June 28, 2012

Mesolithic Iberians (La Braña-Arintero) not ancestors of modern ones

From the press release:
A team of scientists, led by researcher Carles Lalueza-Fox from CSIC (Spanish National Research Council), has recovered - for the first time in history - part of the genome of two individuals living in the Mesolithic Period, 7000 years ago. Remains have been found at La Braña-Arintero site, located at Valdelugueros (León), Spain. The study results, published in the Current Biology magazine, indicate that current Iberian populations don't come from these groups genetically.
This appears quite consistent with my model of mostly recent origins of European populations from a West Asian womb of nations. I can't wait to get my hands on this new data.

From the paper:
In the genomic analysis, it is interesting to see that the La Braña individuals do not cluster with modern populations from Southern Europe, including those from the Iberian Peninsula. The first PC separates a north-south distribution, whereas the second follows a general east-west pattern in modern Europeans. The position of La Braña individuals in the 1000 Genomes Project data and the 1KGPomnichip PCAs suggests that the uniform Mesolithic substrate could be related to modern Northern European populations but may represent a gene pool that is no longer present in contemporary Southern European populations. In the latter PCA, where the origin of each Iberian sample is known, it is possible to see that the Mesolithic specimens are not related to modern Basques, contrary to what has been previously suggested in some recent studies [39].
The global PCA of the two individuals shows a clear shift relative to extant Europeans.

So while they are more related to Northern than to Southern Europeans, they are well outside the range of modern European variation. Indeed, there is a strong hint of "Asian-shift" to these individuals. This is completely consistent with the pattern in modern West Eurasian populations. As I noted:
With respect to the Asian- and African- shift of West Eurasian populations, I note that northern Europeans (and Basques) are less African-shifted than southern Europeans, and, at the same time they are more Asian-shifted: the 16 least Asian-shifted populations have a coastline in the Mediterranean (excluding the Portuguese), while the 16 least African-shifted populations do not (excluding the French).
It now appears clear that the Mesolithic substratum in Europe was:

  1. Well outside the modern range, contributing a little to extant populations
  2. Its contribution in northern populations was higher than in southern ones
  3. It may be responsible for the pattern of Asian-shift observed for non-Mediterranean European populations
Related: Coverage in Science.


UPDATE: Due to the small number of SNPs, I pooled the two Mesolithic individuals into a single composite one; the K7b admixture proportions are: 9.3% African and 90.7% Atlantic_Baltic, which appears consistent with the position of the individuals in the European PCA plot. The sub-1,000 SNPs in common with the K7b do not give me a lot of confidence in the minority element, but, in any case, the high Atlantic_Baltic figure is what I would expect and appears consistent with the similarly high Atlantic_Baltic figure of the Swedish Neolithic hunter-gatherers.


UPDATE II: Using the K12b, the results are: 45% Atlantic_Med, 41.6% North_European, 10.3% East_African, 1% Sub_Saharan.


UPDATE III: In terms of the euro7 calculator, the results are: 89.6% Northwestern, 1.6% Southeastern, and 8.7% Far_Asian.

Current Biology, 28 June 2012 doi: 10.1016/j.cub.2012.06.005

Genomic Affinities of Two 7,000-Year-Old Iberian Hunter-Gatherers

Federico Sánchez-Quinto, Hannes Schroeder, Oscar Ramirez, María C. Ávila-Arcos, Marc Pybus, Iñigo Olalde, Amhed M.V. Velazquez, María Encina Prada Marcos, Julio Manuel Vidal Encinas, Jaume Bertranpetit, Ludovic Orlando, M. Thomas P. Gilbert, Carles Lalueza-Fox


Highlights

  • The first complete Mesolithic mtDNA genome retrieved 
  • There is a remarkable genetic uniformity in Europe during the Mesolithic period 
  • Modern Iberians are not direct descendants of the 7,000-year-old hunter-gatherers 
  • Genetic discontinuity between Mesolithic/Neolithic populations supported by simulations 



Summary The genetic background of the European Mesolithic and the extent of population replacement during the Neolithic [1,2,3,4,5,6,7,8,9,10] is poorly understood, both due to the scarcity of human remains from that period [11,12,13,14,15,16,17,18] and the inherent methodological difficulties of ancient DNA research. However, advances in sequencing technologies are both increasing data yields and providing supporting evidence for data authenticity, such as nucleotide misincorporation patterns [19,20,21,22]. We use these methods to characterize both the mitochondrial DNA genome and generate shotgun genomic data from two exceptionally well-preserved 7,000-year-old Mesolithic individuals from La Braña-Arintero site in León (Northwestern Spain) [23]. 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. This suggests a remarkable genetic uniformity and little phylogeographic structure over a large geographic area of the pre-Neolithic populations. Using Approximate Bayesian Computation, a model of genetic continuity from Mesolithic to Neolithic populations is poorly supported. 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. 

Link

53 comments:

truth said...

It is wrong to say modern iberians are not descendants of these mesolithics, considering Iberians have about 40% of northern-european with consistency on admixture runs, it is clear that there is Mesolithic ancestry, albeit not as high as in Northern Europe.

