September 24, 2012

rolloff analysis of French as a mixture of Sardinian+Burusho

I obtain f3(French; Sardinian, Burusho) = -0.002652 (Z=-13.541) on the basis of 446,917 SNPs. This is the strongest signal of admixture in the French that involves a population that is high on the "West_Asian" component whose influence I have been investigating.

I thus carried out rolloff analysis using the French as a target population and the Sardinians and Burusho as reference populations. The exponential fit can be seen below:

The jackknife gives 239.556 +/- 50.553 generations for this admixture, which corresponds (assuming a generation length of 29 years) to 6,950 +/- 1,470 years.

Analysis of autosomal DNA from the Tyrolean Iceman and a Neolithic TRB farmer from Sweden have revealed an absence of the West Asian ancestral component and a Sardinian-like Neolithic population c. 5ka in Europe. This population may have extended to at least to the Balkans in space and down to the Iron Age in time.

In my opinion, the simplest explanaton for the evidence is that the admixture picked up by rolloff took place in West Asia itself c. 7ka, and then this population begun its movement into Europe at some post-5ka time period.

Importantly, the K=12 Caucasus component appears as a mixture of the K=7 West_Asian and Southern components. The former (West_Asian) is the most important one in the Burusho, and the latter (Southern) is the most important one in Sardinians.

European Neolithic farmers, of presumably West Asian origin only possessed Y-haplogroup G2a out of the wide variety of haplogroups found in West Asia today. They also lacked the West_Asian component which is modal in West Asia today. There is also physical anthropological evidence from Greece and Anatolia, for an introduction of new population elements during the Bronze Age.

These facts combine to make me believe that there were population movements across West Asia which preceded the Indo-European invasion of Europe during late pre-history. That event is then best seen as an extension of a broader Eurasian phenomenon that affected substantially both the western parts of Asia and Europe.

Taking all the evidence into account, I hypothesize that:
  • a "Southern"/"Atlantic_Med"/Sardinian-like population substratum existed in West Asia, and this spawned the early European Neolithic.
  • a new "West_Asian"/Burusho-like population arrived from the east, perhaps associated with the Halaf/Hassuna cultures, or from some other unknown center of dispersal in the Transcaucasus or Iran. Mobility may have been encouraged post-8.2 kiloyear event.
  • these two elements began mixing ~7 thousand years ago in West Asia
  • the admixed population expanded at some post-5ka period into Western Europe.
This scenario is also compatible with the lack of "Southern"/"Atlantic_Med" influences in the Indian subcontinent and Central Asia: if the West_Asian component originated to the east of the Sardinian-like population then it would not have the opportunity to incorporate "Southern" elements in its eastern expansion.

(Obviously, more rolloff analyses are needed to study these ideas; the current one took about ~3 days, which was a little faster than I expected.)

Related (?): Is Burushaski Indo-European?

Image credit: Don Perrault (source)

19 comments:

eurologist said...

In my opinion, the simplest explanaton for the evidence is that the admixture picked up by rolloff took place in West Asia itself c. 7ka, and then this population begun its movement into Europe at some post-5ka time period.

That seems reasonable to me - although I think that both Eastern and Western flows have occurred intermittently for the past ~40,000 years.

Here are some random (and perhaps incorrect) google image returns for Burusho:

http://pastmist.files.wordpress.com/2009/02/hunzagirls_.jpg?w=460&h=345

http://pastmist.files.wordpress.com/2009/02/_pakistan_hunzagilrs.jpg (left)

Burusho aren't really a good proxy for West Asian, IMO.

Crimson Guard said...

Behar has the Burusho mixed with alotve South Indian and a good percentage of East Asian. His light green could be either from the Caucasus or Iran.

Behar:
http://img203.imageshack.us/img203/3720/behar2010k10.png

For central and western Europe, the likelihood is more reasonable to assign to the migrations of Iranian and Turkic tribes that invaded Europe during the final years of the Roman Empire or even to the Germanic or Slavic and mixed Germanic-Slavic tribes which could have mixed with those peoples somewhere in eastern Europe or central Asia(including Russia).

