just some misconceptions I'd like to correct.

Chad Rohlfsen said...

SLC24A5 is responsible for 25-40% of the skin tone difference of Europeans and West Africans. Not 100% You can't focus on the one gene.

Chad Rohlfsen said...

Sorry, for the 3rd post. I believe the Mesolithic Europeans had the ancestral allele, not the derived European, West Asian one. 93 to 100% of Sub-Saharan Africans, East Asians and Native Americans have this allele. So, they did not have the one that creates the lightening of skin by 25-40%.

They were likely dark skinned. Not a big deal, since they were bred out by the lighter EEF. Other alleles lightened the skin more, later on.

Hamar Fox said...

terryt,

Your use of the term 'Australasians' actually includes three quite distinct populations: Australian Aborigines (mainly Y-DNA C4 and mt-DNA N including R), Papuans (Mainly Y-DNA MNOPS and mt-DNA M) and Austronesians (Mainly Y-DNA C2 and mt-DNA B4a). And to say 'there's simply no autosomal sign' is wrong. Western Eurasians are almost completely Y-DNA MNOPS and mt-DNA R.

I was opposing autosomal and haploid data, meaning overall affinities across the greater portion of the genome. I know there are different populations within the region, but data from these populations is sketchy, unfortunately, so I'm simply going on what little information I'm aware of. This paper compares various modern and ancient Europeans with an ancient aboriginal Australian genome via f3, with no significant values produced. The same was done with Papuans in Raghavan et al.'s paper.

Dziebel,

I don't understand when our wires got crossed (I must have misspelled something again or you misdirected me and I didn't catch it early on), but the issue at hand is the special proximity between Amerindians and MA-1 (Lazaridis, Raghavan) and then between MA-1 and LB (Olalde). IMO, this of necessity means proximity between LB and Amerindians. We should not be talking about LB, MA-1 and modern Europeans.

Even though you're obviously backpedaling from a career-shattering catastrophe of reasoning, you still commit the almost-as-hilarious mistake of not realising that one population being directly related to Amerindians (MA-1) and another related population (WHG) being significantly less related to the same population is better explained by admixture from MA-1 to Amerindians than Amerindian admixture into (or, rather, the complete Amerindian ancestry of) MA-1 and WHG. This is because one relationship is primary and the other only secondary (plus time for differentiation), while you want both to be essentially primary (plus time for differentiation).

See above. There was a mistake upstream that derailed the whole conversation. Next time instead of seeking comedic fun, read the whole string and help people by identifying mistakes.

Right, a 'mistake'. Let's face it, you were confused by a basic logical problem that even Hector could solve. The only way you could lose more credibility would be to fake emails from researchers from a serious lab in a laughable plea to be taken seriously, only to have the same lab embarrassingly release a paper about three weeks later contradicting everything the email said. Let's hope you never do something like that. I'm sure you won't.

If you continue to claim, like a broken record, that Amerindians is an admixed population, while MA-1 is a pure one without providing any evidence for this claim, I'm going to call you on this. This is because by doing so repeatedly you are being impolite. And don't congratulate yourself on being verbally "polite" if you can't live up to an ethical standard when it comes to a scholarly discussion.

I was polite because I genuinely pitied you. You're certainly a comical figure, but I felt bad that I didn't dislike you enough to be laughing at you as hard as I frequently did. It just wasn't right. Now you've annoyed me enough that I can laugh freely, no compunction.

That's my "second nature" as an anthropologist. I read people's stuff. And it's not because it's always great. If you don't want to be read, don't post. It's just that simple.

No, I just don't want to be read by you. It decreases the chances that you might bother me. Your writing style is garbage, btw.

terryt said...

"I think this is the impact of a later coastal wave carrying south indian phenotypes (caucasoid), perhaps sweeping future Australian C1b/C4 along with the flow".

The newly proposed phylogeny for Y-DNA C suggests you have the direction wrong here. The most basal C haplogroups (without the K29 mutation) are all eastern, notably C3 and C*. C* is spread around the South China Sea but apparently a basal branch has been discovered in the Malay peninsula. That would indicate that the mutation happened in a C* population who moved from the Malay Peninsula into Wallacea and Australia. C1 is part of that K29 group and so must also have emerged somewhere in the east.

That is my guess. And archaic admixture offers the best explanation as to why Y-DNA C3 is not 'Caucasoid'.

"the issue at hand is the special proximity between Amerindians and MA-1 (Lazaridis, Raghavan) and then between MA-1 and LB (Olalde). IMO, this of necessity means proximity between LB and Amerindians".

I'll leave everyone else to explain the lack of logic in this A = some proportion of B, and C= some proportion of B, therefore A=C totally argument.

German Dziebel said...

@Hamar Fox

"Even though you're obviously backpedaling from a career-shattering catastrophe of reasoning"

No, I'm not. Just read the beginning of the string. I can assure you that I'm interested in MA-1, LB and Amerindians and not in MA-1, LB and Western Europeans. I regret that the moment you decided to put your best rhetorical foot forward, you fell of the stage.

"you still commit the almost-as-hilarious mistake of not realising that one population being directly related to Amerindians (MA-1) and another related population (WHG) being significantly less related to the same population is better explained by admixture from MA-1 to Amerindians than Amerindian admixture into (or, rather, the complete Amerindian ancestry of) MA-1 and WHG. This is because one relationship is primary and the other only secondary (plus time for differentiation), while you want both to be essentially primary (plus time for differentiation)."

Is that your "proof" that Amerindians are admixed? You are in the wrong business, my friend, as you are incapable of scientific reasoning. Not everybody is, so you can easily excel in something else. LB/WHG and MA-1 are related to Amerindians proportionately to their geographic distance from Amerindians and the time elapsed since their split. If Amerindians were derived from MA-1, we wouldn't have seen any pull toward Amerindians in Western Europe. Plus we would've found mtDNA hg U and Y-DNA hg R in them but we don't. Your "primary" and "secondary" relationship is just a made-up play of words. Go to school and learn some proper terminology.

"The only way you could lose more credibility would be to fake emails from researchers from a serious lab in a laughable plea to be taken seriously"

I didn't fake anything. You are just jealous that I have all the degrees, all the big ideas and all the connections I need, while you are a nobody. The only thing you are left with is worshipping people from "respectable labs."

"You're certainly a comical figure."

That's nothing compared to a bunch of out-of-Africa-and-into-the-Americas believers making fool of themselves almost every year now.

"Your writing style is garbage, btw."

For clowns like you my matter-of-fact style will always sound like garbage. Because it leaves them with no room to hide.

Tobus said...

@Grey:
Albinos have a mutation in OCA2 (the gene is actually named for this condition - "Oculocutaneous Albinism II") which means they are unable to synthesis melanin at all. The skin lightening alleles like SLC24A5 affect melanin production and expression so I expect would have no effect at all in albinos.

Tobus said...

@German: Nice backpedalling, but the "misdirection" came from your statement "Now that la Brana has turned out to be closer to Mal'ta than to any other West Eurasian population, this means it's closer to Amerindians", followed up with "I think the data bears out both statements to the effect that LB and MA-1 share an affinity with each other, which is stronger than either of them has toward any other West Eurasian population" and the now classic "I can't wrap my mind around the logic whereby MA-1 is closer to LB than to other Europeans but LB is closer to other Europeans than to MA-1. Since La Brana is definitely closer to modern Europeans than to Mal'ta, your statements and the original conclusion are incorrect. I understand that a public formal retraction is not your style, so I'll settle for you silently accepting the point and not repeating the "LB is closer to MA-1 than to Europeans" nonsense in future.

but the issue at hand is the special proximity between Amerindians and MA-1 (Lazaridis, Raghavan) and then between MA-1 and LB (Olalde). IMO, this of necessity means proximity between LB and Amerindians

As I stated earlier, I agree with your logic and had the same expectation, but that expectation is simply not supported by the data - in fact the f3 statistics fail to show any increased affinity between La Brana and Amerindians at all. So while the logic in this case makes sense, the theory doesn't fit the facts and we should abandon it in favour of one of the many explanations that does fit the facts. One such explanation is that the increased affinity between La Brana and MA-1 is from the 70% "European" part of MA-1 and is expected to be greater than modern Europeans given that La Brana is some 7,000 years closer to MA-1 than modern Europeans. It's also possible that the affinity comes from a common ancestor of both La Brana and Mal'ta but only the Mal'ta branch contributed to the Amerindian gene pool, leaving the LB branch with no special Amerindian affinity over other European populations. Whatever the reason, the fact is that unlike Mal'ta, La Brana doesn't show any increased affinity with Amerindians - he's the same distance to them as modern Europeans are, and any plausible explanation has to fit with this fact.

Tobus said...

@eurologist: Nothing could be further from the truth. We know that extant Northern Europeans are much, much lighter in skin color than N Africans, SW Asians, or S Asians that share their SLC24A5 allele. Ergo, we know that the SLC24A5 allele is not responsible for the light pigmentation of N Europeans.

There have been a number of papers directly associating SLC24A5 (along with SLC45A2 and TYR) with melanin levels in a number of admixed populations with a range of skin pigmentations (such as Stokowski 2007 on South Asians) - there is simply no question that each derived allele of these genes will lighten the carrier's skin colour to some degree. I note that all the populations you refer to as "sharing SLC24A5" show a range of skin colours as well as a range of SLC24A5 frequencies - and you can be 100% sure that those individuals with derived SLC24A5 (and/or SLC45A2/TYR) alleles directly coincide with those individuals exhibiting lighter pigmentation, since that's how the connection between these alleles and skin colour was determined in the first place. Given that each allele acts differently, independently and additively and that any individual can have one of 3 configurations of each of the 3 genes (both dark, both light or one of each), this gives us 27 possible pigmentation phenotypes - Europeans are at one extreme with all derived, Sub-saharan Africans, Sri Lankans and Melanesians are at the other with all ancestral, and populations in North Africa, West Asia, South Asia etc are in between with various combinations of ancestral/derived, typically becoming more derived as you move towards Europe. Most studies attribute the bulk of the African/European melanin difference to these 3 genes however one study (Belaza 2012 on Cape Verdeans) suggested the combined effect was less than half.

It's certainly possible (probable even) that further genes will discovered in the future that will give us a better picture of La Brana's skin colour than we do now, but given what we know at this point we have to accept the possibility (and perhaps I should stress possibility) that La Brana had the same skin colour as modern Papuans and Australian Aborigines - he certainly has the same genetic profile as these populations in terms of skin colour alleles known to science today. If this DNA was from a murder weapon recovered in NY this morning, the police would definitely be looking for a black man, it's only our presupposed notion of Europeans being "white" that makes us question these results in regard to La Brana.

I'm happy to post links to the papers that confirm what I'm saying but you should be able to find most of them with Google/Pubmed - the "Human Skin Color" page on Wikipedia cites most of them in it's Genetics section as well.

eurologist said...

You think hunter gatherers would be inside more than farmers? You must have never done either. Farming doesn't require many days in the sun to tend crops. Hunters and gatherers are always on the move looking for herds, berries and such.

Chad,

Not inside more in general, but not nearly as much out in the full sun, either. Mesolithic Central and Northern Europe was largely covered in dense, old-growth forest, where gathering would have taken place. Hunting in the few available clearings would likely occurred at dawn or dusk, when the animals are there to graze.

Neolithic (and later) farming was extremely laborious: clearing forests/ bush land adjacent to rivers, driving the animals onto the grazing areas, making hey for winter, preparing the fields, and harvesting. Harvesting is in late summer (and when it does not rain), when the sunshine hours are the most in Central Europe.

Sorry - I maintain that early farmers would have spend much, much more time in intense sunlight than HGs.

Of course, I agree that northern cultures who relied largely on fishing, and even more so on hunting sea mammals, are an exception when it comes to vitamin D intake.

Gary Moore said...

On a lighter note, perhaps a little off the thread but still on the general topic of demic succession in Europe, I have a plausible explanation for the collapse of the Early European Farmer populations that does not involve the EEFs being cut down by the sharp bronze of implacable chariot-riding Kurgan hordes. About the time of the disappearance of the EEFs, give or take a few centuries, an Egyptian papyrus recorded a cattle plague in approximately 3,000 BC. This presents a likely scenario in which the immigrants from the Eurasian steppes may have introduced rinderpest to the livestock of the EEF communities when they came in contact with them. The effects of cattle plagues on 19th century Africans is well documented, and they would probably have been worse for the EEFs, who also had to deal with surviving long, cold winters.

Eastern View said...

@Kurti, no, I was referring to one of terryt's comments about C1a2 being West Asian.

@Eurologist: I wouldn't say Mal'ta is Gravettian but rather shows Gravettian influences. Archaeologists also say it has a Central Asian character in its lithics, so I would think the R2 is from Central Asia. in ADMIXTURE we also see this Ancestral South Asian component in Mal'ta, which is lacking in Mesolithic Europeans and Amerinds.

Chad Rohlfsen said...

@eurologist

We would not see the forests we do today, during the Mesolithic. The glaciers were retreating, and we are talking about a polar desert here that was inhabited by mammoths and other giants. Plots of trees would've been few and far between for quite some time. Dark skin would've been nice in a land with today's climate and few trees. Darker skin is also handy for these people hunting in a white-out environment as dark skin will act as eye-black does for a ball-player.