Dienekes said...

Equating "Northern_European" with "Mesolithic" is wrong; both the Iberian Mesolithic and Swedish Neolithic hunter-gatherers were outside the range of modern variation.

Clearly there is _some_ continuity, since mtDNA haplogroup U still exists in Europe, but this is the minority element everywhere, although higher in Northeastern Europe. European foragers have been mostly replaced, but to a smaller degree in Northeastern Europe.

In the case of Iberia we must be doubly cautious because the "Atlantic_Baltic" element may in part reflect Mesolithic populations, and in part later Celto-Germanic intrusions.

truth said...

ok thank for the answer. Anywways judging from the position in the global PCA, I bet he is about 6% SSA, 15% Asian and the 78% restant mostly Northeast Euro, maybe this could be due to the low SNPs, like the SSA that appeared on the Swedish HG's.

jeanlohizun said...

Dienekes said:

Equating "Northern_European" with "Mesolithic" is wrong; both the Iberian Mesolithic and Swedish Neolithic hunter-gatherers were outside the range of modern variation.

Of course, they are going to be outside of the range of modern variation, there are 7000 years worth of genetic drifting between these Hunter Gatherers and modern Europeans, so that is something expected, but in Figure-S3

http://i1133.photobucket.com/albums/m582/jeanlohizun/Sanchez-Quintoetal2012-Figure-S3.jpg

We see that Finns (Northeastern Europeans) are as equally removed, or even more removed from the Mesolithic Iberians as the modern day Iberians are, so I’m not sure how that fits with a greater survival of a Mesolithic substratum in Northeastern Europe. The NW Europeans appear to have the closest affinity to these samples thus far, and they aren’t “heavy” on haplogroup U5 last I checked.

Dienekes said...

Of course, they are going to be outside of the range of modern variation, there are 7000 years worth of genetic drifting between these Hunter Gatherers and modern Europeans, so that is something expected, but in Figure-S3

Both Oetzi and Gok4 are not outside the range of modern variation though. Drift doesn't cut it as an explanation.

We see that Finns (Northeastern Europeans) are as equally removed, or even more removed from the Mesolithic Iberians as the modern day Iberians are

Finns are removed because of their ~10% East Eurasian admixture.

jeanlohizun said...

Dienekes said:

Both Oetzi and Gok4 are not outside the range of modern variation though. Drift doesn't cut it as an explanation.

Oetzi is indeed a bit outside of modern variation, see here, of course he is not as outside as these Mesolithic folks:

http://www.nature.com/ncomms/journal/v3/n2/fig_tab/ncomms1701_F3.html

Gok4 is indeed inside modern variation:

http://1.bp.blogspot.com/-rl1bw36vh5E/T5WCj36-KyI/AAAAAAAAEyQ/IgbXJAn3FPo/s1600/skoglund.png

Either one though are still 2000 years younger than these two folks.

Dienekes said:

Finns are removed because of their ~10% East Eurasian admixture.

But aren’t the HG also Asian shifted in the plot shown on Figure-S2, in any case, I would expect the Finns to be closer due to this shift.

Dienekes said...

Oetzi clusters with modern Sardinians remarkably well given that he's 5,000 years removed from them.

Gok4 clusters with modern Southern Europeans as well, and he's contemporaneous to the Ajv hunter-gatherers who are outside the modern European range.

Brana clearly is well outside modern European variation.

All signs point to massive upheavals in European prehistory: Gok4 is more like Southern Europeans, and Brana 1/2 are more like Northern Europeans; that's the opposite of geography.

Also, all Europeans so far seem to lack the West_Asian element that must have been added late in prehistory.

Anonymous said...

I wonder what to expect when we finally have good Y-DNA data on samples like these. Hg "I" ? If it's something else, it would be an interesting surprise, given recent developments.

princenuadha said...

>Well outside the modern range, contributing a little to extant populations

Then why is brana extremely close to some Northwest Europeans on the first map?

The paper even said "the uniform Mesolithic substrate could be related to modern Northern European populations"

As for the African vs Asian shift how reliable is that when it when it gives a different picture? We are talking about ancient DNA and comparing it to modern populations that drift. And why is brana "Asian shifted" when he shows 9% SS African?

>Equating "Northern_European" with "Mesolithic" is wrong; both the Iberian Mesolithic and Swedish Neolithic hunter-gatherers were outside the range of modern variation.

"Northern European" doesn't explain the variation of any modern European population either, as none of them are 100%.

That's off point anyways. How how much E. meso decent do modern European have? The Swedish stuffy suggested that North European were mostly Meso in decent. http://www.evoandproud.blogspot.com/2012/05/who-were-ancestors-of-modern-europeans.html?m=1

The European only maps in this study also seem to support significant meso in northern Europeans.

jackson_montgomery_devoni said...

Wow both samples are U5b2c1! I find this to be quite amazing since I am myself also U5b2c. Well I am U5b2c2 while they are U5b2c1 but we are both still U5b2c which is very cool! :)

Dienekes said...