Or we could be left with that plus some intrusion of Gypsy admixture into Europe as they came in large numbers and have good size communities in most of Europe. Strange, though in any case.

Only early alternative, not related to the Neolihtic IE would be to Scythian tribes possibly, who were living in parts of southeastern Europe( that somehow got around and maybe mixed with the Celts.

Levels of South Asian admixture in Europeans has been found before. In the fairly recent Tishkoff study, where they have between 4% and 17% of the "Indian" component. With 9% in Northern Italians, 10% in the French and 13% in Orcadian Scots.



Tishoff:
http://www.imagebam.com/image/38fd8034351627

In lei the the Barusho light blue is more seen more in French, Orcadians and Russians and its influence in Italy is not that great(Tuscany has a bit more whereas Sardinia has none)

http://img89.imageshack.us/img89/2673/li2008k6.png

Davidski said...

You're picking up Mesolithic admixture in the French. The signal is probably more distant, because the reference group less precise than the Karitiana, so the age of admixture comes out older.

Every time you use a modern group from North, Central or East Asia to test continental Europeans vs Sardinians, you'll mostly be picking up Mesolithic admixture.

No doubt, there are genuine post-Neolithic signals from Central Asia in modern European genomes, but these are minor compared to the effect of the mixing between Neolithic and Mesolithic groups in Europe.

Dienekes said...

You're picking up Mesolithic admixture in the French.

Not really, the two signals are separate. The North European/Karitiana signal represents the Mesolithic-type admixture, and the West Asian/Burusho signal represents an east-west population movement.

This was also clear in the analysis of West Asian and North European admixture:

http://dienekes.blogspot.com/2012/09/west-asian-and-north-european-admixture.html

The Burusho have very little North European admixture, so the signal is unlikely to be driven by admixture of a North-European-like population with Sardinians.

eurologist said...

Perhaps it would be worthwhile to repeat this test with Armenians.

In your K12b they are mostly "Caucasus" (i.e., Anatolia/Caucasus/Highlands) but only ~17% Gedrosia. Their North_European, South_Asian, and East_Asian is ~4%, 0%, and 0% (i.e., truly negligible), while that of Burusho is 12%, 25%, and 8%, respectively. So, the Burusho are almost 50% "contaminated" with what you might not want to include if you are looking at a relatively early admixture.

Kurti said...

What some people seem to have missed, no one is talking about modern Burushos being representative for the West Asian populations simply because modern Burusho people have significant South Indian admixture BUT the West Asian component in Burushos seem to come closest to what West Asian immigrants into Europe might have been like (Gedrosia?) What I start to believe is that the proto West Asian component might had some connection to South_Central Asia and was Gedrosia like and played a huge role in forming the West Asian and North European components. I believe Gedrosia is the originaly proto West Asian component which was frequent all over ancient West and Central Asia and the source component for Caucasus-West Asian through mixing with the Southern (Proto Mediterranean) component. And the same with North European which I think is simply a northern-northeastern drifted form of Gedrosia.

Slumbery said...

eurologist

This analysis is after the Southern + West Asian (/Gedrosia) admixture. This is the very admixture that created the Caucasus component itself, so it is hard to imagine any worse reference point than Armenians.

eurologist said...

Slumbery,

I completely disagree with your hypothesis. Europe has very little South_Asian and little East_Asian (the French have 0% and 0% in K12b) - and the latter in the wrong place (in the North) and for different reasons, many (if not tens of) millennia before. Burusho have very little in common with Eastern Anatolians or highlanders or neolithic Iranians that would have admixed with neolithic Western Anatolians (before further admixture into Western Europeans). They basically are Central Asian (i.e., early European/Caucasian and East Asian) with likely Mongolian, Turk, Iranian, Pakistani, and Indian (Gypsie?) admixture (the latter three adding more Gedrosia and South_Asian to them). But they have a great initial affinity with today's Gedrosia because Gedrosia has a strong affinity with all of Central Asia - not because they are somehow ancestral.

