search this blog

Wednesday, August 2, 2017

Steppe admixture in Mycenaeans, lots of Caucasus admixture already in Minoans (Lazaridis et al. 2017)


Over at Nature at this LINK. Why is the presence of steppe admixture in Mycenaeans important? And why does it matter if the Minoans already had a lot of ancestry from the Caucasus or surrounds? Because Mycenaeans were Indo-Europeans and Minoans weren't. I'm still reading the paper and will update this entry regularly over the next few days. Below is the abstract and, in my opinion, a key quote. Emphasis is mine.

The origins of the Bronze Age Minoan and Mycenaean cultures have puzzled archaeologists for more than a century. We have assembled genome-wide data from 19 ancient individuals, including Minoans from Crete, Mycenaeans from mainland Greece, and their eastern neighbours from southwestern Anatolia. Here we show that Minoans and Mycenaeans were genetically similar, having at least three-quarters of their ancestry from the first Neolithic farmers of western Anatolia and the Aegean [1, 2], and most of the remainder from ancient populations related to those of the Caucasus [3] and Iran [4, 5]. However, the Mycenaeans differed from Minoans in deriving additional ancestry from an ultimate source related to the hunter–gatherers of eastern Europe and Siberia [6, 7, 8], introduced via a proximal source related to the inhabitants of either the Eurasian steppe [1, 6, 9] or Armenia [4, 9]. Modern Greeks resemble the Mycenaeans, but with some additional dilution of the Early Neolithic ancestry. Our results support the idea of continuity but not isolation in the history of populations of the Aegean, before and after the time of its earliest civilizations.

...

The simulation framework also allows us to compare different models directly. Suppose that there are two models (Simulated1, Simulated2) and we wish to examine whether either of them is a better description of a population of interest (in this case, Mycenaeans). We test f4(Simulated1, Simulated2; Mycenaean, Chimp), which directly determines whether the observed Mycenaeans shares more alleles with one or the other of the two models. When we apply this intuition to the best models for the Mycenaeans (Extended Data Fig. 6), we observe that none of them clearly outperforms the others as there are no statistics with |Z|>3 (Table S2.28). However, we do notice that the model 79%Minoan_Lasithi+21%Europe_LNBA tends to share more drift with Mycenaeans (at the |Z|>2 level). Europe_LNBA is a diverse group of steppe-admixed Late Neolithic/Bronze Age individuals from mainland Europe, and we think that the further study of areas to the north of Greece might identify a surrogate for this admixture event – if, indeed, the Minoan_Lasithi+Europe_LNBA model represents the true history.

Lazaridis, Mittnik et al., Genetic origins of the Minoans and Mycenaeans, Nature, Published online 02 August 2017, doi:10.1038/nature23310

Update 03/08/2017: This is my own Principal Component Analysis (PCA) of the Minoan and Mycenaean samples, which are freely available at the Reich Lab website here. The Armenian angle for the eastern admixture in Mycenaeans looks forced. The trajectory of this admixture obviously runs from Northern or Eastern Europe to the Minoans. If it did arrive from Armenia, then realistically only via a heavily steppe-admixed population. Right click and open in a new tab to enlarge:


Update 05/08/2017: Much like Lazaridis et al., I ran a series to qpAdm analyses to find the best mixture model for the Mycenaeans. However, just to see what would happen, unlike Lazaridis et al., I didn't group any of the archaeological populations into larger clusters based on their genetic affinities. The three models below stood out from the rest in terms of their statistical fits.

Mycenaean
Minoan_Lasithi 0.786±0.049
Sintashta 0.214±0.049
taildiff: 0.96574059
chisq: 6.030
Full output

Mycenaean
Corded_Ware_Germany 0.210±0.043
Minoan_Lasithi 0.790±0.043
taildiff: 0.961238695
chisq: 6.198
Full output

Mycenaean
Minoan_Lasithi 0.791±0.043
Srubnaya 0.209±0.043
taildiff: 0.950419642
chisq: 6.558
Full output

So it's essentially the same outcome as the one obtained by Lazaridis et al., because Sintashta and Srubnaya are part of their Steppe_MLBA cluster, while Corded Ware is part of their Europe_LNBA cluster, and it's these clusters that, along with Minoan_Lasithi, provided their most successful mixture models for the Mycenaeans. But it's nice to see Sintashta at the top of my results, because it fits so well with the long postulated archaeological links between Sintashta and the Mycenaeans (for instance, see here).

By the way, here's what I said back in May when the Mathieson et al. 2017 preprint came out (see here). So things are falling into place rather nicely.

The same paper also includes the following individual from present-day Bulgaria dated to the start of the Late Bronze Age (LBA), which is roughly when the Mycenaeans appeared nearby in what is now Greece:

Bulgaria_MLBA I2163: Y-hg R1a1a1b2 mt-hg U5a2 1750-1625 calBCE

This guy is the most Yamnaya-like of all of the Balkan samples in Mathieson et al. 2017, and, as far as I can see based on his overall genome-wide results, probably indistinguishable from the contemporaneous Srubnaya people of the Pontic-Caspian steppe. He also belongs to Y-haplogroup R1a-Z93, which is a marker typical of Srubnaya and other closely related steppe groups such as Andronovo, Potapovka and Sintashta. So there's very little doubt that he's either a migrant or a recent descendant of migrants to the Balkans from the Pontic-Caspian steppe.