Then why is brana extremely close to some Northwest Europeans on the first map?

Because they're not Finnish and hence cluster with unmixed Europeans.

As for the African vs Asian shift how reliable is that when it when it gives a different picture? We are talking about ancient DNA and comparing it to modern populations that drift. And why is brana "Asian shifted" when he shows 9% SS African?

Atlantic_Baltic is shifted towards Asians.

http://dodecad.blogspot.gr/2012/01/k12b-and-k7b-calculators.html

So, basically you had a Neolithic intrusion by "Southern" that is much more distant to Asians than Atlantic_Baltic is.

That's off point anyways. How how much E. meso decent do modern European have? The Swedish stuffy suggested that North European were mostly Meso in decent. http://www.evoandproud.blogspot.com/2012/05/who-were-ancestors-of-modern-europeans.html?m=1

That is inaccurate. The Swedish study never quantified the Neolithic hunter-gatherer ancestry of Europeans, only their Gok4 related ancestry. The latter was not quantified vis a vis the ancient hunter-gatherers, but using a particular model that presumed that farmers were a sister clade of Druze and h/g were a sister clade of Finns.

princenuadha said...

"Because they're not Finnish and hence cluster with unmixed Europeans."

Fair enough, but the study did say that the mesolithic groups may be related to the modern north Europeans. though, I don't know exactly what that means.

"Atlantic_Baltic is shifted towards Asians."

I am positive that the makeup of the Lithuanians is more Asian shifted than a 91% "Atlantic-Baltic" 9% "SS African" composite. Both AB and WA are very far from SSA.

That makes it look like the global estimates aren't reliable.

Unknown said...

They dont look Iberian thats for sure but they cluster darned close to the Brits. Not bad for 7,000 years. La Brana is NW Spain, an area with long-term cultural connections to France, Britain and Ireland. The Irish even quote Portugal-ish as their genetic homeland

The nice pure global PCA plots you show do not fit with modern European reality either. They look like Northern Europeans to me.

Didn't we already know about this La Brana U5b2c1 ?

German Dziebel said...

"This appears quite consistent with my model of mostly recent origins of European populations from a West Asian womb of nations."

"Indeed, there is a strong hint of "Asian-shift" to these individuals."

The Asian shift doesn't mean it's a West Asian shift. Unless I missed the data showing that. But I think you're just fishing for data supporting the idea of Indo-European origin from West Asia. The Asian shift means rather the East Eurasian shift. U5b2c1 is closely related to the U5a and U5b found at high frequencies in Saami.

Dienekes said...

The Asian shift doesn't mean it's a West Asian shift.

Who said it's a West Asian shift? Please don't project your delusions onto my writings.

But I think you're just fishing for data supporting the idea of Indo-European origin from West Asia.

That is irrelevant; actually, the fact that the "West_Asian" component is lacking in all tested European individuals down to 5,000 years ago is quite consistent with IE origins in West Asia.

Anonymous said...

The PCAs drawn by Dienekes show the ancients as being closer to modern sub Saharan Africans and modern Japanese and Chinese than any of the European groups used in the PCA. Modern North Europeans have low sub Saharan ancestry indicators, and modern South European populations have low East Asian ancestry indicators. Seems like neither European groups are close to these ancients. Modern British populations have more East Asian ancestry and sub Saharan ancestry indicators than other Northwest Europeans. It is no wonder the ancients are "drawn" to them. It is the minor ancestry in the British that is the reason for the apparent closeness.

mtDNA U haplogroup spread out of the Mediterranean zone after the LGM and Holocene epoch. It is not particularly Northern European, just where it has high spots as in the Saami, but using frequency as indicators of origin is rather simplistic as in the saga over R1b.

princenuadha said...

BTW

17.3 is the k7 distance between "African" and "Atlantic-Baltic"

So the distance from "African" to Brana, 90.7% "Atlantic-Baltic" 9.3% "African", on the k7 map is.

17.3 x .907 = 15.69 [I hope the math is this simple]

Besides "African" itself, 15.69 is a smaller distance than any cluster in k7. So a person who is 100% "West Asian" is less African than Brana.

Also, any person who has a trivial amount of "African" is less "African" than Brana.

Davidski said...

The Finns aren't shifted to the left on that PCA due to North Asian admix. They're shifted because they're drifted. If there were as many Ashkenazim on that PCA, they'd be shifted in the same way.

So that's a really poor PCA in the context of the study. They're basically looking at Finnish genetic drift, rather than the Mesolithic samples.

In order to do these samples justice, they should be compared to a few Finns, plus Lithuanians. Oh, and Poles. :)

German Dziebel said...

"This appears quite consistent with my model of mostly recent origins of European populations from a West Asian womb of nations."

"Indeed, there is a strong hint of "Asian-shift" to these individuals."


You wrote both of these statements in the same post. Try to read the whole thing over before posting.
But the main point is that U5 has highest frequencies in modern Saami, not in any West Asian population. There's nothing in the study that's "consistent" with the origin of modern Europeans from "West Asia." It's your wishful thinking.