Furthermore, if Burusho were even close, how come they supposedly admixed so easily with Western Anatolians but left the people in between (Armenians and Georgeans) untouched? The reality is that these latter populations escaped the much later Mongolian, Turk, Iranian, and Pakistani admixture that the Burusho have (which makes them IMO completely unsuited as a neolithic stand-in).

Armenians are only ~10% Atlantic_Med, Georgians 0%, and in K7b Armenians are >50% West_Asian, Georgians >60% - so they are clearly a good stand-in for what would have entered Europe - even if they are admixed with early eastern Anatolian neolithic. At least, unlike the Burusho (who are only 45% West_Asian!), they don't have ~50% "contamination" that makes any results using the latter very questionable.

Kurti,

I somewhat agree with your last statement, but I think that Gedrosia is much derived (long ago) from the same population from which eventually all European and West Asian derived. It is by definition closer to Caucasus or South Asian due to proximity.

Slumbery said...

eurologist

"Europe has very little South_Asian and little East_Asian"

Nobody said that Burusho are a pure ancestral population in their current form (I do not think such a population possibly exist), also nobody said that their current area is particularly close to the West Asian (/Gedrosia) ancestral land. They simply represent a high West Asian population that got no Southern influence, so ideal as a reference for the admixture with Southern. Unlike the Caucasian population that got a lot of Southern, therefore utterly useless as a reference population in this analysis.

"Furthermore, if Burusho were even close, how come they supposedly admixed so easily with Western Anatolians but left the people in between (Armenians and Georgeans) untouched?"

Where you got the idea that they are untouched? Not by the Burusho of course, they probably did not even exist, but by the investigated admixture generally. They are very much touched. That is my whole point that you totally did not understand based on this question.

Do you really want to use a 35% Southern population (Armenians) as a reference to the admixture with a population represented by Sardinians? Indeed, Burusho have potentially disturbing admixtures, but for this analysis those are relatively insignificant compared to the high Southern among Armenians.

Burusho are not perfect for the purpose, but ancestral populations do not exist in the present (they are ancestral...), so Burusho is one of the best we have. Not because they are ancestral, but because they lack the investigated admixture.

Dienekes said...

"Furthermore, if Burusho were even close, how come they supposedly admixed so easily with Western Anatolians but left the people in between (Armenians and Georgeans) untouched?"

There is a strong negative f3(Armenian; Sardinian, Burusho) signal. It actually exists for the Brahu, Balochi as well, and for South Asian populations in general. This doubtlessly is due to the east-west population movement I talked about.

But, this was not Out-of-South Asia, since South Asian populations (including those of Balochistan) are ASI-admixed. It was more likely from some region between Anatolia and South Asia. As this element moved west, it admixed with the Southern early Neolithic population, and as it moved into South Asia it mixed with the ASI element, and as it moved into the steppe it mixed with aboriginal North_European-like people.

eurologist said...

They simply represent a high West Asian population...

@Slumbery:
But one that has less than Armenians and Georgians!

Where you got the idea that they are untouched? Not by the Burusho of course, they probably did not even exist, but by the investigated admixture generally. They are very much touched. That is my whole point that you totally did not understand based on this question.

I am not commenting on this any further because it contradicts all available existing research.

Dienekes,

You could shed so much more light onto this by doing the same analysis with Armenians and Georgians. Either way. Just do it. ;)

Slumbery said...

eurologist

I seriously do not get you and this is strange, so please do not back out.

Please specify which of my statements contradicts all available existing research and how. It may be evident for you, but you seemingly do not count with the possibility that you may be wrong and it is hard figure out anything if you do not actually say what exactly is your problem. Quite possible that I overlooked something, you are likely more literate on the topic, but in this discussion you have not written anything convincingly contradicting yet.

Especially do not get you, because I stated essentially the same here and you agreed.