I've read somewhere that blue-eyes help against Seasonal Affective Disorder. I am trying to locate the article to verify the sources. Even Arctic Reindeer change from brown to blue eyes in the winter.

Hunters are always on the move, taking down lodging and moving it. Farmers build a home and stay there. They aren't always putting it up and down to follow animals. They would be busy hoeing the ground, dropping seeds, pulling weeds a couple times, and then harvesting. It is not the daily chore that hunting and gathering had to be.

I would compare their farming to the one acre garden we had as a kid. There was maybe 15-20 days of work during the growing season. They may go out daily to let a few animals graze, but it's not the same. Were Native Americans hiding in teepees and the woods while the Americans were outdoors working all day?

The Neolithic brought a change in society. It was more like every man for himself. It's not the same as the hunting and gathering to keep your tribe alive. These hunting people are out gathering nettles, acorns and such as well. They don't just sit indoors until the meat runs out. As you spoke of, darker skin would be nice if you are sitting over the reflective water for hours fishing. I am sure that was the main staple of their diet. Fish livers (also large game liver) are a great source of Vitamin D to replace the lack of sunlight.

Another point is that you want to move into placement on a herd during mid-day for your dusk hunt. Pre-dawn darkness and mid-day are the best times to move.

Show me any hunter groups today with light skin that do not have post Neolithic West Eurasian ancestry. They are not there.

Hamar Fox said...

Dziebel,

Is that your "proof" that Amerindians are admixed?

It wasn't intended to be proof of anything but your inability to reason soundly. I was merely highlighting that of the two solutions on the table for solving the quandary at hand, you preferred the one with the least explanatory power.

LB/WHG and MA-1 are related to Amerindians proportionately to their geographic distance from Amerindians and the time elapsed since their split.

According to the calculations of a man who doesn't understand even the basics of logic. Obviously there's no simple distance-by-age relationship between WHG, MA1 and Amerindians, and the importance of geography (presumably by closer populations being subject to continual outpourings from America) is unquantifiable and an obvious get out of jail free card. It's also been thoroughly debunked by the simple fact that Amerindians have global relationships that MA-1 does not. So even if you think this is because Amerindians were the initial centre of it all and their progeny split into W. and E. Eurasians, which developed their own way, essentially erasing detectable bonds with the other Eurasian half of the picture, any subsequent admixture from Amerindians into Mal'ta should not be part of that process and hence the global relationships of the most recent admixture from America shouldn't be disguised in MA-1.

If Amerindians were derived from MA-1, we wouldn't have seen any pull toward Amerindians in Western Europe.

You're probably not correctly interpreting whatever source you have to support this. Pull towards Amerindians relative to whom? Tobus has painstakingly explained to you how f3 and f4 statistics work, and if you're talking about PCAs, then Western European Sardinians and Basques are at the W. Eurasian 'apex'. They're not 'Eastern' shifted relative to any other Western Eurasian populations.

Plus we would've found mtDNA hg U and Y-DNA hg R in them but we don't.

Q.

Your "primary" and "secondary" relationship is just a made-up play of words. Go to school and learn some proper terminology.

A direct ancestor or descendant = primary relationship. An uncle, cousin, nephew etc. = secondary relationship. The genes of MA1 and kin are in Amerindians. The genes of La Brana are only related (by some distance) to genes that are in Amerindians. I thought you'd be able to work this out. And I'm actually laughing at the 'get educated' remark, because you're right: I have no formal education in anthropology or anything else that might qualify me as an authority in any of this. But I'm still, way, way smarter and more correct than you. Some people are just born smarter than others. Imagine if I were formally educated in this field. I think I'd melt your brain.

You are just jealous that I have all the degrees, all the big ideas and all the connections I need, while you are a nobody.

You have to be joking. I can see for myself that you are not a particularly exceptional man, and that all those soft subject degrees you have are just so many ways of stifling your insecurity about the fact. If you want to talk about my being a 'nobody', try having almost dead-cert terminal cancer twice at ages 20 and 21, having all future plans dashed, and nearly ten years later bouncing back, healthy(ish) again, and just breezing through a debate like this with a so-called 'accomplished man' who obviously tried so hard through all the years my mind was on other things. That's a life worth talking about. Then I might at least respect your life, if not envy it. How many people want to hear the life story of someone who has two or three degrees?

The only thing you are left with is worshipping people from "respectable labs."

How do I worship them? I respect them over you, but I review their arguments on their own merits.

Chad Rohlfsen said...

I would bet on the excess allele(s) associated with the extra 'whitening' of Europe to originate near the Iron Age. This was the time where farming was made easier by iron plows, sickles, and the development of new grain grinding technology.

This could be the population boom that spread the new allele(s) continent wide.

German Dziebel said...

@Tobus

" Nice backpedalling, but the "misdirection" came from your statement...I understand that a public formal retraction is not your style"

I don't mind admitting that an error came from me. We are juggling different things, corresponding remotely with 24-hour breaks between postings, sorting through multiple posts and having less than user-friendly interfaces. But it's a "human error" that has no bearing on the integrity of my argument. Your argument and your data mastery are still flawed, although looks like you didn't admit any technical errors.

So let me straighten it out. The fact remains that since 1) Mal'ta is closer to La Brana than it is to a) modern West Eurasians and b) East Asians and 2) Mal'ta is closer to Amerindians than it is to West Eurasians and b) East Asians, La Brana and Amerindians share common ancestry. Since Mal'ta and Amerindians do not share the West Eurasians BLUE component and Amerindians don't have mtDNA U and Y-DNA R (Raghavan), it's impossible that Amerindians have West Eurasian ancestry. The opposite must be true: West Eurasians from Mal'ta in the east to La Brana in the west have Amerindian ancestry. This is supported by the presence of the "Amerindian component" in WHG, Motala and Shuttgart in Europe. And by Y-DNA C6 in La Brana. Since La Brana is younger than Mal'ta and further removed geographically from America than Mal'ta (don't forget the geography!), it's more distantly related to Amerindians than Mal'ta. Again, consistently with the rest of ancient DNA samples from Europe. Hence, sometimes the original Amerindian component gets overwhelmed by subsequent waves of admixture, and La Brana looks completely West Eurasian without Amerindian admixture.

Note that Olalde conveniently omitted Amerindians from their tables, hence it's hard to say what the numbers are, but I think the plot discussed by Razib shows that La Brana is pulled eastward away from the rest of West Eurasians.

Now you need to admit your fundamental lapse of logic and ignorance of the data patterns just like I willingly admitted that my writing was the source of the technical error in our discussion. You and Hamar can continue to gloat over my typos but I'm holding all the real cards in this debate.

Annie Mouse said...

"The newly proposed phylogeny for Y-DNA C suggests you have the direction wrong here. The most basal C haplogroups (without the K29 mutation) are all eastern, notably C3 and C*."

I don't have a problem with this. Clearly C1 made it to central Asia, by whatever route. I am suggesting that a C1b population flowed down into India-ish (or developed there). Then when Caucasoids travelled east in a later wave, C1b travelled with it into Australasia.

Maybe C* had already travelled east in the first coastal wave. Maybe C* travelled eastwards via Central asia instead. It does not matter, because I am talking about C1, which we know was somewhere near to central asia in order to split into C1a and C1b.

Grey said...

@Tobus

"The skin lightening alleles like SLC24A5 affect melanin production and expression so I expect would have no effect at all in albinos."

Yes. Now there would be no reason to expect Europeans to already be de-pigmented before the arrival of the farmers except *all* the ancient writers describe the same phenotype: red hair, light skin, grey/green eyes. If that's not discounted - even if just for the sake of argument - then where did the red phenotype come from and why did it mostly disappear except in the region coincidentally furthest from both the farmers and the IE?

Also, what is the percentage of dark skinned, blue-eyed people in India and what is the percentage of mesolithic Euro samples so far supposedly dark-skinned and blue-eyed?

Also Neanderthals.

I think the presumption ought to be they were already de-pigmented - somehow connected to MC1R or IRF4.

.

(The description of grey/green eyes is significant as that's partway to blue so if for example a population like that got an infusion of East Asian skin-lightening genes which added some more de-pigmentation on top of what was already there then red hair and grey/green eyes would turn blond and blue - so red in the isles, blond in the Baltic.)

.

Initially this

http://i.telegraph.co.uk/multimedia/archive/00632/news-graphics-2006-_632931a.jpg

de-pigmentation to this

http://i.imgur.com/Ew5G7OL.jpg

eurologist said...

there is simply no question that each derived allele of these genes will lighten the carrier's skin colour to some degree

Tobus,

Of course - but not in the European context, as I have stated numerous times. There is no way these alleles come even close to being responsible to extant Central or N/ NW European skin color. Likewise, there are populations in NE Asia that rival the lightest in the world and don't harbor those mutations.

It is mind-bogglingly ridiculous to assert that Mesolithic Europeans at the common 50 to 60+ degree (i.e., easily-habitable Northern-most PNW and Alaska-like)latitudes had anything else but light skin color, unless in those few instances their diet was mostly sea mammals.

Simon_W said...

@ Gary Moore

It has to be stressed that the Yamnaya/Pit Grave culture didn't really expand westwards into central Europe in the sense of Poland & Germany. It only advanced into southeastern Europe and the Carpathian Basin. Maybe it was really more R1b than R1a, but hopefully we'll see in the future some ancient y-DNA from the Yamnaya culture.

Yet there is an obvious eastern influence in the core area of the Corded Ware immediately preceding it, and related with regards to pottery: The Globular Amphora culture, which perhaps developped from the eastern group of the TRB.

As Raetzel-Fabian noted in 2002: There was a global climate crisis following heavy volcanic eruptions in 2958 and 2955 BC. It's possible that this gave the Corded Ware Culture with its different ideology and/or economy an edge over the preceding cultures. However, he also noted that the cultural changes leading to the Corded Ware already started to develop about a century earlier.

terryt said...

"It does not matter, because I am talking about C1, which we know was somewhere near to central asia in order to split into C1a and C1b".

But to me it is most likely that C1a entered Central Asia from South Asia rather than the other way round. I'll explain why:

"Then when Caucasoids travelled east in a later wave, C1b travelled with it into Australasia".

Two things wrong with that. 'Australasian' C is not C1b, nor is it C1a, even under the newly proposed phylogeny. Australian C is C4-M347 and Wallacean/Pacific C is C2-M38. Either, or both, may prove to have the K29 mutation, which would make them C1, but they would be neither C1a nor C1b. I think Hector has said the C2 individual examined had the K29 mutation but not the F 3393. That would mean C2-M38 is more basal than either C1a or C1b. Also C3 and C* now appear to be the only C(xK29) haplogroups, and both are eastern. This all means the C1 haplogroup did not travel east to Australia but west from somewhere near that continent.

"Maybe C* travelled eastwards via Central asia instead".

That is how I see it. Eurologist suggested a time of 120,000 years for the first OoA. At that time the climate was as warm as it is today, if not warmer. Later climate cooling would have eliminated haplogroups from large swathes of Central Asia, only for southern survivors to return as technolgy allowing survival was developed. That return was primarily from the western end to begin with.

"I am suggesting that a C1b population flowed down into India-ish (or developed there)".

If C-K29 originated in the east from C(xK29), which now seems extremely likely, it would mean that C1b 'flowed' into South Asia from the east. C1b actually breaks into two clades: C1b1 and C1b2. C1b1 is the old C5 while newly discovered C1b is confined to Bangladesh. As I said elsewhere that is exactly the type of habitat we would expect to find a population newly adapted to an island habitat. The related C1a would therefore have passed through South Asia on its voyage from SE Asia.

"Maybe C* had already travelled east in the first coastal wave".

More likely not. C*'s diversity seems greatest in South China/SE Asia with just a smattering in South Asia, especially near the east coast. And it is difficult to account for the scarcity of C in India if it was the first haplogroup to enter the subcontinent. Any sort of rapid early 'coastal migration' has always seemed most unlikely to me.

Annie Mouse said...

"then where did the red phenotype come from and why did it mostly disappear"

It could partly be nutritional.

Check out Jose Salvador Albarengo, a Mexican fisherman who looks to have a lot of Latino/Spanish ancestry. A largely mediterranean ethnic type. He transitioned from a very dark almost black brunette to a clear redhead on a diet devoid of grains and vegetables (turtles and fish apparently).

http://www.bbc.co.uk/news/world-latin-america-26041727

Gary Moore said...

@ German Dziebel - I am of Native American descent, and I have no doubt we are admixed with East Asians. The $1,200 I had to spend out of pocket to have a root canal procedure repeated because my European-American endodontist did not realize that some people's first lower molar has a third root is proof. Fortunately, I went to a Chinese endodontist the second time, and she recognized the issue immediately. (I also have winged incisors.) Interestingly, although I have typical Native American facial structure in addition to sinodonty and have even been pointed out as an "Indian" by tourists once when visiting a reservation despite my fair coloring, DNA testing failed to detect any East Asian or Native American ancestry. (My nearest Old World YDNA match, BTW, is Volga Tatar. Swedes, OTOH, often identify me as Saami.)

@ Simon_W - The two scenarios are not necessarily incompatible. Often, outbreaks of disease are often associated with climate change - an example being Black Plague in Europe. If the peoples associated with Corded Ware had livestock that had already developed resistance to cattle plague, they may have been in a better position to ride out the crisis and fill the void left by the collapse of the EEFs.