GailT said...

The sample appears to be U5b2c1 with no additional mutations reported in Table S2. We have 4 FGS test results in the U5 project for U5b2c1 and 1 partial result from Hernstadt. One person in the project has ancestry in Spain and also has no extra mutations, while the other 2 have ancestry in Ireland, and they have some extra mutations. So this seems to confirm a Spain-Ireland connection in the mtDNA.

I get an age estimate of about 5200 years for U5b2c1 (compared to 4200 in Behar et al), which is younger than the remains that are 7,000 years old. However, my estimate is based on a small sample size and has large uncertainty. The test result is very useful for an independent estimate of the age of U5b2c1, and it confirms a very old date for this group and also confirms that it was present in Europe during the late Mesolithic.

One of the U5b2c1 project members has a back mutation at 16270, so their HVR1 result is simply 16192T. Also, 16192 is a very frequent mutation site in U5, so this raises the possibility that CRS test results could be either U5b2a1 or U5b2c1.

GailT said...

U5b2c1 is definitely NOT closely related to the Saami signature U5b1b1a. Their common ancestor U5b lived more than 20,000 years ago.

You really cannot make simple generalizations about U5a and U5b, and we really need full genome tests in most cases of ancient mtDNA to sort out their possible connections to current populations. Kudos to the authors for doing the extra work on the full sequence - it would be great if other researchers would revisit their previously published HVR test results of ancient remains.

GailT said...

For comparison with other U5b2c in the FTDNA U5 project, we have one U5b2c* with ancestry in Ireland, we have two U5b2c2 with ancestry in Ireland and Scotland.

We also have 5 people in "U5b2c2 Group A" (defined by 16249C) with ancestry in: Scotland, Sweden, Ireland, North Ireland and England. This "Group A" also appears to be quite old with an average of 2.2 extra mutations in addition to the mutation at 16249.

U5b2c has been estimated to be some 8,000 years older than U5b2c1, so there is a very long history of U5b2c in Europe before the people in La Braña 7000 years ago. The simplest story would be an Iberian ice age refuge followed by a very early expansion through Spain to England and Ireland. But that migration could have happened very long ago, perhaps with the first people to repopulate England and Ireland after the glacial maximum. I would think this would be to ancient to reflect any cultural stories of Irish origins in Iberia. Those legends might reflect more recent population movements.

The very old age estimates for U5b2c and its diversity in the UK, Ireland and Sweden could leave open the possibility of other migration routes for early U5b2c to arrive in England and Ireland. It could even be possible that the La Brana individual arrived in Spain from Ireland after U5b2c was established in the UK. Not that I think this is likely, but I do think we need to consider the possibility of very complex migrations over a period of many thousands of years.

Onur Dincer said...

Pre-Neolithic Europeans might have been Caucasoids with varying amounts of Australoid-like admixture. This not only explains the results of this study but also the Australoid-like physical features seen in some hunter-gatherer European skeletons.

Fanty said...

Hm.

From the experiences, I have seen on a series of MDS experiments from Davidski, I would say, its absolutely important what populations are on the MDS.

I am sure, a single Pole or Russian would cluster with the CEU and UK on the first MDS, just because those are the best choice between Finns and Spanish.

And for things "outside of the range" I recall that freak results of "ADMIXTURE" when Davidski "forced" North-Africans and West-Asians to decide if they are rather Ireland based "Western European", Sweden based "Northsea", Lithuania based "Southbaltic" and I think "Chuvash". (all components for wich non-Europeans should be "out of the range")

Northafricans, Arabs, Iraq and Turks, all decided to be like 90-100% North Sea (and 10% Westeuropean). Iranians werent sure if they are Northsea or Chuvash (50% Northsea and 50% Chuvash) but for Indians it was clearly 100% Chuvash.

Not for a second, anyone outside of Europe thought he could even be a tiny but "Southbaltic".

Wich maded David to think, "Southbaltic" beeing the best candidate for European Mesos.

"Northsea" beeing the most cosmopolitican component he explained by this component possibly originating in central Europe and beeing in the center of all kind of migration routes.

Well maybe, but atleast results like this, should us make carefull of taking some things too literate.

Dienekes said...

The Finns aren't shifted to the left on that PCA due to North Asian admix.

It's not an "either or". The Finns differ from other Europeans both because they are drifted and because they carry East Eurasian admixture. I do agree that it would be better if more populations were used, however.

You wrote both of these statements in the same post. Try to read the whole thing over before posting.

Thank you for your suggestion, but you should really follow your own advice: if I write "Asian" and "West Asian", chances are I mean something different. And, if you had bothered to read the original post in which I detected the Asian shift in non-Mediterranean populations, you'd know what kind of "Asian" I meant.

jeanlohizun said...

Dienekes said:
Oetzi clusters with modern Sardinians remarkably well given that he's 5,000 years removed from them.