See also...

A Mycenaean and an Iron Age Iranian walk into a bar...

Main candidates for the precursors of the proto-Greeks in the ancient DNA record to date

Late PIE ground zero now obvious; location of PIE homeland still uncertain, but...

436 comments:

«Oldest   ‹Older   201 – 400 of 436   Newer›   Newest»
Rob said...

@ Chad

"Minoans certainly arent local. They're not close Greek N, Boncuklu, or Levant. They have the same process going on, but with the northern ancestry. Minoan>Late Helladic would only bring them closer. It needs to be considered. We need samples that stretch 3200-1300 BCE. I dont think Mycenaeans have just two mixes from Greek N starting point."

That's true because the late Neolithic Greek samples already show minor shift, differentially, toward the Tepecik - South Anatolia BA axis. But i don't think that is particular important for neither Minoans nor Myceneans.

The greatest impact for Bronze Age mainland Greece is definitely of a Balkan Copper Age origin. And that can be approximated as ~ 75% Baden, ~ 25% Kumtepe IV.
In turn, Kumptepe IV is ~ 60% ANF, 40% Ukraine Eneolithic.

Putting it all together, the 'northern signal' in Mycenean Greeks comes from something like Cernavoda culture which formed in the East Carpathian region after 4000 BC and appers in southern Thrace & Macedonia c. 3400 BC.

But that signal is missing in Minoans. They have the 'northwest Asian' shift also seen in Myceneans, but to lesser degree. I'd say most of their ancestry is 'local'.

Chad said...

It may not be a popular opinion, but I still expect pretty steppe-heavy samples between 2200-1600BCE, for sure, and some lingering during Mycenaean times. Z2103 is my pick for that time.

1Mero171 said...

@ Davidski

I fully understand and agree! The CHG type migration occurred prior to steppe migration. What bothers me is that now some people are mistakenly associating CHG ancestry and DNA with Indo-European language (e.g. Sarno et al, 2017).

Probably the population of the south Balkans prior to the steppic people migration was composed of a mixture of Neolitic farmers + CHG/Iran and high proportion of ydna J like minoans.

Chad said...

Ill go with Myceaean= "Minyan+Minoan-like".

Rob said...

@ Chad
What percentage admixture from the steppe-rich prediction would there be ?

Rob said...

I think they'd have influences Thracian etc

Samuel Andrews said...

@Rob,
"The greatest impact for Bronze Age mainland Greece is definitely of a Balkan Copper Age origin."

There's a million different ways Myceanean could have goten its EEF/CHG/EHG ratios. But the way you think it happened doesn't fit with ancient DNA. The vast majority of Balkan Copper age genomes show little or no CHG and EHG ancestry yet you give the impression both CHG and EHG were prevalent.

You always pick the hypothesis where Steppe and especially IE speaking Steppe does nothing. Don't forget some of your past suggestions turned out wrong. I remember you were one of the people who suggested Steppe ancestry in Europe could be a reflection of local Mesolithic ancestry combined with more recent ancestry directly from CHGs. That turned out to be wrong.

And now you're trying to twist the sort of confusing results from ancient Bulgaria to make the Steppe signal in Myceaneans a reflection of something else.

a said...

It will be interesting to see when Vucedol R1b results are released. The-bell-beaker-behemoth-has a sample ... (2884-2666 BC), Beli Manastir, Croatia, Vučedol culture R1b-Z2103+ I believe.

https://abload.de/img/z93w9use.png
I3499 does appear to have more Yamnaya component than I2792 sample.
Also Kefalonia has a KMS-67* sample,, besides R1b-9219+ found in Albanians

https://www.yfull.com/tree/R-KMS67/
id:YF09904GRC [GR-23]Kefalonia. Of course sister branch R-Y20993*id:YF03134 is recently found in ancient Sarmatian sample from Steppe.

Alexandros said...

@Matt

Thanks for the very useful comments. May I just ask what do you mean by "an elite female from Cyprus" in your previous comment??

Gioiello said...

@ a
"https://abload.de/img/z93w9use.png
I3499 does appear to have more Yamnaya component than I2792 sample.
Also Kefalonia has a KMS-67* sample,, besides R1b-9219+ found in Albanians

https://www.yfull.com/tree/R-KMS67/
id:YF09904GRC [GR-23]Kefalonia. Of course sister branch R-Y20993*id:YF03134 is recently found in ancient Sarmatian sample from Steppe".

Samples from Samara in Eastern Europe are all KMS75, downstream KMS67, thus Western European KMS67 (xKMS75) don't descend from Samara. You are all newbies in these analyses, but I am doing them from more than ten years. None like me.

Alexandros said...

David, thanks for the Fst matrix!

If I am not mistaken, in that matrix the top modern populations showing the smallest Fst with Minoans are:
1. Italian_South
2. Sicilian_West/Cypriot/Greek/Italian_Tuscan/Sicilian_East (equally close)

The corresponding populations for Mycenaeans are:
1. Italian_South/Sicilian_West (equally close)
2. Cypriot/Greek/Italian_Tuscan/Sicilian_East/Albanian/Italian_Bergamo/Sephardic_Jew (equally close)

Gioiello said...