BTW - The Corded Ware people are most likely Vasconic rather than Indo-European. There also appears to be a Vasconic/Yeniseian/Dene language substrate not only in Western Europe, but also in the Balkans. Some IE languages reflect this influence: words for "water" in Hittite, Germanic, and Slavic seem to be a compound of another language and an apparent Dene '-ta/-tar' ("water") component. In German, this apparent holdover Dene word for 'water' was "re-purposed" as "liquid" and as "wet" in Persian. The Basque language itself seems to have cognates with not only Yeniseian and Dene languages but even Algonkian languages are well.


Alvah said...

The direction of human migration out of the Americas is compatible with the data interpretation(s) German Dziebel outlines, period. Mal’ta I should be seen as ancestral to Europeans and not Native Americans since hgU is not found in Native Amerindian Populations. Mal’ta’s ancestral relationship to Europeans and not Native American Populations confers a recurring point; that derived Old World mtDNAs are not basal to Amerindian Populations but descendant of them. Native Americans maintain basal mtDNAs positions to hgs M, N, and R. They maintain “few, if any” of the derived lineages resulting from their sapient colonization of the Old World.

As for the “Out of Africa” theory the fact remains that all the proposed older ancestral lineages (L mtDNAs) do not accommodate this migration model as it is constrained by a “boot-strapped” exodus. A migration Out of Africa that does not include any of the proposed ancestral linages (including Y hgs A and B) is a very significant point that has never been adequately addressed. A more parsimonious explanation championed by German in this blog and his own website (http://anthropogenesis.kinshipstudies.org/) , counters with a modern dispersal into the African Continent (Johnson et al. 1983). German understands the big picture as he is not compromised by excluding the Western Hemisphere in searching for our Human Origins.

Tobus said...

@German:
My sole reason for entering this discussion was to refute your statement that "this finding confirms the "easternmost" (Amerindian-like) affinities of European hunters-gatherers" when in fact the data clearly show that La Brana doesn't have any Amerindian-like affinity.

You have just agreed with me by saying "La Brana looks completely West Eurasian without Amerindian admixture.", so regardless of how much integrity you think your argument has or what "cards" you might be holding, I rest my case.

Thanks for "straightening" out the facts for me, but we've been through this before - there are other interpretations besides yours and the Amerinidan affinity with Tianyuan and modern East Asians makes your theory unworkable. Given how much effort it's taken just to get you to accept the basic facts of La Brana's affinity, I really don't think there's much in it for me trying to convince your that your fringe theory is wrong - especially when you're just going to liberally sprinkly your arguments with ad hominen attacks and spurious logic. Enjoy your theory, I hope it makes you happy.

I think the plot discussed by Razib shows that La Brana is pulled eastward away from the rest of West Eurasians.

Yes, but this is toward South Asians and East Asians, neither of which is what you are referring to as "Amerindian" in Mal'ta.

Tobus said...

@Grey:
Which ancient writers are you talking about? Since farming came first then writing, I suspect the references to "red hair, light skin, grey/green eyes" are relatively recent, thousands of years after the depigmentation. There are several variants of MC1R are responsible for red hair and non-tanning (pale) skin but none of them are widespread and none show signs of positive selection.

The Neanderthal MC1R variant that suggests they had red hair/pale skin is not one that is found in Sapiens.

The East Asian alleles for light skin are different to the Europeans - light skin evolved separately via different genetic pathways in those two populations, European skin colour isn't due to East Asian admixture.


Red

eurologist said...

We would not see the forests we do today, during the Mesolithic. The glaciers were retreating, and we are talking about a polar desert here that was inhabited by mammoths and other giants...

Chad,

Sorry, you are way off. You are describing some of the environmental scenarios during the Younger Dryas or during or before the Ice Age - but not the very well-know state during and in the several millennia before the Neolithic (and also both ways outside the Younger Dryas).

http://motherjones.com/files/images/sweden_spruce_forest.jpg

http://4.bp.blogspot.com/-pMScFSqq_gw/UuHEmT2VY2I/AAAAAAAAnKU/7B6TEVKP068/s1600/r291610_1248021.jpg

http://upload.wikimedia.org/wikipedia/commons/b/bc/Biogradska_suma.jpg

http://upload.wikimedia.org/wikipedia/commons/e/e7/Tongass_National_Forest_2.jpg

Tobus said...

@Eurologist:
Of course - but not in the European context, as I have stated numerous times

I'm not sure what you mean by "in the European context" here... do you mean skin colour differences between "white" Europeans? Nearly all of the difference in European skin colour is environmental - studies of unexposed skin show very similar melanin levels from Ireland to Spain. It's tanning ability that largely produces the North/South difference, although both TYR and SLC45A2 show a slight cline (from 100-80% frequency) as you move south, which may explain what you are talking about.

There is no way these alleles come even close to being responsible to extant Central or N/ NW European skin color.

What do you base this on? Those populations are fixed for the light skin SLC24A5 and SLC45A2 variants and most of them have the TYR allele as well... every study on these alleles has confirmed that they are directly responsible light skin, so on what basis do you claim the opposite?

there are populations in NE Asia that rival the lightest in the world and don't harbor those mutations.

Which populations are you talking about? Jablonski (2000) took skin reflectance measurements from all over the world and found the darkest Europeans are lighter than the lightest East Asians. Is this just a subjective observation you've made yourself or do you have some evidence to support it?

In any case, East Asian populations evolved a separate genetic mechanism for light skin (proven in OCA2 and possibly in MC1R and DCT) so phenotype comparisons between East Asians and Europeans doesn't contradict the role of the SLCs and TYR in Europeans - you'd need to find a dark-skinned population that has the European alleles to do that.

It is mind-bogglingly ridiculous to assert that Mesolithic Europeans at the common 50 to 60+ degree (i.e., easily-habitable Northern-most PNW and Alaska-like)latitudes had anything else but light skin color

Why do you say it's ridiculous? It'd be much more mind-bogglingly ridiculous to ignore the facts just because they don't agree with our preconceived notions... all the evidence I've seen says that La Brana and Loschbour had dark skin - do you have any actual facts that show they had light skin, or is it just a personal theory of yours?

For the record I'll also point out that the two "dark-skinned" mesolithics found were in Spain (43 degrees) and Luxembourg (49 degrees), so (just) outside your 50-60+ range, and that people don't immediately change colour when they enter a new enviroment - it takes a random mutation plus scores of generations to spread it. People could retain dark skin in high UV enviroments for thousands of years before experiencing depigmentation on a population-wide scale.

German Dziebel said...

@Tobus

" the data clearly show that La Brana doesn't have any Amerindian-like affinity."

Wrong. Here's why. On the one hand, some of the data was not presented by Olalde. Look at their F3 and D statistics tables and you'll see that Amerindians are simply not listed. On the other hand, the data that was presented between Raghavan, Lazaridis and Olalde suggests that La Brana does have Amerindian-like affinity because it's clear that La Brana and Mal'ta share common descent and Mal'ta and Amerindians do, too. The special proximity between Mal'ta and La Brana comes not from their "western Eurasianness" (if this was the case, Mal'ta would have been equally close to modern West Eurasians) but from their Amerindianness (or at least eastern non-Africanness). La Brana's Y-DNA is hg C6 and Mal'ta's mtDNA is hg R which is a sister clade of Q. Amerindians are hsg C and Q.

"the Amerinidan affinity with Tianyuan and modern East Asians makes your theory unworkable. "

It makes my theory work even stronger. Some of the most prominent and oldest easternmost ancient DNA samples such as Tianyan, Mal'ta and even Altai Neandertals are closer to modern Amerindians than they are to other Western and Eastern Eurasian populations. This suggests that Western and Eastern Eurasians are Amerindian offshoots and La Brana and other later samples still carry this signal. Modern East Asians may have taken an additional boost of northern Amerindianness in early Holocene times.

"I really don't think there's much in it for me trying to convince your that your fringe theory is wrong - especially when you're just going to liberally sprinkly your arguments with ad hominen attacks and spurious logic. Enjoy your theory, I hope it makes you happy."

As long as you continue to waive your responsibility to provide data and use scientific arguments, you can't hope for a respectful attitude from me. Nobody needs another believer in mainstream theories. We need people who can actively and tangibly advance our understanding of human prehistory. I gave you a chance to admit your flaws but you didn't take it. The invitation remains open. That's the only road for people like you to avoid what you subjectively perceive as "ad hominem attacks."

"Yes, but this is toward South Asians and East Asians, neither of which is what you are referring to as "Amerindian" in Mal'ta."

Here's a good example of your bias and spurious logic. The plot we are talking about does not even have Amerindians on it (for some inexplicable reason). But it does demonstrate the eastward pull that makes perfect sense in light of all the other data. And hence East Asians and Southeast Asians don't show any special proximity to any ancient West Eurasians, and the Mal'ta sample is a great example of this, it's likely the missing Amerindians that are responsible for the La Brana eastward pull.

Tobus said...

@German:
Look at their F3 and D statistics tables and you'll see that Amerindians are simply not listed

Incorrect, Amerindians are included in the f3 stats (EDF 5) and they show no affinity to La Brana.

As already explained multiple times the D-statistics aren't useful in this discussion (they are all D(Out, MA-1, LB, X) so using Amerindian for X would only tell us if MA-1 was closer to LB or Amerindians, not if LB has any Amerindian affinity... for someone who claims superior understanding it's taking you an awfully long time to twig to how D-stats work).

...suggests that La Brana does have Amerindian-like affinity because it's clear that La Brana and Mal'ta share common descent and Mal'ta and Amerindians do, too

Again, that's a valid and logical possibility but it's not the only valid and logical possibility, and it doesn't match the facts, so you should drop it in favour of a possibility that does fit. Consider that Halle Berry has African affinity, and person X has affinity with Halle Berry... does that mean X must also have African affinity? No, it's certainly a possibility *IF* the affinity is in the African part of their genome (X is Will Smith for example), but since Halle Berry is part-African and part-European, X can have zero African affinity if X shares only European DNA with her (X is George Bush for example). Since a large part of the MA-1 genome is non-Amerindian, LB's affinity with MA-1 doesn't automatically mean LB shares MA-1's Amerindian affinity. You can only say LB has Amerindian affinity if LB actually *has* Amerindian affinity... and the data shows he does not. You yourself acknowledged this in your previous post so I don't know why you are still harping on about it.

It makes my theory work even stronger

That's because you haven't thought it through. Raghavan's data explicitly rejects the possibility of Amerindian admixture into MA-1 based on the assumption that Amerindians and East Asians are descendents of Tianyuan. So it's implicit in your theory (and you stated it explicitly to me in a previous thread) that Amerindians are ancestors of both Tianyuan and modern East Asians as well as contributing to MA-1. In your scenario we'd expect East Asians to be closer to Tianyuan than Amerindians, and we'd expect MA-1 (and Europeans) to be closer (or equidistant in a mutiple wave scenario) to East Asians than Amerindians. Just draw a tree diagram of the theoretical relationships your scenario proposes and you'll see that it doesn't hold up with the measured genetic relationships.

Here's a good example of your bias and spurious logic.

Don't you think implying Amerindian affinity from a chart that doesn't include Amerindians is even more biased and spurious?

The chart shows a slight (very slight - is it even statistically significant?) pull "eastwards", with South Asians being the closest population in that direction. We know LB has an affinity with Mal'ta and from his ADMIXTURE results we know that Mal'ta has a large South Asian component (much larger than his Amerindian component) which is confirmed by the EDF5 f3 stats in this paper where MA-1 plots with South Asians in most charts... it's not unexpected that LB would be closer to South Asians so we don't need a "missing" population to explain this very minor shift. I note you surreptitiously changed my "South Asian" to "South-East Asian" in your last reply - another spelling mistake?

Chad Rohlfsen said...

Has anyone seen the article on mtDNA C in NE Europe during the Mesolithic?

http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0087612

Chad Rohlfsen said...

Some interesting things here in these links. Apparently in Britain, one needs to be in mid-day sun from 4-28 minutes a day for 3-4 days a week to get adequate Vitamin D. People with dark skin need up to six times as much time in the sun. So, around 30min-3 hours a day for 3-4 days a week. Perhaps 5-6 hours at any given time during the day.

I would imagine that would be very possible and the reason that Mesolithic Europeans were able to remain dark-skinned.

One needs around 600 IU of Vitamin D a day, from 1 year old to 70 years old. Aside from time in the sunlight, you can get 400 IU of Vitamin D just from 3 oz. of Salmon, 9 oz. of tuna, and quite a bit more from the liver, which they probably ate. Cod liver has over 1300 IU/tbsp!!! So if these hunters eat a couple fish a day with the liver; it is more than enough to maintain healthy Vitamin D levels during rainy or snowy seasons.

http://www.bio-medicine.org/medicine-news/Dark-Skin-Require-More-Sun-Exposure-for-Vitamin-D-9188-1/

http://ods.od.nih.gov/factsheets/VitaminD-HealthProfessional/

Grey said...

@Tobus

"I suspect the references to "red hair, light skin, grey/green eyes" are relatively recent, thousands of years after the depigmentation."