In figure-3a one can see that Oetzi is indeed outside of the European variation.

http://www.nature.com/ncomms/journal/v3/n2/fig_tab/ncomms1701_F3.html

Dienekes said:
Gok4 clusters with modern Southern Europeans as well, and he's contemporaneous to the Ajv hunter-gatherers who are outside the modern European range.

Well, yes Gok4 clusters with modern Southern Europeans, when only Europeans and West Asians are plotted.

http://1.bp.blogspot.com/-rl1bw36vh5E/T5WCj36-KyI/AAAAAAAAEyQ/IgbXJAn3FPo/s1600/skoglund.png

In fact, when only Europeans are used the La Braña individuals cluster near British samples, which is what we observed in Figure-3, and Figure-3S. In fact, they mention it in the study:

Sanchez.Quito.et.al.2012

A worldwide genomic principal component analysis (PCA) with data from the 1000 Genomes Project [31] places La Braña 1 and 2 near, but not within the variation of current European populations (Figure S2). However, when compared exclusively to European populations, La Braña 1 and 2 fall closer to Northern European populations such as CEU and Great Britons than Southern European groups such as Iberians or Tuscans (Figure 3). With 1KGPomni chip [31] data, the PCA generates a similar pattern (Figure S3), although the general geographic structure is less clear because of the limited number of SNPs (see Supplemental Experimental Procedures).


So neither Gok4, nor Oetzi where plotted in worldwide genomic PCA with data from the 1000 Genomes, so how do we know if either one would shifts towards the East Asians or Africans if neither East Asians nor Africans were included in the plots that featured Gok4 or Oetzi. After all, the La Braña individuals cluster within the European variation when compared with other Europeans only, they are to the UK, what Oetzi is to Sardinians.

Dienekes said...

In figure-3a one can see that Oetzi is indeed outside of the European variation.

http://www.nature.com/ncomms/journal/v3/n2/fig_tab/ncomms1701_F3.html


That is incorrect, Oetzi clusters with Sardinians who are, last time I checked, Europeans. He clusters more specifically with Europeans (Figure 3a).

So neither Gok4, nor Oetzi where plotted in worldwide genomic PCA with data from the 1000 Genomes, so how do we know if either one would shifts towards the East Asians or Africans if neither East Asians nor Africans were included in the plots that featured Gok4 or Oetzi.

Easily enough:


Oetzi (148345 SNPs): http://i46.tinypic.com/i2t3wp.png
Gok4 (14574 SNPs): http://i50.tinypic.com/fkvccp.png

It does not appear that Oetzi and Gok4 are shifted towards Asians/Africans, rather the opposite.

Fanty said...

What about the Swedish HGs with Africa and Asia? Or have we seen such already?

Dienekes said...

http://i46.tinypic.com/2zea2it.png

Onur Dincer said...

Dieneke, have you done your own global PCA analysis of the La Braña samples?

BTW, your global PCA analyses of the ancient samples would be better if you included more reference populations (especially Caucasoid populations).

German Dziebel said...

@GailT

"U5b2c1 is definitely NOT closely related to the Saami signature U5b1b1a..."

Well, it is. In comparison to other non-U5b haplogroups. "Close" is a relative term. What's important is that U5b is found at 4% in modern European populations, but at close to 50% in Saami (other western Uralic populations have elevated frequencies of U5b, too, with Karelians at 13%). This makes La Braña-Arintero individuals more like Saami, provided, of course, the western Euralic excess in U5b lineages wasn't caused by a bottleneck that drove the frequencies up from the European average of 4%.

I wouldn't trust the dates derived from the molecular clock.

GailT said...


Well, it is. In comparison to other non-U5b haplogroups. "Close" is a relative term.


@German Dziebel

Sure, all U5b are closely related in the context of the modern human population as a whole, but not close at all in the context you discussed, the origins of modern Europeans - you wrote:

...the idea of Indo-European origin from West Asia. The Asian shift means rather the East Eurasian shift. U5b2c1 is closely related to the U5a and U5b found at high frequencies in Saami.

The authors of the paper make the same mistake in arguing that the commonality of U5b across early European hunter-gatherer populations is evidence of a cultural link between these peoples. And they have been criticized for this overreaching conclusion.

The link between U5b daughter groups U5b1 and U5b2 was probably around the time of the last glacial maximum, and their expansion to Europe could have originated in refuge populations ranging from Iberia to Ukraine with very different migration histories. So you really cannot use U5b as a evidence of common connections in autosomal DNA among present day European populations.

I wouldn't trust the dates derived from the molecular clock.

That is why the full genome sequence of this sample is so important - it gives us an independent age estimate for U5b2c1 that is older than the mean estimate calculated using the molecular clock and is at the very high end of the uncertainty range. Of course we need more full sequence testing of ancient remains to confirm, but this suggests that current molecular clocks could underestimate the age of mtDNA haplogroups.

All ancient remains should be tested for the mtDNA full genome so that we can refine the molecular clock and so that we can identify real connections among ancient populations.

jeanlohizun said...