@ newbies

"besides R1b-9219+ found in Albanians"

Albanian and Balkan R-L23-CTS9219 are all very recent, they are the "Balkan cluster" that I discovered many years ago, and the oldest samples are in Italy and Western Europe.

Alexandros said...

The paper's PCA plot is rather confusing, as it should be made clearer that not all 'present day' populations are depicted in grey colour. Modern Greeks and Cypriots are depicted in colouration.

Some things which catch the eye from that plot:

The Bronze Age Cretan from Armenoi clusters nicely with modern Greeks!

Pitty they did not distinguish in the plot the Southern Neolithic Greek (Diros, Peloponnese) from the previous Central/Northern Neolithic Greeks.

Curiously, one Anatolia_Kumtepe sample clusters with modern Greeks. Isn't this a Neolithic sample?

Also curiously, Minoans/Myceneans cluster with some 'present day' samples which I assume are Southern Italians/Sicilians and Sephardic Jews?

David's PCA plot, although broadly similar, does have important differences, with much clearer differentiation of ancient and modern populations. I understand this has to do with different methodology used, as well as slightly different samples included.

Simon_W said...

@Arch Hades

"Simon, how do u know Minoans were non IE?"

It looks very non-IE to me, like an exotic African language:

ATAI*301WAE ADIKITETE DA PITERI AKAONE ASASARAME UNARUKANATI IPINAMINA SIRU[TE] INAJAPAQA
ATAI*301WAJA JADIKIRA JASASARA-[…]-SI IPINAMA SIRUTE TANARATEUTINUIDA
ATAI*301WAJA TURUSADU*314RE IDAA UNAKANASI IPINAMA SIRUTE TURUSA

And apart from a few words it has escaped translation. If it were IE it would be easier to translate because of cognates with other languages.

Simon_W said...

@italicroots.lefora.com

"WoW I actually espected ancient Greeks to turn out as Poles and Germans, who later on mixed with Lebanese slaves during the Byzantine period becoming modern Greeks. :-/"

Lol, seriously? Sounds like old school Nordicism, the ancient Greeks as pure Nordics - the modern Greeks as a different, lowly race. And the ancient Greeks produced great art the Germans and British loved because they had Nordic souls. These theories can now be put to the dustbin obviously. Modern Greeks are even more Northern than the ancient ones, because of Slavic and Albanian admixture.

Simon_W said...

I have to correct myself: If we take the position of Klei10 in David's PCAs as a basis, then the Minoans are rather Anatolia_BA-shifted than towards the Levantine samples. But the rest of my observations still holds.

Another observation: Modern Greeks are not just more northern than Mycenaeans, they also have more Caucasus-related admixture! At least judging by the ADMIXTURE analysis of the paper at K=16, where the CHG are almost completely made of one, pink component. So probably Armenian/Anatolian admixture from Byzantine times I suppose.

Simon_W said...

As for the Etruscans, originally they may have been Minoan-like, when they first set foot on Italian soil, but I'm convinced that they assimilated IE Italic people, especially in the Etrurian heartland. So classical Etruscans from Etruria will show Steppe admixture is my prediction. Some parts of the southern Po valley may have been different though, as they are devoid of Protovillanovan finds.

truth said...

@Simon_W

We already have some 2.5K Etruscan samples, see where they plot on PCA :

https://image.ibb.co/kNbqcF/etruscans.jpg

Alogo said...

As I wrote above, to me the Minoans seem like a mix of the local Neolithic population with a new one arriving from Bronze Age Anatolia to Crete which is something that has often been posited (e.g. some papers in, the older but pretty classic I think, 'Bronze Age Migrations in the Aegean' come to mind). The Mycenaeans are a lot like the Minoans with extra steppe ancestry that was likely mediated via a mixed Balkan-steppe group (like some of the ones we saw in the earlier paper) arriving from the north, which is also a pretty commonly accepted scenario. And I'll even consider the possibibility that (pre-)proto-Greek is the result of various Indo-European linguistic streams mixing together, both from the northeast and the northwest (e.g. centum but also ties with Indo-Iranian) until we can connect it more reliably to particular cultures further north.

The language of Linear A might be non-IE but it still might be a Bronze Age Anatolian language of some sort (perhaps some sort of Caucasian-Hattian as sometimes posited) and the pre-Greek linguistic (the toponyms in ss/nth, plenty of words with non-IE clusters etc.) substratum having ties to even post-Neolithic Anatolia is strengthened with these findings imo.

Simon, italic was joking. Like you wrote, the difference to these samples seems like slightly more Anatolian and a lot more Northern European revealing the continuing connections with Anatolia but especially Northern Europe (the Germanic and Slavic migrations, no doubt).

Gioiello said...

@ Alogo

You nick doesn't make us hope well. It means "without reason". In fact what you write is without reason: "and the pre-Greek linguistic (the toponyms in ss/nth, plenty of words with non-IE clusters etc.)". From so long we know that those suffixes are Indo-European...

Matt said...

@ Alexandros: Thanks for the very useful comments. May I just ask what do you mean by "an elite female from Cyprus" in your previous comment??