Yes, that's the point. The red hair, grey or green eyed, light skin phenotype lasted long enough to get into the written record from China to Rome and yet the farmers are dark haired, dark eyed and light skinned so where does the red phenotype come from unless it was already there?

.

"There are several variants of MC1R are responsible for red hair and non-tanning (pale) skin but none of them are widespread and none show signs of positive selection."

If the red hair phenotype has retreated you'd expect that. The farmer's improved (tanning) version of depigmentation would have been better.

Is it possible to test if MC1R *was* positively selected for in the semi-distant past but then selected against in the last few thousand years as the SLC genes spread?

.

"The Neanderthal MC1R variant that suggests they had red hair/pale skin is not one that is found in Sapiens."

Yes but it points at partial depigmentation as a northern adaptation.

The phenotype ties in with the written record on sapiens.

The same phenotype in sapiens survived most in the region furthest from the farmer entry point.

.

"The East Asian alleles for light skin are different to the Europeans - light skin evolved separately via different genetic pathways in those two populations, European skin colour isn't due to East Asian admixture."

Skin color no but they do effect hair and eye color in Europeans.

http://scienceblogs.com/gnxp/2010/03/05/oca2-makes-east-asians-white-a/

Grey said...

"Which ancient writers are you talking about?"

Chinese descriptions of the western barbarians

(which makes me wonder if the Tarim mummies were typed for these genes?)

Greek writers talk about Thrace and the Ukraine region.

Romans talk about Britons and Scots.

IIRC Libyans were described this way also but i can't remember who by.

Grey said...

@Annie Mouse

"It could partly be nutritional."

Interesting thought. Although in this particular case sure it wasn't the sun bleaching his hair though?

terryt said...

"Mal’ta’s ancestral relationship to Europeans and not Native American Populations confers a recurring point; that derived Old World mtDNAs are not basal to Amerindian Populations but descendant of them. Native Americans maintain basal mtDNAs positions to hgs M, N, and R. They maintain 'few, if any' of the derived lineages resulting from their sapient colonization of the Old World".

That is simly not true. American haplogroups are not ancestral to Eurasian haplogroups in any shape or form. American haplogroups are subsets of Eurasian ones. American mt-DNAs are A2 (subset of Eurasian A4), B2 (subset of Eurasian B4b), C1b (subset of Eurasian C1, C4 (subset of Eurasian C), D1 (subset of Eurasian D4, and X2a (Subset of Eurasian X2). The Y-DNAs are likewise subsets of Eurasian haplogroups.

"Mal’ta I should be seen as ancestral to Europeans and not Native Americans since hgU is not found in Native Amerindian Populations".

Haplogroups are not a proxy for diploid DNA. Mal'ta can very easily be ancestral to Americans without having American haplogroups.

"A migration Out of Africa that does not include any of the proposed ancestral linages (including Y hgs A and B) is a very significant point that has never been adequately addressed".

Are you claiming that any movement out of Africa should include a fully representative sample of the whole DNA of that continent? It would follow logically from that that African-Americans too should have a fully representative sample of African haplogroups. The fact they don't requires a very convincing explanation from you.

"Given how much effort it's taken just to get you to accept the basic facts of La Brana's affinity, I really don't think there's much in it for me trying to convince your that your fringe theory is wrong - especially when you're just going to liberally sprinkly your arguments with ad hominen attacks and spurious logic. Enjoy your theory, I hope it makes you happy".

Agreed. In fact Germans ability to dance on the head of a pin is outstanding. He should enter politics (perhaps already has?). At anoth Dienekes post he is arguing strongly for what he claims a doubtful mt-DNA phylogeny supports yet he has a long history of completely dismissing mt-DNA phylogenies. He uses data to suit himself, sometimes to argue contradictory positions. You can't have a rational discussion with someone like that.

Simon_W said...

@ Gary Moore

Well, the ethnic interpretation of the Corded Ware is uncertain, but I'm thinking they're the root for the Balts and Slavs, partly also for the Germanics, and possibly even for the Indo-Iranians.

As for water: A German etymological dictionary informs me that in the different Germanic languages there are two different suffixes used, which go back to a PIE word that had both suffixes, depending on the case. West Germanic (watar), Old High German (wassar), Middle High German (wasser), Old Saxon (watar), Middle Low German, Dutch and English (water) have the -r suffix. In contrast Old Norse (vatn), Swedish (vatten) and Gothic (wato) have an -n suffix. Hittite still had both suffixes: Nom. & Acc. Sing. watar – the other cases weten-. The PIE root is aud-, ued-, ud-, meaning „water“, and being an extension from the verbal root au-, signifying „to sprinkle, to moisten, to flow“. Related are Old Indic udan, Greek hydor (water), Latin unda (wave), Old Irish u(i)sce (water, hence also whisky), Old Slavic voda.
So, the t in water that you associate with Dene -ta/-tar was already part of the PIE root.
And BTW, I'm not sure what you're alluding to with the alleged repurposing of that Dene word in German as „liquid“. The German word for „liquid“ is Flüssigkeit, or as an adjective: flüssig.

As regards the genetical relationships of Basque: The mainstream view among scholars of Basque is to regard it as an isolated language.

Also I would like to point out that, while both non-IE Basques as well as IE insular Celts have very high levels of R1b, the autosomal Ancient North Eurasian admixture is much stronger in the Celts than in the Basques. And it's the amount of the autosomal admixture that reflects the overall genetic admixture - the incidence of a y-haplogroup in contrast may deviate from that due to drift and sex-biased gene-flow. Furthermore, older theories of Pictish being a non-IE language are now mostly discarded, and the dominant view among experts is to regard it as a p-Celtic language.

By the way, for the reason I just explained, y-DNA is pretty much useless if you want to detect your Native American ancestry. I can only recommend an autosomal test and the DIY Dodecad calculators, at least those containing an Amerindian component, especially Globe10 and Globe13. In my experience, they can even detect admixture that's too weak for the Population Finder by FTDNA.

Eastern View said...
This comment has been removed by the author.
Eastern View said...
This comment has been removed by the author.
Chad Rohlfsen said...

@Grey

Greek is not written before 1450BCE and 1200BCE for Chinese. This is 3000 years after the Neolithic reached Britain, which is more than enough time for those features to have arisen in a new hunter/farmer hybrid.

When do these writings date from? Do you honestly believe that there are pure Mesolithic people running around after 1400 BCE?

Tobus said...

@Grey:
I get what you are saying, but I don't think there's any evidence for it - La Brana and the others don't have any of the ancestral MC1R variants known to cause this phenotype so if such a population did exist at that point in time, it's not represented in any of the samples we have.

All the sources you mention (it was Egyptians who painted Libyans with red hair, but only in the New Empire) are more recent than 2000 BC, at least 3000-4000 years after the dark-skinned European hunter-gatherer samples. It's also worth noting that 2 of the 4 references are from within Europe itself, indicating the phenotype was not Europe-wide but more likely individual tribes to the west and north, and in some cases "red" was used to refer to blonde/light brown hair as well.

Given the current distributions and the historical references I think it's fair to theorise that is was a common (predominant if we include blond/light brown?) phenotype in Eastern/Northern Europe (Celts, Gauls etc.) around 2000 BC and has diminished in frequency since that time. But there is no evidence it existed in Europe before the Neolithic revolution ~6000 BC, most likely it was either carried in with the lighter skin farmers, or it appeared in situ afterwards.

which makes me wonder if the Tarim mummies were typed for these genes?

Some of the earliest Tarim mummies (again ~2000BC) are known to feature samples with blue/green eyes, red/blonde hair and European features/haplotypes. The original DNA testing was done in the late 1990's, before SLC24A5 etc. were identified and as AFAIK they have never been tested for them. I think we'd all be very surprised if they didn't have the light skin variants.

Skin color no but they do effect hair and eye color in Europeans.

Not really - the OCA2 mutation that causes light skin in East Asians is rs1800414, the OCA2 mutations associated with blue eyes in Europeans are rs1800407, rs7495174, rs6497268 and rs11855019... same gene (general region of DNA), but different SNPs (individual mutations).


Mikel Mariñelarena said...

I'm just a layman in genetics but find it hard to believe that the prevalent looks of Europeans a mere 7000 years ago were that exotic. We have a much more reliable reconstruction of a 5,300 yo European (Oetzi) who had light skin and would probably pass unnoticed today if he were to wear modern clothes.

And then, 4-3 ky ago we begin to have all sorts of evidence: carvings, mosaics, paintings, writings from Egypt, Crete, Greece, Middle East and, later on, Rome suggesting that western Caucasoids didn't look very different at all from now. Indeed, Greeks and Romans portray a picture of Europeans similar to the current one: light haired in the North and darker in the South. In summary, nothing much has changed in the European phenotype of the last ~4 ky years but we are to believe that in the preceding 3 ky our ancestors had a phenotype that has totally disappeared (La Braña) and then suddenly brand new phenotypes appeared (light hair/eyes/skin)?

I personally find it easier to believe that there are still gaps in the current knowledge of geneticists that make a reconstruction like this one too unreliable to be taken at face value. As a matter of fact, a couple of years ago I did the Genographic Project test for my Y haplogroup and it turned out to be R1b (M-269). Not too surprising, for a Basque. But their description of my "genetic journey" said that this haplogroup was associated with the Cro-Magnon. As I understand it, the current view is that R1b arrived to Europe in Neolithic times instead. For the time being, I'm not taking any more genetic ancestry tests.

German Dziebel said...

@Tobus

"Incorrect, Amerindians are included in the f3 stats (EDF 5) and they show no affinity to La Brana."

You continue to twist the truth. They excluded Amerindians from the tables, which was my point. Why in the world would they exclude them from the tables? Amerindian data is so frequently excluded from studies that I wonder if it's purposeful because it doesn't fit the paradigm. You should be attacking these biases which are objective and wrong. Instead, you're covering them up.

"taking you an awfully long time to twig to how D-stats work)."

Olalde et al. should have published those f3 stats for Amerindians and La Brana. But logic helps one fill in the gaps.

"for someone who claims superior understanding"

Yes, I do claim superior understanding. Because there's a lot of hard work and lot of education behind it. Just get used to it and follow my example.

"Again, that's a valid and logical possibility but it's not the only valid and logical possibility, and it doesn't match the facts, so you should drop it in favour of a possibility that does fit."

I was the one who presented the facts and logic. You countered with pop culture examples. Everything that we know - and our knowledge is imperfect because labs present incomplete results - suggests that Amerindians is the common denominator between LB and MA-1.

"Since a large part of the MA-1 genome is non-Amerindian, LB's affinity with MA-1 doesn't automatically mean LB shares MA-1's Amerindian affinity."

But west eurasianness can't be the reason for their special affinity for the reasons I gave you earlier. Hence, Amerindian must be the reason. This is logic.

"You can only say LB has Amerindian affinity if LB actually *has* Amerindian affinity... and the data shows he does not. "

The data was not presented. Only data visualization was. See above.

"You yourself acknowledged this in your previous post so I don't know why you are still harping on about it."

Tables vs. figures. West and East Eurasian data was presented in both formats. Amerindian one only in one, the less precise one.

"We know LB has an affinity with Mal'ta and from his ADMIXTURE results we know that Mal'ta has a large South Asian component (much larger than his Amerindian component) which is confirmed by the EDF5 f3 stats in this paper where MA-1 plots with South Asians in most charts... it's not unexpected that LB would be closer to South Asians so we don't need a "missing" population to explain this very minor shift."

But Amerindians don't share a component with South Asians. And since f3 stats show Amerindians as the closest to MA-1 in Raghavan and South Asians are among their most distant relatives of MA-1 n Eurasia and since Amerindians share special affinity with other European samples, South Asians are less interesting in this regard. But I admit I haven't researched this issue much. BTW, who are these South Asians in Raghavan and in Olalde - ASI or ANI?

"so we don't need a "missing" population to explain this very minor shift."

But it is missing. And it's not my fault that it's missing. If it weren't missing, you wouldn't have been arguing with me.

"I note you surreptitiously changed my "South Asian" to "South-East Asian" in your last reply - another spelling mistake?"

Yes, that's how typos look like. How come you are so good at detecting my "surreptitious mistakes" and so bad at noticing much more obvious data-doctoring by Olalde et al.'s ? Oh, I have an answer: you are biased.

German Dziebel said...

@Tobus (contd.)

"That's because you haven't thought it through. Raghavan's data explicitly rejects the possibility of Amerindian admixture into MA-1 based on the assumption that Amerindians and East Asians are descendents of Tianyuan. So it's implicit in your theory (and you stated it explicitly to me in a previous thread) that Amerindians are ancestors of both Tianyuan and modern East Asians as well as contributing to MA-1. In your scenario we'd expect East Asians to be closer to Tianyuan than Amerindians, and we'd expect MA-1 (and Europeans) to be closer (or equidistant in a mutiple wave scenario) to East Asians than Amerindians. Just draw a tree diagram of the theoretical relationships your scenario proposes and you'll see that it doesn't hold up with the measured genetic relationships."

Where did Altai Neandertals go from my list of ancient Eurasian populations that show special proximity to Amerindians? How would you explain that - Neandertals admixing into Amerindians? This is definitely not a typo but a purposeful act of data doctoring in the face of a looming epistemological catastrophe.

"Raghavan's data explicitly rejects the possibility of Amerindian admixture into MA-1 based on the assumption that Amerindians and East Asians are descendents of Tianyuan."