Dienekes said:

That is incorrect, Oetzi clusters with Sardinians who are, last time I checked, Europeans. He clusters more specifically with Europeans (Figure 3a).


http://www.nature.com/ncomms/journal/v3/n2/fig_tab/ncomms1701_F3.html


Well, the way I see it, is that Oetzi cluster right outside the European cluster in Figure-3a, but whatever man.

Now, Figure-3b shows Oetzi with respect to Europeans only, would you care to argue how is it that the Swedish Hunter Gatherers do not cluster with Europeans, whereas Oetzi does:


http://1.bp.blogspot.com/-rl1bw36vh5E/T5WCj36-KyI/AAAAAAAAEyQ/IgbXJAn3FPo/s1600/skoglund.png


I mean, if we are to consider the position where Oetzi clusters as being within Europeans, then it is clear than the Swedish HG samples do cluster within the European variation too.


As for your Global PCA plots, Onur beat me to it,you ought to use more European populations in order to recreate the plots of the study.

Eduardo Pinto said...

Hello Dienekes


Don't forget to run the Brana composite with Euro7 calculator.

Unknown said...

Can it be assumed that these guys (brothers?) froze to death? They were crouching, which is what makes me think this. But I suppose they could have been arreanged in the foetal position as has happened in some burials. Is there any evidence of death ritual?

Anonymous said...

I can't wait for true Mesolithic European remains are dna mined to see the results. At the moment everyone is pussyfooting with European remains from the interface between the hunter/gatherers and the farmers in the post Holocene epoch, Neolithic period, about 7,000 years ago. The true test would be using ancient dna from sites in Europe proven to be pre Holocene, 12,000 years old. As I said, the pussyfooting just leaves lots of questions regarding how native European were these people, when did they arrive in Europe, where did they come from, and the results with such poor dna retrieval from the remains are spurious. As far as we know the mtDNA U5 in Europe is 9,500 years old, Cheddar Man. Why not retrieve more of his dna after all his skeleton was complete, lots of teeth? I have had enough of this piecemeal and poor scientific approach to find out the peopling of Europe. Remember all those tags, North European, Atlantic Baltic etc are just artifacts of those Admixture programs, and not truly reflecting the reality of the Mesolithic, Epipaleolithic and Neolithic cultural phases in Europe.

Fanty said...

"Can it be assumed that these guys (brothers?) froze to death?"

Cant imagine this.

1. Caves tend to have a year round stabile temperature. In modern Spain this is suposedly around 17 degress Celsius all 365 day the year.

2. They are suposed to be "Hunter-Gatherer". I rate the ability of a hunter-gatherer to survive in the wilderness by far higher than that of a farmer.

So, guys whos everyday life is to survive with what can be found in the wilderness, in a cave, in Spain, freezing to death like a new yorker, without a cellphone, lost in the forrest at night... no way.

My bet goes for a burrial in a sleeping child position. Back in momies (earth) belly or something like this.

Maju said...

Wow, Dienekes, those Updated zombie analysis are most interesting! It means that, provided we can assume that these people are true Epipaleolithics (it's probable considering some contextual clues and their mountain habitat, not likely for first farmers, but by no means 100% certain), that your "Atlantic-Mediterranean" (Southwestern) component is pre-Neolithic.

Not just that but North African blood was also there before Neolithic, as I speculated in relation with South and West Iberian Solutrean and related North African Oranian.

Adding that to the several mtDNA H findings of the last months, even in Magdalenian contexts, we can really conclude a good deal of pre-Neolithic continuity.

Good work! I'm liking your K12 analysis: it is what was needed in this case, much better, much more informative than the messy PCAs of the paper. Thanks again.

truth said...

Fanty, no way 17 degrees all year round, In León, were they were found, in modern winter the average is about 2-3 degrees, and snowfalls are quite common.

Dienekes said...
This comment has been removed by the author.
Dienekes said...

Wow, Dienekes, those Updated zombie analysis are most interesting! It means that, provided we can assume that these people are true Epipaleolithics (it's probable considering some contextual clues and their mountain habitat, not likely for first farmers, but by no means 100% certain), that your "Atlantic-Mediterranean" (Southwestern) component is pre-Neolithic.

This certainly does NOT mean that the "Atlantic_Med" component is pre-Neolithic.

Even if there was virtually no pre-Neolithic ancestry in modern Europeans, the algorithm would still pick a bunch of numbers that add up to 100% and would prefer clusters that are related to the little pre-Neolithic that there is.

It's like having a scale that goes from white to a very light pink and then testing a bright red with it. Of course the algorithm will call the "bright red" 100% pink, even though the pink is actually composed of white with a very little red thrown in.

Adding that to the several mtDNA H findings of the last months, even in Magdalenian contexts, we can really conclude a good deal of pre-Neolithic continuity.

There is no H in well-dated and well-described contexts in pre-Neolithic Europe, and I doubt there will be.

Dienekes said...

Don't forget to run the Brana composite with Euro7 calculator.

See updated post.

Fanty said...

@Truth:

Thats why I mentioned, they had been in a CAVE.

Caves have the effect of stabile temperatures, after some distance to the entrance.

I googled it and found that German caves are +8 degrees all around the year. even if its -20 degrees right in front of the entrance.