Duh, I messed up. She's from Crete. Check out p22 of their supplement for more info (Armenoi 503). No ancient Cretans, sorry. Still pretty sure about the rest though.

Also I'm actually not sure how elite she was either though, checking that again - her sample is from a very richly furnish necropolis of chamber tombs, but she's not labeled as a member of the "ruling class" like Peristeria4 from the rich Tholos 1 burial at Peristeria in the western Peloponnese.

(Tholos tombs are the famous "beehive tombs" like the Treasury of Atreus. Chamber tombs are more ordinary Mycenaean tombs.)

@ SimonW: Another observation: Modern Greeks are not just more northern than Mycenaeans, they also have more Caucasus-related admixture!

Caucasus admixture should go up with Steppe_EMBA admixture though, for any population that has less than 30-50% CHG.

Alogo said...

We don't 'know' that. Linguists still argue about it, at least in some contexts, quite a bit. And of course leaving that aside, most of those substrate words have never been reliably ascribed to Indo-European.

Anyway it's pretty clear that both pre-Greek and proto-Indo-European are the languages of the Italian refugium.

Alexandros said...

@Aris
RE Cyprus and BA Greece, the special connection has been known to linguists for a long time, so no surprise there; see https://en.wikipedia.org/wiki/Arcadocypriot_Greek

I don't think so.. The Mycenaean origin of Cypriots is an issue debated intensely in the archaeological literature and surely linguists have not done much in clarifying this, other than providing evidence that the ancient Greek dialect spoken in Cyprus was very closely related to the dialect spoken in Arcadia, Peloponnese (hence Arcedocypriot dialect). Over the past 20 years some archaeologists (e.g. Prof Bernard Knapp from the University of Glasgow) have been arguing strongly against a major Mycenaean migration into Cyprus during the LBA/early IA, which has been the mainstream paradigm supported by Greek and Cypriot archaeologists for decades.

@italicroots.lefora.com
With all the respect, why on earth would you think that ancient Greeks would come out as modern Poles and Germans?? Check out some Minoan and Mycenaean frescoes below.. They always depicted themselves with black hair (although they surely had the skills to paint blonde hair in their frescoes if they wanted to) and men in particular had an intense tan, similar to the one I get when I am exposed to the sun continuously for a couple of days during the summer..

Minoan men
http://brewminate.com/wp-content/uploads/2017/03/MinoanFresco02.jpg
http://www.visit-ancient-greece.com/image-files/knossos-fresco.jpg

Minoan women
http://images.metmuseum.org/CRDImages/gr/web-large/DP234770.jpg

Mycenaean women
http://1.bp.blogspot.com/-KoDg7UZyZDQ/VR6SagBhRGI/AAAAAAAAA_s/53hDsUWEkmk/s1600/Mycenaean%2Bwomen%2Bbringing%2Bgifts%2Bfresco.jpg

Ariel said...

That Etruscan PCA is strange, one individual was probably Iberian BB like, but two of those are shifted towards eastern Europe (Or maybe ANE, caucasus, steppe?). Maybe those two individuals were Greek-like or Bulgarian-like. And proably they were around equally or slightly more basal then northern/central Italians. Very much unlike minoans.

Simon_W said...

@truth

Yep I knew that already, I don't even have to check your picture to know what you posted. The information value of such a limited PCA is limited, but if the Etruscans varied between Iberian-like and Bulgarian-like some Steppe admixture is likely.

Simon_W said...

@ Ariel

Like I said: Classical Etruscans from the Etrurian heartland, 2.5 kya will have Italic admixture. Original Etruscans from the Bronze Age who first set foot on Italian soil probably on the Po estuary may be Minoan-like.

Simon_W said...

@Alogo

"Simon, italic was joking."

Maybe, but I doubt it. ;-)

Simon_W said...

@Matt

"Caucasus admixture should go up with Steppe_EMBA admixture though, for any population that has less than 30-50% CHG."

Yes, Steppe_EMBA would do that, but there was no Steppe_EMBA population that postdates the Mycenaeans. And admixture from Europe_LNBA for instance wouldn't suffice to explain the extra-CHG in modern Greeks.

Actually that modern Greeks have more CHG than Mycenaeans is the biggest surprise to me.

Simon_W said...

Steppe admixture in Mycenaeans per se doesn't mean a lot, unless we realise that this was a prediction of the steppe theory. Hence it's a verification of a prediction of the steppe theory. So the only option for the sceptics is to call into question that it's really steppe admixture and not some pseudo-steppe due to a mix between Balkan foragers that had been there since the Mesolithic and CHG from Anatolia.

Matt said...

SimonW: Yes, Steppe_EMBA would do that, but there was no Steppe_EMBA population that postdates the Mycenaeans. And admixture from Europe_LNBA for instance wouldn't suffice to explain the extra-CHG in modern Greeks.

Well, yes, as long as the Europe_LNBA had more CHG than Mycenaeans it would. Of course, "Europe_LNBA" did.

Assume Steppe_EMBA = 50:50 CHG:EHG.

Assume Europe_LNBA 50:50 Steppe_EMBA:Other. Therefore Europe_LNBA 25 CHG.

Per basic models: Minoan 14:86 CHG:Anatolian. Mycenaean 13:87 Minoan:Steppe_EMBA.