The data doesn't reject it. It's Raghavan et al. that assume that Amerindians descended from East Asians (not Tianyuan!) hence they presumably couldn't have contributed to MA-1. The assumption is wrong and is not dictated by the data.

"In your scenario we'd expect East Asians to be closer to Tianyuan than Amerindians, and we'd expect MA-1 (and Europeans) to be closer (or equidistant in a multiple wave scenario) to East Asians than Amerindians. "

West Eurasians and East Asians diverged from a common Amerindian source but they don't share subsequent gene flow, hence they are not closer to each other than to Amerindians. Northern Amerindians have a special connection with modern East Asians because they contributed gene flow to Asia post Mal'ta times. What's so difficult to understand?

"Just draw a tree diagram of the theoretical relationships your scenario proposes and you'll see that it doesn't hold up with the measured genetic relationships."

You're repeating the mistake of lab geneticists who first build a theoretical model, a tree, and then they fit the data in it. I work from the data to a model. This is how science works.

@Eastren View

"it's good thing that DIscover fired razib. He wasn't well-acquainted with any area of the anthropological disciplines but really pretentious about offering his opinions that often were embarrassingly lay. Whenever he put up something technical that he didn't understand, he'll say that he hasn't "internalized it". That guy was such a jackass who kept linking to Amazon books in his comments so he can make money on the links. It's amazing people visited his site for so long and didn't realize what a pretender he is."

Yes, Razib Khan is a science profiteer.

Chad Rohlfsen said...

The oldest record of red hair in Europe is in Greek, from 500BCE. The oldest in China is from around 700BCE. The oldest body found with red hair is from 2000BCE in the Tarim Basin. This is way after the mixing event of 5000-3500BCE.

The last known hunter group, the Pitted Ware was mixed 15-30% EEF (same as root population for Saami and Finns that have SLC24A5 light allele). It is possible to see from this that there may have been no pure, dark-skinned Mesolithics by 3000BCE. That is 1600 years for the light skin alleles to set in before writing is known to have begun in Europe.

SLC45A2(very high occurrence in Europe) and TYR 40-50% in Europe) are important in mixed race people being lighter and showing shade variation. Which would be important in mixing light and dark-skinned people. This makes it possible for a light-skinned person to have light-skinned kids with a dark-skinned person. This is why my daughter is lighter skinned than me, with a Laotian/Native American mother. I am Northern European and Jew.

There are no light skin alleles in La Brana or Loschbour. Therefore, they cannot be light-skinned.

Grey said...

@Tobus
"I get what you are saying,"

cool

.

"but I don't think there's any evidence for it"

Yes i agree, the reason I've been going on about it is *if* there were evidence it might actually already have been found but be tucked away for example in medical studies related to skin problems.

.

"La Brana and the others don't have any of the ancestral MC1R variants known to cause this phenotype"

Yes I was using MC1R as it's better known. I think *if* this phenotype did exist then there might be a dark-haired version also connected to IRF4. For example I bet Cillian Murphy has IRF4. But that has even less evidence than for the red haired phenotype so never mind for now.

.

"All the sources you mention (it was Egyptians who painted Libyans with red hair, but only in the New Empire) are more recent than 2000 BC, at least 3000-4000 years after the dark-skinned European hunter-gatherer samples."

Yes if they started to mix straight away but some of the studies say they *mostly* didn't until much later which would narrow the gap.

That doesn't make it impossible of course.

.

But yes, I'll stop going on about it now. The main reason was to point out if there was evidence it might have already been found but in a separate field.

.

@Chad
"When do these writings date from? Do you honestly believe that there are pure Mesolithic people running around after 1400 BCE?"

Well not if they were brown-skinned as someone would have noticed and written it down.

.

@various
"Razib Khan is a science profiteer"

He seems okay to me.

Tobus said...

@German:
You continue to twist the truth

Fact: Amerindians are included in the f3 stats like everybody else and don't show an affinity with La Brana.
Fact: The D-statistics don't measure La Brana affinity so aren't relevant to the discussion at hand (3rd time now I've had to explain this to you!)
Conclusion: La Brana doesn't have Amerindian affinity.

Where is the twist?

Olalde et al. should have published those f3 stats for Amerindians and La Brana

They did, it's EDF 5. I'd be happy to put it table form for you if you're unable to understand the figure despite your "hard work and a lot of education" superiority.

But west eurasianness can't be the reason for their special affinity for the reasons I gave you earlier. Hence, Amerindian must be the reason. This is logic.

The "reasons you gave earlier" was really just one reason: "(if this was the case, Mal'ta would have been equally close to modern West Eurasians)". But as I've already pointed out, that's an invalid assumption - La Brana is 7,000 years closer to Mal'ta than modern West Eurasians are, so we'd expect him to show increased affinity even with the exact same "west eurasianness". Your statement might be logic, but it's spurious logic.

If it weren't missing, you wouldn't have been arguing with me.

I think I would - this chart's PC1 is essentially the same as Raghavan's PC2 on figure SI 10 (the Europe-East Asia dimension). If Amerindians were included they'd be grouped down with the East Asians and since LB has no affinity with either of them, I'd still be saying this miniscule "shift" (if it's not just statistical noise) is most likely South Asian - a population who are also East-shifted relative to Europeans, and who also share DNA with MA-1. We can only attribute it to Amerindian affinity if we have other evidence of such - if this is your only evidence that LB has Amerindian affinity then it's circular reasoning.

who are these South Asians in Raghavan and in Olalde - ASI or ANI?

They're modern South Asians so obviously a mix of both.

Yes, that's how typos look like. How come you are so good at detecting my "surreptitious mistakes" and so bad at noticing much more obvious data-doctoring by Olalde et al.'s ? Oh, I have an answer: you are biased.

... and as I predicted, the ad hominen attacks start. We've already has spurious logic - if only I'd said "conspiracy theories" I'd have had the trifecta!

Where did Altai Neandertals go from my list of ancient Eurasian populations that show special proximity to Amerindians?

This is the first mention of Altai (or any other) Neandertals in any discussion you've had with me. If you think it affects the Tianyuan/MA-1/East Asian/Amerindian relationship please tell me where it goes in your hypothesis.

It's Raghavan et al. that assume that Amerindians descended from East Asians (not Tianyuan!) hence they presumably couldn't have contributed to MA-1.

They actually say "diverged from East Asian (Han Chinese) ancestors" and since they're talking pre-Mal'ta I think Tianyuan is a good proxy for what they mean. My use of "Tianyuan" in this context doesn't change anything... unless of course you are acknowledging that Amerindians are in fact descendants of Tianyuan?

What's so difficult to understand?

You haven't given a timeline for your scenario, but based on what you've suggested in our previous discussion, under your assumption it's hard to understand:
a) why modern Amerindians are closer to Tianyuan than modern East Asians are - they should be the same distance with East Asian pulled closer to Tianyuan due to admixture with his descendants
b) why Mal'ta doesn't show East Asian affinity - he should be the same distance from Amerindians and East Asians

eurologist said...

I personally find it easier to believe that there are still gaps in the current knowledge of geneticists that make a reconstruction like this one too unreliable to be taken at face value.

Mikel,

Exactly. In fact, all we know for sure is that we don't yet know the mutations that cause the much lighter skin of central and northern Europeans - so we have no clue how Mesolithic people looked from the current genetic state of knowledge. From all other disciplines, however, they most likely looked as light as modern populations at the same latitude (who do not have a diet of predominantly fish and sea mammals) - that is, very light, like Uralic people or NE Asians.

Gary Moore said...

@Siman_W:

"As regards the genetical relationships of Basque: The mainstream view among scholars of Basque is to regard it as an isolated language."

Some linguists such as John D. Bengtson have proposed adding Basque to Dene-Caucasian (1). I personally think that Caucasian languages are not directly generic to Basque, but rather northern Caucasian languages interacted with and influence Dene-Yeseniean languages in the region of the Pontic Steppes as they did with early IE languages when they were passing through the region to influence phonetics and structure. Anyone who looks at Basque Sawdesh list might easily conclude that most is not all of their pronouns were picked out a a grab bag of word lists of various North American languages, mainly Dene and Algonquian. Vitaly V. Shevoroshkin believes that Almosan (Algonquian-Wakashan) languages are affiliated with Dene languages, and Basque appears to show some influence from this language family as well. Basque may be an isolate as far as European are concerned, but if it was found in North America, linguists would be arguing over whether it is a Dene or an Algonquian language.

"The PIE root is aud-, ued-, ud-, meaning „water“, and being an extension from the verbal root au-, signifying „to sprinkle, to moisten, to flow“. Related are Old Indic udan, Greek hydor (water), Latin unda (wave), Old Irish u(i)sce (water, hence also whisky), Old Slavic voda.
So, the t in water that you associate with Dene -ta/-tar was already part of the PIE root."

One of the problems encountered in these studies is conflicting etymologies and reconstructions of PIE. I've been working with Balkan languages (Albanian, Thracian, and Dacian). Walde & Pokorny (2) cite akwa as the IE root word for 'water'. Elsewhere, Russu (3) lists *uer, *au(e)r 'wet, moisten'. Please note that the latter looks a lot like the equivalent in Kott (a Yeniseian language): ūra,

I'm beginning to doubt that there ever was such as language as "Proto Indo-European" and so did the Indo-European scholar Max Muller towards the end of his career. The PIE Swadesh list I'm using has akwa as the IE root word for 'water', and I think it could just as easily have been the Iroquoian awa. Postulating an Iroquoian origin of IE could help fill in gaps in IE philology. For instance, there is no agreed-upon PIE word for 'they'. In Kurdish, it is e-wan-e and iishån in Persian. In Slavic, oni, one, ona. Applying phonetic shift inferred from (a)wa-ta to Slavic woda/voda, we derive ani, ane, ana. The Cherokee pronomial affix equivalent is ani-. (Cherokee has an animate/inanimate gender scheme as opposed to Slavic M/F/N.) In the western centum IE languages, in a language innovation the original pronoun has been replaced by words derived from a relative pronoun, as in English he/here. Iroquoian does have a counterpart to the PIE ki-. 'Here' in Mohawk is kèn:'en and in Cherokee is ahani, Cherokee having undergone a similar shift from 'k-' to 'h-' as Germanic languages did. ('That ' in Cherokee, BTW, is hina.)


(1) Bengston, John D., 1996. "A Final (?) Response to the Basque Debate in Mother Tongue 1." (
(2) Walde-Pokorny, Julius (1959). The Indogermanisches etymologisches Wörterbuch (IEW, "Indo-European Etymological Dictionary") and notes.
(3) Russu, Ion Iosif (1969). Die Sprache der Thrako-Daker

Isidro57 said...

I have absolutely no problem understanding how the 3 mutations for lighter skin arriving with farmers from the Near East to Europe in fact changed the darker skin of Mesolithic Hunters that had no need for it because of their diet.
Unexposed skin in Europeans demonstrates that, slight variations in skin color along the North-South gradient in Europe over let's say 3K years, is plenty of time to attribute it to the environment.
By the way, 70% of Europeans today show vitamin D deficiency.
The mystery part in my opinion is the blue eyes...

Tobus said...

@Mikel:
I'm just a layman in genetics but find it hard to believe that the prevalent looks of Europeans a mere 7000 years ago were that exotic.

Yes, it's very suprising indeed but we're nearly 100% certain that the first Out Of Africans where dark skinned, so the transition to today's phenotype has to have happened sometime. It was usually assumed to be very early after the initial migration but recent studies have suggested it was a much more recent event. The Neolithic revolution has long been suggested as a possible time due to the change in diet reducing the vitamin D intake and making lighter skin a significant benefit in low UV regions. A number of research studies have found that the light skin mutations are some of the most strongly selected for in Europeans populations, so once they occurred they spread extremely quickly through the population... some have suggest in as little as 1,000 years (and remember, we're talking about much much smaller population sizes than today). Genetic evidence also suggests an almost complete replacement of the extant European population during the Neolithic expansion, so it's possible that "white" people existed in Central/West Asia for a substantial amount of time before occupying Europe. The samples we have are right on the edge of this expansion and one other sample (Stuttgart) shows a partially depigmented phenotype in European farmers at roughly the same time as the dark-skinned Mesolithic samples. As you point out, by the time of Otzi a few thousand years later, the new phenotype was in place.

We have a much more reliable reconstruction of a 5,300 yo European (Oetzi) who had light skin and would probably pass unnoticed today if he were to wear modern clothes.

The Otzi reconstruction is based on the exact same alleles that the Mesolithic samples are - do you have a reason for believing it's any "more reliable"?

nothing much has changed in the European phenotype of the last ~4 ky years but we are to believe that in the preceding 3 ky our ancestors had a phenotype that has totally disappeared (La Braña) and then suddenly brand new phenotypes appeared (light hair/eyes/skin)

That's how selective sweeps work, and doubly so in this case as it was likely accompanied by a large scale population replacement and a complete cultural revolution (hunter-gathering to farming).

I personally find it easier to believe that there are still gaps in the current knowledge of geneticists that make a reconstruction like this one too unreliable to be taken at face value

I agree that there are almost certainly gaps that will be filled and help paint a more definite picture, but the methods behind the reconstruction are the same as used everyday in forensic profiling and genetic determination of phenotype. It might not be 100% of the story, but it is extremely reliable - Mesolithic Europeans were undoubtedly darker skinned than modern Europeans.