Then I googled caves in Spain and found that they have a "constant +17 degrees all seasons"

Means, even if there is snow outside.

Iglus (artificial man-made snow caves that Inuit hunter gatherers build if they cant reach the base camp in time. And they build them in hours) for example reach between -5 and 0 degrees without a person and +15 degrees with 2 persons if its -30 degrees cold outside of it.

I cant imagine 2 Greenlandish HGs digging themselfs into the snow and survive the nights while on a 2 weeks hunting trip in the polar ice, while 2 Spanish HGs freeze to death in a cave without a glacial maximum around and the polar ice in northern Italy or so. Hehehe

Maju said...

There are two mtDNA H (H* and H6) in clearly Magdalenian contexts in two different sites of Cantabria, tested for AluI: Hervella 2012.

I personally asked Montserrat Hervella and her team chief, Concepción de la Rúa, if they tested for that, because some people (including myself) had doubts and the answer was a double and rotund YES, we tested for AluI.

Also the distinct Epipaleolithic burial site of Linatzeta (or Linazeta) is well dated, whatever you say. On human bone itself: 7315±35 BP (radio-carbon, there are two more dates on charcoal) - ref.

The only blind people are those who don't wish to see.

And sure: your own Dodecad results are an illusion, c'mon! Go tell that to your Dodecad affiliates... all is a game of mirrors and what not! Sorry but the division of the K7 Atlantic-Baltic into K12 (North European + Atlantic-Med) happens all the time, regularly: it's almost an exact equation and happens here too, exactly as a expected.

You can also try to do direct comparisons instead of using zombies... if it is an artifact it will become apparent but so far it seems not.

The only think anomalous (and very much so I admit) is the Eastern African component. That is very intriguing and may lead to a large number of speculations.

truth said...

If you could run it with a more pure mediterranean than Atlanto-Med, like the Med component of the v3 K12, and that would give us a better picture on terms of his european composition, between East-West-Med

Dienekes said...

The only blind people are those who don't wish to see.

I.e., you:

" In the latter PCA, where the origin of each Iberian sample is known, it is possible to see that the Mesolithic specimens are not related to modern Basques, contrary to what has been previously suggested in some recent studies [39]."

And sure: your own Dodecad results are an illusion, c'mon! Go tell that to your Dodecad affiliates... all is a game of mirrors and what not! Sorry but the division of the K7 Atlantic-Baltic into K12 (North European + Atlantic-Med) happens all the time, regularly: it's almost an exact equation and happens here too, exactly as a expected.

Of course my results are not an illusion, what you fail to understand -or perhaps don't want to understand- is that no matter what one puts in the program, it will come up with a set of numbers adding up to 100%. You erroneously interpret this as indicating that "Atlantic_Med" is all pre-Neolithic. Well, I can't help you with your lack of education or common sense, the information is there for anyone who is intellectually honest.

Maju said...

I understand that perfectly Dienekes but it does not come up with gibberish, as one can interpret the East African component. It does not come up with a random split in many components... it comes up precisely with those components that are specifically European.

Otherwise, seriously, do you think that a quote is more important than the factual data? That's scholasticism, which is the opposite of science.

jeanlohizun said...

Dienekes said:


" In the latter PCA, where the origin of each Iberian sample is known, it is possible to see that the Mesolithic specimens are not related to modern Basques, contrary to what has been previously suggested in some recent studies [39]."


A quick look at Figure-3 and Figure-S3 reveals that by related the authors had assume that the Basques would be exact genetic copies of Mesolithic Europeans, or rather that the Basques could be the closest modern Europeans to them.

Nonetheless Figure-S3 which was used by the study to assert the relationship between Basques and Mesolithic Europeans was also said by the authors to yield an unclear geographic structure due to the small number of SNPs, i.e.

Sanchez.Quinto.et.al.2012

A worldwide genomic principal component analysis (PCA) with data from the 1000 Genomes Project [31] places La Braña 1 and 2 near, but not within the variation of current European populations (Figure S2). However, when compared exclusively to European populations, La Braña 1 and 2 fall closer to Northern European populations such as CEU and Great Britons than Southern European groups such as
Iberians or Tuscans (Figure 3). [b]With 1KGPomni chip [31] data, the PCA generates a similar pattern (Figure S3), although the general geographic structure is less clear because of the limited number of SNPs (see Supplemental Experimental Procedures).[/b]



In fact, the distribution (scattered all over the graph instead of plotting in a single spot) of Basques should give clues as to the effect of using a low number of SNP is causing.


Dienekes said:
Of course my results are not an illusion, what you fail to understand -or perhaps don't want to understand- is that no matter what one puts in the program, it will come up with a set of numbers adding up to 100%. You erroneously interpret this as indicating that "Atlantic_Med" is all pre-Neolithic. Well, I can't help you with your lack of education or common sense, the information is there for anyone who is intellectually honest.