Assuming above levels of CHG for Steppe_EMBA, then Mycenaean 18.5:6.5:75 CHG:EHG:Anatolian.

Since 18.5 is lower than 25, any admixture from Europe_LNBA into Mycenaean should cause CHG to rise.

I'm not saying you're wrong, just that it's a more complex matter of looking at and quantifying the ratio of whether changes in CHG are excessive to expectation based on more admixture from a European source. We can't just look at the level and presume an Antolian source.

batman said...

@ andrew

"we've pretty much figured out the Minoan number word system and some of the words for farming products and quite a few proper nouns (that have Mycenaean cognates that sound the same). And, basic words for numbers and farming products are among the words which are most resistant to linguistic drift."

Most interesting. Was their counting decimal or duodecimal?

Unknown said...

@Davidski

Would you be able to fit the BA Anatolians, The Minoans from Lassithi, and the Mycenaeans each as an admixture of 1) Neolithic Tepecik 2) Iran Chalcolithic and 3) Yamnaya as 3 source populations?

batman said...

"Mycenaean have more CHG and Eastern Paternal DNA J2a .

Link with J2 and Indo-European is directly confirmed."

Same as with G2 and I2. Together they covered the entire area were the IE languages came to spread, already during the LME/ENE transition. Well before the lactose-persistance started to spread, along with R1a and R1b.

Anthro Survey said...

@Simon

That Modern_Iberian-like Etruscan needn't have had Italic admixture. There is a simpler explanation: he was merely a plain, Etruscanized North Italian who autosomally resembled the Parma sample. I'd venture to say that most people in both Padanian and Tyrrhenian Etruria would have been like this, w/Minoan-like admixture restricted to upper castes.

Gioiello said...

@ Alogo
"Anyway it's pretty clear that both pre-Greek and proto-Indo-European are the languages of the Italian refugium".

Pre-Greek, if linked to Pelasgians or to the language of Lemnos, could be linked with Etruscan and Rhaetian and Camun which ended to come to Italy from central Europe. We'll see if my hypothesis will be proved, better, I think, than the Schrjver (great cathedratic) one.
About the "Italian Refugium" and a possible link with hgs R1b (and I think also R1a and much other) we'll see next: let that your friends test Italy at least how much all the other European countries.
About names in "ss/nth" read also what you never read in the past, and in other languages and not only in English.

"And of course leaving that aside, most of those substrate words have never been reliably ascribed to Indo-European".

The Balkans have many IE substrates, and I don't repeat what I wrote many times in the past.

Grey said...

"It's still pretty mysterious to me why the early Armenian Chalcolithic and Bronze Age are like they are, though, rather than just on a Anatolia_N to CHG cline. Why are they EuroHG shifted from positions on this cline? Early ancestry from steppe / North Caucasus?"

listing possible options for how it *might* have happened is useful imo

1) conquest

2) mercenaries

3) artisans/traders

4) infiltration migration by mode of food-getting (for want of a shorter phrase) i.e. settlement of regions of land that were marginal for crop farming but optimal for cattle within settled territory (possibly by invitation)

5) other

#

conquest might be a bit early
mercenaries - maybe
if artisans/traders then of what? (horses/metal?)
if cattle then are there regions in Old Armenia that were/are good cattle country and if so does the EHG component map to those regions?

(some years back i read hayasdan was derived from "metal-worker" which has skewed ny view but that was in a computer game so prob not the best evidence)

#

the other thing is are rich burials sometimes just rich burials as opposed to political elites? this might align with the artisan/trader option more than the conqueror option (at least in some cases)

Rob said...

@ Sam

"There's a million different ways Myceanean could have goten its EEF/CHG/EHG ratios. But the way you think it happened doesn't fit with ancient DNA. The vast majority of Balkan Copper age genomes show little or no CHG and EHG ancestry yet you give the impression both CHG and EHG were prevalent. "

I think someone has major gaps in their understanding, and I don;t think it's me.

Anatolia 3600 BC (middle Chalccolithic)
(600 years before Yamnaya)

Kumtepe_LN:kum4
"Tepecik_Ciftlik_N:Tep006" 61.9
"Kotias:KK1" 14.65
"Ukraine_N1:StPet12" 17.75

Davidski said...

Oops, forgot to include the Tepecik-Ciftlik Anatolians in the Anatolia_N set in my PCA. They're there now, and interestingly they sort of overlap with the Minoans.

https://3.bp.blogspot.com/-qDAN5pZ8Qm0/WYUXuKv8fPI/AAAAAAAAF80/ltaSA7RXUEMCELq5ceMKYmdDqxHpH4nNwCLcBGAs/s480/Minoans_%2526_Mycenaeans.png

Seinundzeit said...

For what it's worth, in this case nMonte (with Global_10) strikes me as being rather effective.

As per the paper, 79% Minoan_Lasithi + 21% Europe_LNBA is the most sensible solution for the Mycenaeans as a whole.

For comparison, this is what I find using PCA data (scaled).

84.85% Minoan_Lasithi

8.6% Andronovo + 3.65% Afanasievo + 2.05% EHG

0.85% Gambian

distance=0.0271

Very similar to what we see in the paper.

I didn't even try to create an optimal setup for these ancient samples; I simply threw them into the same setup I use for South Central Asians and Indians.