Tobus said...


@eurologist:
all we know for sure is that we don't yet know the mutations that cause the much lighter skin of central and northern Europeans

There is almost no measurable difference between the melanin content of central and northern Europeans and other Europeans, and there are mutations that we know for sure cause all modern Europeans to be lighter than Africans, South Asians, Papuans, East Asians etc.... mutations which these Mesolithic samples are lacking.

so we have no clue how Mesolithic people looked from the current genetic state of knowledge

Except of course that they don't have any of the "light-skin" alleles that modern light-skinned European populations do, so are undoubtedly darker than them.

they most likely looked as light as modern populations at the same latitude... - that is, very light, like Uralic people or NE Asians.

Uralic people all have the European light-skin alleles. NE Asians have other mutations that are known to lighten their skin - La Brana and Loschbourg have none of these. The lightest modern populations I'm aware of that have the same genetic profile as these Mesolithic samples are central and southern Native Americans, the darkest are southern South Asians and Melanesians. Given the current state of knowledge it's extremely likely that Mesolithic Europeans had a skin colour somewhere in the range between these two population - it's extremely unlikely that they were as light as Uralic/NE Asian populations are today... that would require some alternative genetic mutation with a similar effect that has since disappeared from the gene pool, despite strong selection for that exact phenotype in other alleles.

German Dziebel said...

@Tobus

"Fact: Amerindians are included in the f3 stats like everybody else and don't show an affinity with La Brana.
Fact: The D-statistics don't measure La Brana affinity so aren't relevant to the discussion at hand (3rd time now I've had to explain this to you!)
Conclusion: La Brana doesn't have Amerindian affinity."

Nonsense. Look at EDF 5 in Olalde and compare the d) graph with Fig SI 21 in Raghavan. In both graphs, along the X axis, you'll see Karitiana/Amerindians at 0.194 and in Olalde you'll find LB at 0.190. East Asians are in the area of 0.165 just like in Raghavan. You can see how the LB black star is pulled all the way to the right toward Amerindians way beyond East Asians, Central Asians and South Asians.

Using your model:

Fact: all of West Eurasians are closer to Amerindians than to East Asians (Raghavan).
Fact: MA-1 is closer to Amerindians than to East Asians (or any other population) (Raghavan)
Fact: MA-1 is closer to LB than to other West Eurasians or East Asians (Olalde)
Fact: LB is closer to West Eurasians than to East Asians (Olalde)
Prediction: LB and Amerindians share common ancestry
Test: EDF 5d in Olalde and SI 21 in Raghavan prove that the prediction is right.

"They did, it's EDF 5. I'd be happy to put it table form for you"

I'll take your generous offer. Instead of wasting my time without doing any actual work, please put it in a table format and list side by side with the LB data the data from Raghavan showing MA-1-Amerindian affinity. Also please add Amerindian data to the D stat tables (S10-S11 in Olalde), which currently feature East Asians, West Eurasians but no Amerindians. My prediction is that LB, just like MA-1, will be closer to Amerindians than to East Asians.

"La Brana is 7,000 years closer to Mal'ta than modern West Eurasians are, so we'd expect him to show increased affinity even with the exact same "west eurasianness".

Fact: MA-1 is closer to Amerindians than to other West Eurasians (Ragahavan).
Fact: MA-1 is closer to LB than to other West Eurasians (Olalde).
Fact: LB is "7,000 years closer" to Mal'ta in time
Conclusion: LB is "7,000 years closer" to Amerindians than other West Eurasians

"This is the first mention of Altai (or any other) Neandertals in any discussion you've had with me."

Because it came up after our initial discussion. See Prufer 2014 Suppl Mat, S13.1. This is consistent with several other indications that Neandertals and Amerindians share special affinity.

"If you think it affects the Tianyuan/MA-1/East Asian/Amerindian relationship please tell me where it goes in your hypothesis."

Let's wrap up the LB discussion first.

"They're modern South Asians so obviously a mix of both. "

Then we can ignore them, as ANI would pull them up to West Eurasians (or the other way around) and it's a recent link.

German Dziebel said...

@tobus (contd.)

"as I predicted, the ad hominen attacks start."

What you subjectively and defensively perceive as ad hominem attacks are in reality attacks ad bias, ad heckling, ad whistleblowing, ad filibustering... Stop those discursive practices and everything will be fine.

"a) why modern Amerindians are closer to Tianyuan than modern East Asians are - they should be the same distance with East Asian pulled closer to Tianyuan due to admixture with his descendants
b) why Mal'ta doesn't show East Asian affinity - he should be the same distance from Amerindians and East Asians."

a) Eskimo-Aleuts and Na-Dene are closer to East Asians than Karitiana (Raghavan). But Karitiana is closer to Tianyuan than East Asians. Admittedly, we don't have the data comparing Tianyan to EA/ND vs. Karitiana vs. East Asians. I predict that modern East Asians are closer to Tianyuan than EA/ND because of admixture with Tianyuan but farther removed from Tianyuan than Karitiana because they are a more recent spinoff from northern Amerindians, while Tianyuan is an ancient relic related to more conservative Karitiana; b) when populations diverge from a common source they become different. Raghavan assumed that Amerindians descended from East Asians, hence, by this logic, if MA-1 got gene flow from the descendants of East Asians, then MA-1 should be closer to East Asians than, say, to Papuans. But if West Eurasians and East Asians both descended from Amerindians, diverged and didn't admix ever since, then they won't share any similarity with each other, unless Amerindians are involved in the comparison.

eurologist said...

Genetic evidence also suggests an almost complete replacement of the extant European population during the Neolithic expansion

Tobus,

That statement is incorrect, and in fact the lightest skin in Europeans today occurs in those populations that have the least Neolithic and Bronze/ Iron age admixture.

There is almost no measurable difference between the melanin content of central and northern Europeans and other Europeans

I am simply assuming you are not serious with that statement.

it's extremely unlikely that they were as light as Uralic/NE Asian populations are today... that would require some alternative genetic mutation with a similar effect that has since disappeared from the gene pool

Why? The opposite is the case, the mutations that code for the lighter skin color of Central and Northern Europeans are obviously still present - except that we haven't figured out yet which they are.

Grey said...

@Tobus

I was going to leave it but as you're now talking complete nonsense...

"Genetic evidence also suggests an almost complete replacement of the extant European population during the Neolithic expansion"

This has been shown to be totally wrong.

The genetic evidence (Laziridis etc) shows the exact opposite - that there was a neolithic expansion and then collapse with the vast majority of surviving European DNA above the Alps being pre-neolithic.

WHG, ANE and 50% of EEF are all pre-neolithic.

This is one of the main reasons why the dark-skinned argument is so unlikely because the farmers were over-run in the north by the supposedly brown-skinned WHG and ANE. It makes no sense.

.

"that would require some alternative genetic mutation with a similar effect that has since disappeared from the gene pool, despite strong selection for that exact phenotype in other alleles."

It does not require anything of the sort. If the *known* skin-lightening alleles are fixed in white people then by definition any earlier de-pigmentation will be hidden. That's why the studies on skin-lightening alleles have had to be conducted on mixed-ancestry people. You can't tell what percentage of skin-lightening SLC24A5 causes on white skin.

The entire basis of this argument is the *assumption* they had brown skin at this time. There is no way to prove that assumption is correct. However it can be proved incorrect by looking for white-skinned people who don't have the SLC genes.

.

nb

SLC24A5 has a c. 33% skin-lightening effect on mixed-ancestry people

the other SLC gene has 15-20%

KITLG is also 15-20%

plus there's a bunch of others with smaller effects.

so what does that mean?

It means Europeans had to have *all* of them at fixation to not be various shades of brown by the time the Romans, Greeks and Chinese wrote about them.

If the IE were already white somehow - thus adding to the farmer portion of the total - and it was only the WHG who were brown-skinned then maybe but otherwise it's nonsense.

.

Isidro57

"The mystery part in my opinion is the blue eyes..."

That to my mind is another clear hint they were already at least partially depigmented.

Mikel Mariñelarena said...

Hi Tobus,

Thanks for your answer. For all I know, you could well be right but I'm still surprised about the fast changes that are proposed for the Europeans' phenotypes and a bit skeptical about the degree of certainty that we can currently have on such issues.

The reason why I said that Ötzi's reconstructions are probably more reliable is because, well, we have the actual frozen corpse of that guy.

On the other hand, your explanations make me wonder about why similar changes didn't happen in the Americas. For ~20k years Amerindians had to adapt to a huge variety of climates/diets and went through their own Neolithic revolution. However, we see very little variation in their phenotypes. Those living in the tropics or in the Andes Plateau do seem to have acquired a darker skin tone but, generally speaking, they still all look quite similar to their ancestors in Northern Siberia and to each other: raven-black hair, dark eyes and sallow skin. In some places of the Americas (Northwestern Pacific coast or Patagonia) the climate is as harsh as anything you can find in Europe but I've seen Patagonian Amerindians and they look quite dark. A huge contrast to what is proposed for Europe.

Simon_W said...

Tobus said: The lightest modern populations I'm aware of that have the same genetic profile as these Mesolithic samples are central and southern Native Americans, the darkest are southern South Asians and Melanesians.

So, definitely still a wide range of possible pigmentation. I think the idea of Mesolithic Europeans with a skin colour similar to central and southern Native Americans shouldn't be that hard to swallow, given the different nutrition.

Simon_W said...

Gary Moore, I'm not the right person here to argue over such linguistic issues in any detail, there are other commenters around with better linguistic training. I just know that perhaps the main problem of these supposed macro-families is that analogies between short syllables may occur by chance, especially if you've got the whole wide range of all Amerindian or Caucasian, or whatever world languages, to select from. Moreover the analogies most often seem to be based on nothing more than apparent similarity. While in Indo-European studies, one thing that strikes the layman is that accepted cognates are often not easily recognizable as such. The thing that proves their standing as cognates is that they can be deduced with known and regular sound changes. To my understanding, the sceptical mainstream linguists who are not convinced of macrofamilies like Dene-Caucasian don't deny that it's possible, they rather think that the evidence is not convincing enough. And lastly I want to add that, for example, you suggested that something akin to Dene -ta/-tar, denoting water, was present in a Vasconic substrate in Europe. But Basque, which is Vasconic of course, has ur for water, which is hardly similar to -ta/-tar.

eurologist said...

Yes, it's very suprising indeed but we're nearly 100% certain that the first Out Of Africans where dark skinned, so the transition to today's phenotype has to have happened sometime. It was usually assumed to be very early after the initial migration but recent studies have suggested it was a much more recent event.

Tobus,

First of all, from many studies about skin color, vitamin D, Calcium, and folic acid, we are very certain that Northeast Africans 130,000 to 80,000 ya were not dark colored. They simply could not have been by the biochemistry and UV illumination we have known for about 20-30 years.

Genetic evidence also suggests an almost complete replacement of the extant European population during the Neolithic expansion,...

There is zero evidence for this, and strong evidence that N and E Europeans experienced very little gene replacement from outside Europe, and that other Europeans mostly experienced gene replacement from other Europeans.

Uralic people all have the European light-skin alleles.

The alleles you are talking about are not European. Many of them are are extremely common in Western Eurasian people, S Asians, SW Asians, and N Africans. They are not particular to Europeans, and even less so to C or N Europeans. What clearly sets C and N Europeans apart is nothing like that, but a similar parallel evolution as NE Asians.

C and N Europeans Like Uralic and N Asian populations have been very well known for quite some to be lighter in skin color because of vitamin D necessity (latitude and other UV suppressing factors).

Simon_W said...

I've got a map of human skin colour by Fiorenzo Facchini, and according to this, the native Americans in central and south America vary a lot! The darkest areas (e.g. Mayans, eastern Brazil, southern Argentina) have skin pigmentation values similar to the darkest parts of India and of northern New Guinea. On the other hand, large parts of central and south America (e.g. northwestern Brazil, the countries bordering the Caribbean, southern central America) harbour Amerindians with pigmentation values similar to Egypt, for instance. And the Amerindians of the southernmost third of Chile were, always according to the map, even as light as East Asians or northern Maghrebis! If true, the western European hunter-gatherers may have had quite a light skin colour. But we may speculate that these southernmost native Americans have had some unknown genes for light skin.

Tobus said...

@German:
Nonsense. Look at EDF 5 in Olalde and compare the d) graph...You can see how the LB black star is pulled all the way to the right toward Amerindians

If you take a look under the X-axis on EDF 5 you can see some words that say "f3(Mal'ta, X, Yoruba)". What this means is the the X-axis represents affinity with Mal'ta, and what you have interpreted as "pulled ... towards Amerindians" is actually pulled towards Mal'ta.

In contrast, take a look at the X-axis on the a) graph - it's got the same label but with "Karitiana" instead of "Mal'ta". This means the X-axis represents affinity with Amerindians. You can see that on this axis the black star is in the range of the blue European dots. This means that LB has no increased affinity with Amerindians compared to modern Europeans.

Furthermore, if you look at the same axis on graph b) (affinity with Han), you can see that the Europeans and LB are grouped almost identically as they were in graph a). What this means is that in terms of "Eastern" affinity LB is the same whether East Asians or Amerindians are used (and hence, doesn't have any special Amerindian affinity).