Let me start from the beginning, maybe we can find the root of the misunderstanding.

Dienekes posted an analysis that tried to find the admixture date between the pre-bronze age European population, that notably did not have any West Asian, and the West Asian component.

He is after the West Asian coming time into Europe for a long time in multiple post, because he thinks it is IE arriving, but this is a side story.

His choice of mixed population was the French.
His choice as pre-West Asian reference population is the Sardinian as it looks the possible closest to the Neolithic people of the region.
Now, he had to choose a population as a reference West Asian.But it can't be a population that is itself highly admixed with the other reference population. Well, it could be, but it would point on a different admixture. This is a problem with Armenians. They indeed have very high West Asian, but they also have very high Southern. That is the worst possible contamination, because pre-West Asian Neolithic population was also had Southern as main component. Burusho are better, because they are much less Sardian like.

What part of this is in contradiction with existing researches?

Actually Armenians would be more interesting in the place of the French in this analysis.

Kurti said...

@eurologist

The breakup of the comoponents has shown us that Gedrosia is much more West Asian and ancestral than Caucasus which seem to have developed as a cross between West Asian + Mediterranean.

The Gedrosia component is much more ancestral and we have seen that North European appears to be Gedrosia + some Northeast Asian, or simply North European is Gedrosia shifted to North Europe since Gedrosia also shows affinities to East Asians.

Caucasus also appears to be Gedrosia+Mediterranean.

and ANI (Ancestral North Indian) seems also to be West Asian (Gedrosia likely) shifted towards South Asia.


I might be wrong but this seems very logical to me. And dont forget the place where it peaks is not always the place of origin. Gedrosia most likely originated somewhere in Southeast Anatolia or Northwestern Iran. Just like Mediterranean which peaks today in Sardinians was most probably the original Neolithic farmer component and originated in Southeastern Anatolia-Levant.

eurologist said...

Slumbery,

I apologize - I did not want to come over that harsh, and my comment initially wasn't even really directed at you, but more at the concepts. I like the ideas Dienekes pursues, but I also think there are oversimplifications. And there are huge time scales of admixture, here, almost always ignored.

I just think that all of these components we talk about float in space and time, and only condense perhaps in a meaningful way between the Mesolithic, Neolithic, and Iron Age. None of them are ancestral, but some can be identified with groups of populations at an earlier time than others. To me, the formation of "Caucasian" is not late Bronze Age or early Iron Age - as would, e.g., be indicated by Dienekes' latest Armenian analysis. It can't be, because by that time high-Gedrosia populations undoubtedly were already Eastern European, South Asian, and East Asian admixed.

To me, "Caucasian" started to form right when populations moved out of South and Central Asia towards Anatolia and Europe, perhaps mixing with a few native populations, and has consolidated since then. I also don't think "Sardinian" is Anatolian Neolithic. It's a mixture of Western Anatolian/Northern Levantine Neolithic with contemporaneous Balkan and Mesolithic Mediterranean (the significant diversity of ancient European neolithic mtDNA supports this).

For some reason, eventually, the still-forming Caucasian component started to dominate much of Anatolia and the Caucasus and the highlands, without mixing all that much with the similarly consolidating Gedrosia. Likely due to climatic conditions, while the formerly greatly-expanded Gedrosia contracted, the "Causasus" component blossomed, but somewhat isolated until the late bronze age.

The Western_Asian in Armenians or Georgians is not any more recent than that in Pakistani - IMO it is just much more characteristic (and less "contaminated") of populations that have inhabited NE Anatolia and the highlands all the way to Iran since the early neolithic and before. IMO, this is a population that potentially could have had a rather late impact in Europe - not populations that existed at that point in Afghanistan or Pakistan.

Slumbery said...

eurologist

OK, I think I get your concerns now, so I can possibly answer.

"To me, the formation of "Caucasian" is not late Bronze Age or early Iron Age - as would, e.g., be indicated by Dienekes' latest Armenian analysis. It can't be, because by that time high-Gedrosia populations undoubtedly were already Eastern European, South Asian, and East Asian admixed."