Unknown said...

@Davidski
"Oops, forgot to include the Tepecik-Ciftlik Anatolians in the Anatolia_N set in my PCA. They're there now, and interestingly they sort of overlap with the Minoans.

https://3.bp.blogspot.com/-qDAN5pZ8Qm0/WYUXuKv8fPI/AAAAAAAAF80/ltaSA7RXUEMCELq5ceMKYmdDqxHpH4nNwCLcBGAs/s480/Minoans_%2526_Mycenaeans.png"

Thanks! But I cannot read the legends or discern the data point plots because the resolution of your png is too low. Could you possibly save it at higher visual resolution? I agree that Tepecik could be a proxy/source for the Minoan immigration to Crete.

Davidski said...

@Roy

Right click on image > Open link in new tab

Davidski said...

@Matt or anyone else who might know

Have there been any detailed attempts in any papers to model Iran_IA F38 with qpAdm yet?

Unknown said...

@Davidski

Are the Tepecik samples in your plot labeled Anatolian Neolithic and are the brown x's that overlap the Minoan samples?

Davidski said...

Are the Tepecik samples in your plot labeled Anatolian Neolithic and are the brown x's that overlap the Minoan samples?

Yep

Alexandros said...

After a reality check last night and despite my initial astonishment, it now seems clear that Cypriots are not particularly close to either the Minoans or the Mycenaeans. This is clear by looking at David's PCA plot, which apparently does not suffer from projection bias, which seems to affect the paper's PCA plot. The Admixture results also show a clear differentiation between Minoans/Mycenaeans and any modern population. The Fst (inbreeding coefficient, as per paper authors) indicate however a clear genetic affinity between Mycenaeans/Minoans and modern Greeks, including Cypriots. The f4 results also indicate clearly the additional shared drift of modern Greeks and Cypriots with other populations, as compared to Mycenaeans.

I am surely no expert in autosomal DNA analysis, but this is my crude interpretation after a careful first read.

Alexandros said...

Very interesting to see the predicted hair and eye colour! If you notice the images shared in a previous comment of mine, both Minoans and Mycenaeans presented themselves in their frescoes as brown/black hair/eye people. This is now confirmed by the genetic data! Interestingly, there is a Lasithi Minoan with predicted 'blond or brown' hair. Actually her probability of blond hair is just below 50%. I would assume this individual had light brown hair, a phenotype quite common in modern Greeks and Cypriots. Maybe something like this??
https://s-media-cache-ak0.pinimg.com/236x/f6/cd/b5/f6cdb5d98515f912433e06faffc520c2--hairstyles-haircuts-hairdos.jpg

Alberto said...

@Roy King

I can only use the data from the Global 10 PCA, so maybe Davidski or Chad can verify this with qpAdm:

Anatolia_BA
Tepecik_Ciftlik_N 68.8 %
Iran_ChL 24 %
Yamnaya_Samara 7.2 %

Minoan_Lasithi
Tepecik_Ciftlik_N 93.6 %
Yamnaya_Samara 6.4 %
Iran_ChL 0 %

Mycenaean
Tepecik_Ciftlik_N 81.2 %
Yamnaya_Samara 18.8 %
Iran_ChL 0 %

(These are weighted values, but I checked with unweighted and they are exactly the same in these cases, just decimals apart).

Aram said...

Davidski

Thanks for that second PCA with Tepecik. But it seems that the image is too much compressed and letters are blurry. Whatever.

Their overlap with Minoans was predictable.

But are You sure that Peloponnese Neolithic samples from Mathieson paper about Balkans who had the same Levantine/CHG shift will not be in the Minoan neighbourhood? This samples are not still released as far as I know and they are absent from Your PCA.






Davidski said...

@Roy

Would you be able to fit the BA Anatolians, The Minoans from Lassithi, and the Mycenaeans each as an admixture of 1) Neolithic Tepecik 2) Iran Chalcolithic and 3) Yamnaya as 3 source populations?

Results are very similar to those using G10/nMonte, and the fit for Anatolia_BA is great.

https://drive.google.com/file/d/0B8XSV9HEoqpFaERCQ3BVSHE0R1E/view?usp=sharing

https://drive.google.com/file/d/0B8XSV9HEoqpFSXVuQkg4cHFOYkk/view?usp=sharing

But Mycenaeans take a lot more Yamnaya (25%).

https://drive.google.com/file/d/0B8XSV9HEoqpFdUVlMElhN1dpMGc/view?usp=sharing

Davidski said...

@Aram

But are You sure that Peloponnese Neolithic samples from Mathieson paper about Balkans who had the same Levantine/CHG shift will not be in the Minoan neighbourhood? This samples are not still released as far as I know and they are absent from Your PCA.

Could be. I didn't know you were referring to these samples.

Aram said...

Anthro Survey

Are we looking in the same place?

This are the ages of Peloponnese Neolithic samples from Mathieson

5000-3200 BCE Greece_Peloponnese_Neolithic
5000-3200 BCE Greece_Peloponnese_Neolithic
5000-3200 BCE Greece_Peloponnese_Neolithic
4043-3947 calBCE (5170±30 BP, Poz-81125) Greece_Peloponnese_Neolithic

Aren't they older than Minoans?