Raghavan's SI 21 doesn't feature La Brana at all, so can't be used to confirm or deny your proposition.

Prediction: LB and Amerindians share common ancestry
Test: EDF 5d in Olalde and SI 21 in Raghavan prove that the prediction is right.


The prediction I'm arguing about is LB's "increased affinity" with Amerindians relative to Europeans and as explained above neither EDF 5d nor SI 21 prove this - if you're now saying that LB has the exact same Amerindian affinity as modern Europeans then we're in agreement.

I'll take your generous offer

My apologies, I assumed the rhetoric was implied. I think I've explained how the graphs work simply enough above that you can understand them now. If not, let me know and I'll walk you through it again.

Also, for the 4th time, the D-statistics in this paper don't measure LB's relative affinity to anything, just Mal'ta's - what's so hard to understand about that?

Conclusion: LB is "7,000 years closer" to Amerindians than other West Eurasians

That conclusion is based on a flawed assumption - affinity to Mal'ta doesn't necessarily mean affinity to Amerindians, just like affinity to Halle Berry doesn't necessarily mean that George Bush has affinity with Desmond Tutu. Again, what's so hard to understand about that?

What you subjectively and defensively perceive as ad hominem attacks are in reality attacks ad bias, ad heckling, ad whistleblowing, ad filibustering

"Ad hominen" is a logical fallacy whereby the offender targets the person making the argument rather than responding to the facts and/or reasoning of the argument itself. Ignoring my reasoning and calling me "biased" is an example of this, as is, somewhat ironically, accusing me of "heckling", "whistleblowing", "filibustering" etc.

Tobus said...

@German: (cont)

In regards to the flaws in your hypothesis:

a) but farther removed from Tianyuan than Karitiana because they are a more recent spinoff from northern Amerindians, while Tianyuan is an ancient relic related to more conservative Karitiana;

So walk me through this - sometime at least 45kya ago a population "ABC" lives in America. It splits into "A" and "BC" and A migrates to East Asia and leaves us the Tianyuan sample. Sometime later (you've never said how much) "B" and "C" diverge, and a branch of B also migrates to East Asia and breeds with A's children... how am I wrong in expecting the migrant Bs to be closer to A in this scenario?

b) But if West Eurasians and East Asians both descended from Amerindians, diverged and didn't admix ever since, then they won't share any similarity with each other

In that scenario whichever population diverged first would share the same relative distance to the other one as they do to Amerindians. If MA-1/Europeans split from Amerindians first they'd have just as much East Asian affinity as Amerindian, but if East Asians came before MA-1 then they shouldn't be so much closer to Amerindians... so which is it?

Tobus said...

@eurologist and Grey:

That statement is incorrect
This has been shown to be totally wrong.

My apologies you are both correct - Y/mt haplogroups suggested a large scale replacement but autosomnal DNA proves it was more of an assimilation. I stand corrected.

My main point still remains however that this was a time of huge shifts in genetics and lifestyle... this period is a good candidate for when such changes might have occurred, and it fits reasonable well with the timing indicated by the genetics.

@euroloist: I am simply assuming you are not serious with that statement.

I am totally serious. My source is Jablonski 2000 - she includes a table of direct skin reflectance measurements from around the world. The difference between the lightest (67.43) and darkest (62.3) European samples represents about 10% of the difference between the lightest (67.43) and darkest (19.3) samples worldwide. I remind you that we're talking about unexposed skin here - environmental factors can cause a much greater visible difference.

Why? The opposite is the case, the mutations that code for the lighter skin color of Central and Northern Europeans are obviously still present

Why? Because we know what the mutations that code for most of the lighter skin colour already are, and La Brana and Loschbour don't have them. If you posit they had the same skin colour but they don't have the same alleles, then it follows that there must be different alleles that caused the same effect... and if those alleles were still around today, then they'd make modern Europeans even lighter again (and hence not the same colour as LB), so they must have disappeared.

@Gray and eurologist:
WHG, ANE and 50% of EEF are all pre-neolithic.
the lightest skin in Europeans today occurs in those populations that have the least Neolithic and Bronze/ Iron age admixture

Admixture proportions don't really matter if an allele is under positive selection - it will be carried forward more often than the rest of the ancestry it was originally associated with. If the selection is strong enough it will end up in 100% of individuals including those that may only have a relative small proportion of it's original ancestry... that's how selection works. It only follows that ancestry represents phenotype if the phenotype was never under selection.

@eurologist: First of all, from many studies about skin color, vitamin D, Calcium, and folic acid, we are very certain that Northeast Africans 130,000 to 80,000 ya were not dark colored

Do you have any sources for that? I've read many studies on these topics but haven't ever come across that assertion. The commonly accepted coastal migration theory has the OOA migration staying pretty much in the tropics all the way to India, hinting that they stayed dark until well into Eurasia.

Tobus said...

@Mikel:
a bit skeptical about the degree of certainty that we can currently have on such issues.

Your skepticism is well founded, but I think it's fair to say given the current knowledge that LB/Loschbour were definitely darker than modern Europeans, probably at least as dark as Native Americans and possibly as dark as Melanesians. Hopefully there'll be more discoveries in the future that will give us a more definite view.

On the other hand, your explanations make me wonder about why similar changes didn't happen in the Americas.

There are a few different theories but very little evidence - since they presumably ate a lot of fish/sea mammals high in Vitamin D during the migration it's possible they retained darker skin (from a partially depigmented paleo-East Asian ancestor?) the whole time. Others suggest they depigmented on the way up and repigmented on the way down. They seem to share most of the candidate genes with East Asians, but not the OCA2 allele found in the lightest Chinese/Japanese so I personally suspect they represent a "first wave" of lighter skin colour in all East Asians ~20-30kya and haven't changed much since then - perhaps suggesting that functional skin colour relies on fairly rare mutations and is not as fluid as many think. That's all conjecture though, as I said there's not a lot of actual evidence.

Tobus said...

@simon_w:I've got a map of human skin colour by Fiorenzo Facchini

Do you have a link to it? I doubt it's very accurate but would like to check it out.

eurologist said...

Yes, the general ideas about the coloration of Amerindians are these following:

* They started out fairly light-colored from Beringia, with some exceptions due to fish and sea mammal consumption in part of the population.

* They have not been able to fully recover the lost tropical pigmentation of Africans, because insufficient time has passed.

* In the tropics, they mostly live in the dense tree canopy shade during the day, not in a Savanna-like environment.

* Groups have moved around in the past 2-3 thousand years, so some are now not quite adapted to where they should be.

* Natives in the extreme North and South, especially the ones with not an extremely high fish/ sea mammal diet, indeed have very light skin.

* You cannot rely on old B&W pictures - same story for the Sami. Those emulsions did not have an even sensitivity throughout the spectrum. Sami, who are very light in color (see today's pictures) but then even more so than today retained the ability to tan, got a reddish-orangey tone that shows up as rather dark in these old photographs.

Wikipedia on this topic supports me:

Early photographic plates and films were usefully sensitive only to blue, violet and ultraviolet light. As a result, the relative tonal values in a scene registered roughly as they would appear if viewed through a piece of deep blue glass. Blue skies with interesting cloud formations photographed as a white blank. Any detail visible in masses of green foliage was due mainly to the colorless surface gloss. Bright yellows and reds appeared nearly black. Most skin tones came out unnaturally dark,...

German Dziebel said...

@Tobus

"If you take a look under the X-axis on EDF 5 you can see some words that say "f3(Mal'ta, X, Yoruba)". What this means is the the X-axis represents affinity with Mal'ta, and what you have interpreted as "pulled ... towards Amerindians" is actually pulled towards Mal'ta."

Wrong. The pull is towards both. It's very easy to see that LB is closer to Amerindians than East Asians, Central and South Asians when they are plotted against MA-1. As Lazaridis noted, MA-1 is closer to West Eurasians than are Amerindians (save Russians and Oracdians) but this is because Mal'ta is 25,000 years older than modern Amerindians in Lazaridis and Olalde samples. If you look at Rasmussen 2014 (the Anzick study, EDF6), Anzick is closer to Mal'ta than modern Amerindians which precisely what we would expect considering the 10,000 years less time separating MA-1 and Anzick.

"The prediction I'm arguing about is LB's "increased affinity" with Amerindians relative to Europeans and as explained above neither EDF 5d nor SI 21 prove this - if you're now saying that LB has the exact same Amerindian affinity as modern Europeans then we're in agreement. "

LB has more affinity to MA-1 and to Amerindians (especially, Anzick in the light of Rasmussen et al. ) than modern Europeans. You just have to accept it. I proved it to you beyond any reasonable doubt.

"I think I've explained how the graphs work simply enough above that you can understand them now. If not, let me know and I'll walk you through it again."

I know how the graphs work. What I'm teaching you is how to compare different graphs and make logical conclusions. And how these logical conclusions should prevail over your biases.

"That conclusion is based on a flawed assumption - affinity to Mal'ta doesn't necessarily mean affinity to Amerindians, just like affinity to Halle Berry doesn't necessarily mean that George Bush has affinity with Desmond Tutu. Again, what's so hard to understand about that?"

It's very easy to understand but your analogy is just some screens as the data shows that it's the same ancestry in the case of LB, MA-1 and modern Amerindians. There's no way you can dismiss Olalde's EDF 5d. LB is less West Eurasian (Sardinian) when it's compared with MA-1 and Amerindians. I wish you worked on the tables supplying actual values for Olalde's EDF5a and EDF5b because it looks like LB is closer to MA-1 and to Amerindians on the Karitiana axis than it's to East Asians on the Han axis. The pull is just softer but it's likely there.

German Dziebel said...

@Tobus (contd.)

""Ad hominen" is a logical fallacy whereby the offender targets the person making the argument rather than responding to the facts and/or reasoning of the argument itself."

Then your whistleblowing, etc. practices are ad hominem arguments because they are invariably accompanied by a) your refusal to address the substance of my argument and b) your frequent twisting and turning of the data in an obvious attempt to discredit a theory simply because it's novel. My reference to your biases is just the identification of a fact just like it's a fact that LB has Amerindian affinity. Your constant denial of this fact betrays your bias.

"In that scenario whichever population diverged first would share the same relative distance to the other one as they do to Amerindians. If MA-1/Europeans split from Amerindians first they'd have just as much East Asian affinity as Amerindian, but if East Asians came before MA-1 then they shouldn't be so much closer to Amerindians... so which is it?"

West Eurasians split from (what was to become southern) Amerindians first. Tens of thousands of years later East Asians split from a diverged group of Amerindians that I call "northern Amerindians." How can West Eurasians possibly have as much East Asian affinity as southern Amerindians if they separated from the ancestors of East Asians' parent population tens of thousands of years earlier? Tobus, I feel like I'm flying over a cuckoo's nest.

"It splits into "A" and "BC" and A migrates to East Asia and leaves us the Tianyuan sample..."

After Anzick's study, I don't know if Tianyuan is closer to southern Amerindians. They published data that contradicts Fu et al.'s data. So let's table this.

Chad Rohlfsen said...

You guys are going around in circles here. There are no light skinned people in the north, that do not have the light skin alleles from West or East Eurasians.

SLC45A2 and TYR are important in lightening skin for people with parents of different skin tones. My daughter's mother is Laotian and Native American, so she is very brown. I am Northern European, with some Jew. I am light skinned with dark brown hair and blue eyes. My daughter is lighter skinned than I am, with medium brown hair and blue eyes. Just as mixed Europeans could show up as. The skin reflectiveness of all Europeans is very close, as someone else stated.


IE people are likely to be part EEF. Probably at least 15-30% for the Corded Ware people. Pitted Ware people to the North appear to be 15%(Ire8)-30%(ajv70 and 52). The reason for this claim is that they show 15-30% of the way from La Brana to Gok4 (Plus they have farmer mtDNA). These Pitted Ware people are at the base of Finns and Saami. So Finns and Saami have the EEF and likely are also fixed with the SLC24A5 and likely the SLC45A2.

Lastly, there isn't a pure WHG/ANE comeback in the north. These people are herders (plus barley farmers) that are mixed with EEF. The mixing with EEF occurred from 5000BCE-3500BCE. There likely is no pure Mesolithic in Northern Europe after this date. So we can easily see Europeans colored similar to my daughter by this time. Mixing with these alleles that show high selection (natural and maybe sexual)can easily create people the same color as EEF(permitting for variance within populations that we see even today/not all Brits or Norwegians are the same color) with no recognizable skin coloring traits of darker ancestors.

eurologist said...

I am totally serious. My source is Jablonski 2000

Then you are aware that source mostly relies on previously attained measurements and basically does not differentiate between Europeans at all, and also does not differentiate between non-tanned (under-arm) and tanned complexion?

Do you have any sources for that?

This goes back 15-20 years. See, e.g., N. G. Jablonski & G. Chaplin, The evolution of human skin coloration, 2000. ;)

Tobus said...

@German:
It's very easy to see that LB is closer to Amerindians than East Asians, Central and South Asians when they are plotted against MA-1.
the data shows that it's the same ancestry in the case of LB, MA-1 and modern Amerindians. There's no way you can dismiss Olalde's EDF 5d. LB is less West Eurasian (Sardinian) when it's compared with MA-1 and Amerindians

This is where your logic is flawed - graph d) only measures affinity to MA-1. Saying that all the groups on this graph that are closer to MA-1 must therefore also be closer to each other is not a logical conclusion. Compare graph e) where both MA-1 and East Asians are closer to Karitiana - does this mean MA-1 must be closer to East Asians?