Well what do you make of the Northwestern European component being the most akin to describe the genome of these Hunter Gatherers. You said not too long ago that:

“The "Northwestern" component represents the pre-Megalithic first farmers of Northwestern Europe, consisting of Linearbandkeramik farmers emanating from Central Europe and admixing with pre-farming Atlantic hunter-gatherers.



So let me guess, because the Northwestern component has some pre-farming Atlantic HG genes, it probably worked best, right!!! But, if these component is heavy on the farmers gene, as you claimed before that any pre-farming substratum is minimal, how come it fits at 89%. Of course we know that the program has to give the genomes the best fit from a limited number of populations, but wasn’t you who said that they would turn out to be mostly of not fully Northern European in K12b?


Also, how come that when given the choice of only European populations in a PCA analysis, these individuals plot right next to the NW Europeans, why don’t the cluster east of the Finns, or South of the Spaniards, or on the Southeast corner, since their outlier status in the European variation should manifest itself regardless of whether they are plot on a European or Worldwide PCA. After all, don’t Swedish Hunter Gatherers plot “East” of Russians and Finns, even in a European only PCA, thus showing that they had higher Eastern affinities than both, and that even in an intra-European plot they would still appear to be slightly outside of the European variation.

Justin said...

I hope Maju and Dienekes never stop; it's so entertaining.

Anonymous said...

I am not interested in people arguing just to reinforce their belief system.

Those Epipaleolithic Europeans and that Copper Age European are thousands of years removed from modern Europeans. Logically no one in Europe would share any dna with those antique Europeans, after all, most people don't share dna with all their great grandparents other than general ones that are part of the general human genome.

All those studies retrieved less than the optimum amount of dna and the studies using those truncated amounts of dna are dubious. It is logical to compare dna from living humans or recently living humans with each other using varous algorithms to work out relationships and admixture but with humans separated by thousands of years compared to living humans is only going to produce strange results. It is better to compare like with like, that is, living humans with living humans, Epipaleolithic Europeans with other Mesolithic and early Neolithic Europeans or humans from same era and epoch. Forcing antique dna into the modern human pattern can only produce spurious results.

Anders Pålsen said...

Here is another analysis of the La Braña individuals.

http://fennoscandia.blogspot.no/2013/05/la-brana-and-modern-european-variation.html

Anders Pålsen said...

I have narrowed it further down to the Saamis.

http://fennoscandia.blogspot.no/2013/06/la-brana-and-saamis-ii.html

Model on a Bottle (tm) said...

For whatever reason, I have a lot of La-Brana-1. Looking at my Eurogene's only on Gedmatch, it is 25%, the rest is Irish and Spanish Galica. Since I already know which Grandparent that comes from (mostly), I would say all the way back to the Fitzgerald's when they left Italy and came to Ireland. There is an King Niall line in there too so that's some more Irish or Scottish. I have a little Swedish and Finish as well but not sure where that comes from. My Mom has very dark skin and Emerald Green Eyes. My Brother looks like her. If I get any sun at all I start to look like that too, very light eyes, dark skin and my Daughter with her Blonde hair and light eyes would get so dark her classmates would stare at her and ask "What are you"? Here is a copy of the result if it will help.
Gedmatch.Com
Eurogenes K13 4-Ancestors Oracle
Using 2 populations approximation:
1 50% South_Dutch +50% Swedish @ 8.009503


Using 3 populations approximation:
1 50% Irish +25% La_Brana-1 +25% Spanish_Galicia @ 7.315225


Using 4 populations approximation:
1 French + La_Brana-1 + West_Scottish + West_Scottish @ 7.061912
2 Tatar + West_Scottish + West_Scottish + West_Scottish @ 7.076351
3 French + Irish + La_Brana-1 + West_Scottish @ 7.104671
4 French + Irish + Irish + La_Brana-1 @ 7.166987
5 Irish + Tatar + West_Scottish + West_Scottish @ 7.206831
6 French + La_Brana-1 + Orcadian + West_Scottish @ 7.235892
7 La_Brana-1 + Spanish_Galicia + West_Scottish + West_Scottish @ 7.244399
8 Southwest_English + Tatar + West_Scottish + West_Scottish @ 7.247712
9 Irish + La_Brana-1 + Spanish_Galicia + West_Scottish @ 7.271399
10 French + Irish + La_Brana-1 + Orcadian @ 7.290482
11 Orcadian + Tatar + West_Scottish + West_Scottish @ 7.299511
12 Irish + Irish + La_Brana-1 + Spanish_Galicia @ 7.315225
13 French + La_Brana-1 + Southwest_English + West_Scottish @ 7.332187
14 French + North_Swedish + Swedish + West_Scottish @ 7.337672
15 Irish + Irish + Tatar + West_Scottish @ 7.354294
16 French + Irish + North_Swedish + Swedish @ 7.370587
17 French + Irish + La_Brana-1 + Southwest_English @ 7.371592
18 Irish + Southwest_English + Tatar + West_Scottish @ 7.373393
19 Finnish + French + West_Scottish + West_Scottish @ 7.385004
20 La_Brana-1 + Portuguese + West_Scottish + West_Scottish @ 7.387310