I am also perplexed why Lazaridis didn't notice it? Maybe this samples were not available when he was preparing his paper?


Matt said...

Application of PCoA and Neighbour Joining to the Fst table:

http://imgur.com/a/7JH20

The PCoA dimension 1 and 2 seems to fit Mycenaeans better as Minoan+Steppe_EMBA, while Minoan+EuropeLNBA/SteppeMLBA and GreekLN+ArmeniaMBLA are slightly worse. Wouldn't read too much into this though.

(PCoA on the West Eurasian subset of Fst looks pretty unclear above 2 dimensions; lots of dominance of additional drift that may either be ancient population specific stuff caused by isolation, or not real caused by the problem of using ancient samples.)

As always, caveat neighbour joining links populations to the closest other population, strongly influenced by recent and local ancestry, and is not necessarily totally in line with overall "deep" ancestry.

Anthro Survey said...

@Aram

"Aren't they older than Minoans?"

They are not older than the Neolithic Peloponesian sample from Diros used in this study, which predates them by 500-1000 yrs. As you can see, it doesn't differ substantially to Thessalian or Anatolian Neolithic individuals. It's not Levant-shifted as those in Mathieson.

Had they been older, my initial assertion that Mathieson's samples represented a dead-end population in the Peloponese would have remained strong because it would imply there was no population shift. Starting point: Anatolian-like. End-point: Anatolian-like.

But, again, the Neolithic samples from Mathieson POST-DATE the one from Dirost that Lazaridis used. This challenges my "dead-end" assertion and raises the possibility of widespread Levant-like influence in Peloponese in later Neolithic times arriving via a southern Anatolian coastal route(as we both speculate). This Levant-shifted population, in turn, would have received CHG-rich admixture and later (Balkan?) steppe admixture to form Mycenaneans.

Alexandros said...

@Anthro Survey
There is some confusion (to me at least) regarding the Greek Neolithic samples used in the Lazaridis et al and Mathieson et al papers. In their preprint, Mathieson et al state the following:
"We find that four southern Greek (Peloponnese) Neolithic individuals – three from Diros Cave and one from Franchthi Cave, plus one previously published individual from Diros(27) – are not consistent with descending from the same source population as other European farmers". Interestingly, reference #27 is "Lazaridis, I. et al. Manuscript in revision. (2017)" (i.e. the current publication by Lazaridis).

In other words, according to the above, 5 in total Southern Greek Neolithic individuals were analysed in the Mathieson paper and the Diros Neolithic individual, analysed in the current Lazaridis paper, was one of them. In contrast, when you see the actual results from the Mathieson paper, there are only 4 data points for Peloponnese Neolithic. Even if they did not analyse this specific sample, they clearly state that she clusters with the rest of the Peloponnese Neolithic samples from that study. Any thoughts??

Alberto said...

Looking at the Fst differences between Mycenaean Greeks and modern Greeks, on one end we find that the top populations that are closer to modern Greeks relative to Mycenaeans are:

Latvian
Lithuanian
Mari
Steppe_Eneolithic
Vepsa

All with the same difference. Not very surprising due to Slavic migrations. While on the opposite end, the populations closer to Mycenaeans relative to modern Greeks:

Natufian
Levant_N
Europe_N
Barcin_N
Anatolia_BA

It is normal to see ancient ones first. But among modern populations (leaving SS-African ones aside):

Sardinian
BedouinB
Egyptian
Lebanese_Christian
Palestinian
Saudi

Simon_W said...

It's a useful thing that the authors of the paper point out that IE Pisidian has been attested close to the area where their 3 Anatolia_BA samples were excavated, I wasn't aware of this. However, if you consider the dates of these samples - two of them having calibrated C14 dates: The one is roughly from 2800-2500 BC and the other from 2500-2300 BC then it becomes obvious that you could just as well or even with more likelihood assume that they spoke something close to non-IE Hattic.

Simon_W said...

@Matt

Agreed, your calculation seems to make sense. In that case the ADMIXTURE analysis is misleading, at least the run with K=16 that I looked at, because there the Mycenaeans have more of the pink CHG-related component than Europe_LNBA.

Davidski said...

I've updated the post with the best mixture models that I could find for the Mycenaeans, using no less than 16 outgroups.

Simon_W said...

@Anthro Survey

That may be possible if that Etruscan really was from Northern Italy, which I somewhat doubt, but we don't know for sure.

Then again Chalcolithic influence from the southwestern Balkans which might have reduced the Iberian-affinity by adding CHG admixture was not that strong in Tuscany. I've heard cultural influence from the Campanian Gaudo culture upon the Rinaldone culture may represent such a vector of admixture.

As for the strength of the Minoan-like admixture, it may have varied geographically; in my theory it was strongest around Frattesina, until that site was abandoned in the Iron Age. Afterwards these people may have moved somewhat, maybe to the particularly dark haired spots around Forli and Mirandola. At least I can say that 13.75% Minoan_Lasithi works very well in modeling my Global10 coordinates with nMonte, it makes the Jordan_EBA almost completely disappear.

Carlos Aramayo said...