It's graph a) that measures direct Amerindian affinity there is absolutely no pull there. This proves conclusively that the pull we see in graph d) can only be to MA-1, an not to both MA-1 and Amerindians.

LB has more affinity to MA-1 and to Amerindians... than modern Europeans. You just have to accept it. I proved it to you beyond any reasonable doubt.
I know how the graphs work. What I'm teaching you is how to compare different graphs and make logical conclusions.

Your supposed "proof" is based on a invalid assumption that affinity to MA-1 necessitates affinity to Amerindians, something which is demonstrably false. Myself and other have pointed out this mistake multiple times but you seem to be having a hard time understanding it. There a multitude of real-world examples where such logic doesn't apply - European->Halle Berry->African, MA-1->Amerindian->East Asian, Father->Child->Mother etc. etc. It's simply a false proposition that if A and C are both closer to B, then A and C must both be closer to each other.

Your conclusions aren't logical, they are based on a false assumption and are inconsistent with direct measurements of LB/Amerindian affinity.

My reference to your biases is just the identification of a fact just like it's a fact that LB has Amerindian affinity

Cool, I'm not biased at all then :)

West Eurasians split from (what was to become southern) Amerindians first. Tens of thousands of years later East Asians split from a diverged group of Amerindians that I call "northern Amerindians."

So you are proposing a north/south Amerindian split, with Europeans subsequently coming from one branch and East Asians coming from the other?

This doesn't work either, if the north/south American split happened before Americans/Europeans split then south Americans are closer to Europeans than to East Asians (and to North Amerindians as well for that matter). On the other hand if the north/south split happened after Europeans diverged, then Europeans are just as East Asian as Amerindians. Neither model matches the facts.

Tobus said...

@eurologist:
Then you are aware that source mostly relies on previously attained measurements and basically does not differentiate between Europeans at all, and also does not differentiate between non-tanned (under-arm) and tanned complexion?

It contains separate measurements for United Kingdom (London, Wales, "Northern", Cumberland), Belgium, Ireland (Carnew, Rossmore, Longford, Ballinlough), Spain (Leon, Basques), Germany and the
Netherlands - probably not as conclusive as it would have been had it contained Swedes and Sicilians.

And, yes, as I mentioned earlier, it deliberately avoided environmental factors by only sampling unexposed skin - hence my point that skin colour difference between southern (Spain) and northern (UK/Netherlands) is primarily environmental.

This goes back 15-20 years. See, e.g., N. G. Jablonski & G. Chaplin, The evolution of human skin coloration, 2000. ;)

What a coincidence, I actually have a copy of that handy :)... but I can't find any suggestion in there about Northeast African skin colour 130-80kya. The only reference to historical populations I found was about people living above 40 degrees N 50-10kya.

Joakim said...

If the La Brana man didnt eat starch or lactose...he was eating high fat low carb diet...paleo diet. Am I right?

Simon_W said...

@ Tobus

That's the map:
http://imageshack.com/a/img838/6383/f825.png

It's from the German version of a 1990 book by Facchini. The eight classes of colours are based on the scale by von Luschan (1 – 36).

Annie Mouse said...

"How can West Eurasians possibly have as much East Asian affinity as southern Amerindians if they separated from the ancestors of East Asians' parent population tens of thousands of years earlier? "

East Asian admixture. Which we know happened, and is visible in the physical features of some Western Europeans.

Grey said...

@Chad

"You guys are going around in circles here."

Neolithic y dna was replaced by mesolithic y dna while neolithic mtdna increased in frequency over time.

The simplest explanation for that is the farmers were over-run by the WHG/ANE but the surviving farmer dna in the females was strongly selected for afterwards e.g. because of the SLC genes among other things.

If farmer mtdna H increased as a percentage over time while mesolithic mtdna U decreased as a percentage that implies the EEF percentage was *much lower* at the time of admixture and has increased since.

This means that if the WHG/ANE were brown then you have a majority brown population over-running a minority white population and producing a 100% white population.

It makes no sense.

However that is enough for now until we get some ancient dna from an IE population. If the IE were already white then the WHG being brown would be more plausible.

German Dziebel said...

@Tobus

"This is where your logic is flawed - graph d) only measures affinity to MA-1. Saying that all the groups on this graph that are closer to MA-1 must therefore also be closer to each other is not a logical conclusion."

You've decided to add to your portfolio of pseudoscientific antics one more - intellectual dishonesty. You know very well that every plot we're discussing compares different populations (X) to population A using outgroup B (Yoruba). If populations cluster together or show a pull toward another population, they share ancestry (through common descent or admixture). In EDF 5a,b LB is firmly part of the West Eurasian cluster on the Y axis. You would't question it right? Or you would dare to say that axis Y in EDF 5a,b only measures LB to Sardinian but not to French or Russian? Now, I want all of your remaining brainpower to focus on one simple truth: On EDF 5d LB would've been part of the Amerindian cluster (LB 0.19 and Amerindians range from 0.185 to 0.195) if only Mal'ta, X, Yoruba values were plotted.

This proves that LB has Amerindian and Mal'ta affinity. The reason EDF a, b are less representative in this regard is simply because LB and MA-1 are ancient samples and Amerindians are modern samples. Although if you furnish me with the actual numbers, you'll see an Amerindian pull there as well.

So what is your profile now? Whistleblowing, heckling, exhibiting bias, practicing intellectual dishonesty... What's next? Just plea guilty and I'll erase all of this from your record. Deal?

"Compare graph e) where both MA-1 and East Asians are closer to Karitiana - does this mean MA-1 must be closer to East Asians?"

Duh, you've just provided evidence for what you thought was impossible. When measured against Karitiana, East Asians and MA-1 are indeed close partners. And lo and behold, both are located in East Eurasia between the New World and West Eurasia.

Tobus said...

@German:

You've decided to add to your portfolio of pseudoscientific antics one more - intellectual dishonesty.

Ad hominen, ignoring.

You know very well that every plot we're discussing compares different populations (X) to population A using outgroup B (Yoruba).

Agreed.

If populations cluster together or show a pull toward another population, they share ancestry (through common descent or admixture).

No! This is your mistake, shared ancestry is just one possible reason why populations will get similar f3 scores.

In EDF 5a,b LB is firmly part of the West Eurasian cluster on the Y axis. You would't question it right? Or you would dare to say that axis Y in EDF 5a,b only measures LB to Sardinian but not to French or Russian?

I would absolutely say that - EDF a/b's Y-axes only measures affinity to Sardinians, not to Russians or French. In this case I don't question that LB's similar result to other Europeans is due to shared ancestry because this is confirmed by all the other data and fits with the geographic and temporal locations of the samples.

Now, I want all of your remaining brainpower to focus on one simple truth: On EDF 5d LB would've been part of the Amerindian cluster (LB 0.19 and Amerindians range from 0.185 to 0.195) if only Mal'ta, X, Yoruba values were plotted.

My remaining brainpower also notes that Oceanians would've been smack bang in the middle of the Middle East range, and that there are a host of similar non-sensical associations if you do the same thing with other axes in the other graphs. This is not a valid way to interpret the data.

This proves that LB has Amerindian and Mal'ta affinity.

No. It proves that LB and Amerindians both have a similar amount of Mal'ta affinity. If you only consider this single axis of this single graph then you might hypothesise that this is due to shared ancestry, but such a hypothesis is explicitly rejected when you include all the other available data - like graph a) which measures LB/Amerindian affinity directly.

The reason EDF a, b are less representative in this regard is simply because LB and MA-1 are ancient samples and Amerindians are modern samples.

Yet this age difference doesn't affect the other modern samples in the graphs?

So what is your profile now? Whistleblowing, heckling, exhibiting bias, practicing intellectual dishonesty... What's next? Just plea guilty and I'll erase all of this from your record. Deal?

More ad hominen, ignored.

Duh, you've just provided evidence for what you thought was impossible. When measured against Karitiana, East Asians and MA-1 are indeed close partners. And lo and behold, both are located in East Eurasia between the New World and West Eurasia.

East Asians and MA-1 are not "close partners" at all, if you think graph e) proves otherwise then your method of interpretation is flawed - you're failing to consider any of the alternative possibilities that could lead to two unrelated populations being a similar distance to a third population.

eurologist said...

There is ample evidence from Northern Europe, Siberia, NE Asia, N America, S America, and even S Africa that extant skin color is strongly regulated by latitude (or, more accurately, by UV exposure).

Classic historical texts from more than 2,000 years ago support this assessment.

It would require incredible new evidence and argumentation to dispute this, or to argue that things were largely different 5,000 or 10,000 years ago -except for some rapid migrations that found people with a different skin color at a place they didn't originally belong to.

German Dziebel said...

@Tobus

"No! This is your mistake, shared ancestry is just one possible reason why populations will get similar f3 scores. "

It's common descent or admixture. There's no other reason.

"If you only consider this single axis of this single graph then you might hypothesise that this is due to shared ancestry, but such a hypothesis is explicitly rejected when you include all the other available data - like graph a) which measures LB/Amerindian affinity directly."

No, it's not. EDF 5a,b are consistent with the inference of shared ancestry between MA-1, LB and Amerindians. And in the absence of actual F3 values, which you promised to provide but failed to do so, it's hard to say whether they add additional weight to the data in EDF 5d but they don't contradict it. And if you put EDF 5d in the broader context of evidence Amerindian admixture in West Eurasia, you have a perfectly proven connection. In one of our earlier exchanges, you mentioned that LB is an exception to this regularity, but EDF 5d shows that it's not.

"I would absolutely say that - EDF a/b's Y-axes only measures affinity to Sardinians, not to Russians or French."

I think you've managed to come up with another pseudoscientific antic, namely intellectual suicide. I have to give it to you - you are very creative (and creationist) with your ways of denying scientific evidence.

"In this case I don't question that LB's similar result to other Europeans is due to shared ancestry because this is confirmed by all the other data and fits with the geographic and temporal locations of the samples."

Amerindian admixture in Europe has been ascertained in a number of studies. MA-1 and LB don't share the same location or the temporal horizon but they share common ancestry. In EDF 5 d LN falls in the Amerindian cluster on axis X just like it falls on the European cluster on axis Y in EDF 5a,b. LB is very clearly admixed population.

"Oceanians would've been smack bang in the middle of the Middle East range and that there are a host of similar non-sensical associations if you do the same thing with other axes in the other graphs."

For you real data is a "non-sensical association," while your own bias and mental laziness apparently provide you with a "magnetic north." Oceanians and East Asians are indeed shifted toward West Asians and Africans. This can be found on many plots and has to be explained, not dismissed.

"Yet this age difference doesn't affect the other modern samples in the graphs?"

Not true. We do know from Rasmussen et al. that Anzick is closer to MA-1 than modern Amerindians. And LB is closer to modern Europeans than MA-1 because, as you yourself argued a while back, LB is a younger sample than MA-1.

"you're failing to consider any of the alternative possibilities that could lead to two unrelated populations being a similar distance to a third population."

Don't play hide and seek with me - what are those "alternative possibilities"? Convergence?

"More ad hominen."

No, just facts. So, here's your updated tally: whistleblowing, heckling, exhibiting bias, intellectual dishonesty, intellectual suicide, bailing on your own earlier claims. Keep them coming.

German Dziebel said...

@Tobus

"It proves that LB and Amerindians both have a similar amount of Mal'ta affinity."

There was a day, about a month ago, when you vehemently denied this. So we are making progress. Let's take another baby step and slowly learn how to walk without leaning on a bias. This affinity is through MA-1's Amerindian component (ORANGE in Raghavan), not through MA-1 West Eurasian component (BLUE in Raghavan). That's why on EDF 5d LB is pulled out of the West Eurasian cluster in the direction of Amerindians on the MA-1 ascertained axis.

So effectively MA-1 and LB both have an Amerindian component or, let's be honest with ourselves, Amerindian ancestry.

Tobus said...

@eurologist: It would require incredible new evidence and argumentation to dispute this

We have incredible new evidence - 2 Europeans from 7500 years ago and a Siberian from 24,000 years ago, none of whom had the known skin lightening alleles that are common today.

I personally don't dispute that adaptation to UV generates the skin colour distribution we see today, but I dispute that a population will automatically become light just by moving to a lower UV area. As well as the low UV the population also has to receive or generate de novo a genetic mutation that lightens skin colour which can then be subject to selection by the environment - without such a mutation there's no way the skin can change colour, regardless of the UV level.

Part of the accepted UV theory is an assumption that these mutations are common and that skin colour will change almost immediately upon entering a different UV environment. But despite at least 40-50kya of modern humans in low UV environments we only have evidence for light skin evolving just twice, and both of these in the last 10 or at most 20kya... perhaps functional light skin mutations aren't so common - once they occur they will spread rapidly to match the UV, but until they do the population will stay the same colour.

That at least is a theory that can explain today's high correlation with UV and also the lack of such correlation in the Mal'ta, La Brana and Loschbour samples.

«Oldest ‹Older   1 – 200 of 338   Newer› Newest»