I totally agree, but as far as I understand this post does not say that "Caucasus" has Bronze Age origin. This ROLLOFF analysis places the admixture that formed the "Caucasisan" at around 7000 BP and may not very accurate because of inevitable noises (later interfering admixtures), so can be easily a bit older.

Sardinian is of course admixed. I think that the K7 Atlantic_Baltic component itself entirely created by the Neolithic admixture. The point is that it is free from West_Asian admixture. Dienekes is after the spread of West_Asian component and Sardinian is a good reference for this particular purpose.

"The Western_Asian in Armenians or Georgians is not any more recent than that in Pakistani - IMO it is just much more characteristic (and less "contaminated") of populations that have inhabited NE Anatolia and the highlands all the way to Iran since the early neolithic and before."

We disagree on this point.

"IMO, this is a population that potentially could have had a rather late impact in Europe - not populations that existed at that point in Afghanistan or Pakistan."

But agree on this and actually I think Dienekes says this too. Especially that Dienekes stated he suspects the source population was in Iran, just a bit South/East from the Armenian Highland.

That is why I said that there is a misunderstanding.

The conception: a West Asian source population in Iran that started to spread in every direction in the Neolothic, but not before the first wave of farmers left Anatolia and the Levant to Europe.
In the West it covered basically all the Middle East and formed the K12 Caucasus (7000-8000 BP). After that this K12 Caucasus was the main source of the following immigration into Europe from this region.
In the East it formed the K12 Gedrosia population, for example on Afghanistan and Pakistan.

You seems to disagree with the point that K12 Gedrosia is closer to the Ancestral K7 West_Asian that the K12 Caucasus, but the cross-analysis of the K7 and K12 components supports that.


As for Dienekes newest Rolloff about Armenians, it is not the origin of the K12 Caucasus, but a much later admixture and probably both participating population had Caucasus. The life did not stop after the K12 Caucasus was born.

Slumbery said...

Sorry for the double, I would like to clarify something.

Eurologist

You are quite possibly right that the West_Asian in the Caucasus is not more recent than in Pakistan. They are likely about the same age. Still, K12 Gedrosia (and not the local current population as a whole) seems to be the less contaminated version. Possibly because the Eastern migration of the West_Asian went into a less densely populated area (outside the fertile crescent).


Dienekes

You probably do not have anything to do with it, but the captchas of this site are sometimes bewildering. I can't prove I am not a robot. :)

eurologist said...

Slumbery,

I think we disagree on very little, at this point. My gut feeling is that Gedrosia and Caucasian started separating very early on (i.e., in Paleolithic & Mesolithic times, Caucasus at that time started as a subset of Gedrosia's original gene pool with perhaps some local contribution (if any), and Gedrosia is roughly local to its center today, and not due to migration from farther West. As to the cross-analysis, I still find that difficult to interpret for a number of reasons; in particular, it could indicated relations extremely distant in the past.

Let's be hopeful for further ancient DNA. The good news seems to be that before the late Bronze and Iron Ages, people seemed to have moved and mixed rather little.

eurologist said...

But, this was not Out-of-South Asia, since South Asian populations (including those of Balochistan) are ASI-admixed. It was more likely from some region between Anatolia and South Asia. As this element moved west, it admixed with the Southern early Neolithic population, and as it moved into South Asia it mixed with the ASI element, and as it moved into the steppe it mixed with aboriginal North_European-like people.

Dienekes,

Yes - and accordingly, it must have been (initially) very early - before substantial East Asian and South Asian admixture in this core population.

Unknown said...

I want to thank the Geneologist Dieneke for researching this link. It turns out I have the Burusho - Franco U8a. These answers are hard to find with rare polymorphisms.
In addition to add kudos to this subject, I harbor Maternal H17 and H27. The High Altitude regional mutations as well harbor LHON - medically diagnosed (as well).