@ Davidski and all

Based on the recent paper`s abstract by Lazaridis et al 2017 on Mycenaeans and Minoans together with Ann Gibbons commentary article in Science web page, it is clear to me that it can be inferred 2 migrations from Eurasian Steppe towards Aegean (Greece), The first one, with Eastern Caucasus and Iran ancestry, bringing Anatolian IE language to Aegean around 4000 BC, and the second one affecting Mycenaens, (who had already adquired genetic influence from the first wave), maybe around 2000 BC.

This is very similar to the possible three IE waves affecting South Asia, which I mentioned in a previous post, taking place since 2500 BC. But paradoxically the two waves towards Greece left behind only 25% or less of IE genetic ancestry. But to India with a massive genetic flux.

Ariel said...

In 4000 BC steppe people were still in the steppe. And the Greek language isn't that all.

Matt said...

Re:qpAdm, Lazaridis 2017 pretty much goes all out with their qpAdm right pops.

For the final fits, if I understand correctly they use the right set

All++: Mota, Ust_Ishim, Kostenki14, GoyetQ116-1, Vestonice16, MA1, AfontovaGora3, ElMiron,Villabruna, WHG, EHG, CHG, Iran_N, Natufian, Levant_N, Anatolia_N, Anatolia_ChL,Armenia_ChL, Armenia_EBA, Armenia_MLBA, Europe_LNBA, Europe_MNChL, Iberia_BA, Iran_ChL, Levant_BA, Steppe_EMBA, Steppe_MLBA, Minoan_Lasithi, Mycenaean, Anatolia_BA

Basically dropping any in right if they also use it in left (e.g. if you're modeling Mycenaeans with Minoan_Lasithi and Europe_LNBA, then they ain't in the right).

Really guess they don't have any concerns any more about including closely related w/Eurasians in both right and left any more at all.
Further they state:

Adding these later populations has one disadvantage: if populations A and B are both included in the larger set and are composed of the same ancestral elements in similar proportions then A may be modeled as deriving most of its ancestry from B and vice versa. This does not clarify the ancestral origins of either population.

However, this approach also has the advantage of identifying mixture when the admixing populations are themselves complex. For example, if a population A is a mix of B and C, and B and C are themselves 2- or 3-way mixtures, then this approach might identify a simpler ix in the origin of A than would be possible if B and C were not considered as source populations.


So it's kind of inverse approach to the "use unrelated outgroups only" idea that originally came out with qpAdm.

Matt said...

Also, re; the simulations approach (the fully new methodology here), if I'm getting it right and if anyone is interested, this is how it works:

Remember Lazaridis and Reich's rebuttal of Goldberg's paper, where they used simulated combinations of Steppe+Anatolian+WHG ancestry and then ran f stats against them?

This is kind of the same thing (across the autosome, not just X).

qpAdm IIUC compares a combinations of correlations of D-stat output, finding the one which minimises against the real population by using the p test.

This simulation approach is (I think) instead to simulate a genotype for all possible combinations of the streams of ancestry in pleft, then run direct f stats against all combinations of pright.

Then the best fit is the one with the lowest Z against the real population.

You can see the comparisons between the two methods in http://www.nature.com/nature/journal/vaop/ncurrent/fig_tab/nature23310_SF6.html for fitting Mycenaean.

(These all use the same, large pRight set I mentioned above for their qpAdm).

qpAdm minimises fit (p=0.25) over 88:11:1 Minoan_Lasithi:Steppe_EMBA:Armenia_MLBA. However the best fitting combination of these three elements in the simulation (80:18:2) has a Z of 3.2 (f stat difference between the real and simulated population).

On the other hand, the best simulation is 62:31:7 Anatolian:Armenia_MLBA:Steppe_MLBA, with worst Z of 2. The analogous qpAdm fit at 65:33:3 has p=0.62.

The final use of their simulations is in the stat f4(Simulated1, Simulated2; Mycenaean, Chimp), where they compare the simulants against each other in how much drift they share with the real population.

In this one, the most optimal simulant appears to be 79:21 Minoan_Lasithi:Europe_LNBA and the worst and least good fitting compared against it is the 78Minoan_Lasithi:22Steppe_EMBA simulation.(The similar Minoan_Lasithi:Steppe_EMBA:Armenia_MLBA 80:18:2 fit is also comparably poor).

Alexandros said...

@Carlos Aramayo

Overall agree, although not sure of the origin of the first migration you are describing. Just a clarification on the second steppe migration you mention though.. It did not AFFECT the Mycenaeans, it has CREATED them.. In other words, the Mycenaean culture was in its origin an Aegean IE culture, with distinct customs closely resembling those of BA steppe cultures. Therefore, the 'Mycenaeans' per se did not exist prior to the steppe invasion/migration. They are a hybrid population of indigenous BA (Helladic) people (closely resembling Minoans) mixed with the invading Steppe people.

Anthro Survey said...

@Simon:

Yeah, Frattesina is a pretty strong candidate for that. Hopefully they're thinking about searching for ancient DNA there! The Po Delta's role in being a landing ground for arrivals from the East Med has been recapitulated multiple times in history. Ancient sailors and traders tended to be conservative in their choice of harbors and trading posts. I wouldn't read too precisely into modern distribution of dark hair, but yes, that kind of influence could have radiated outwards from the Polesine area into proximal regions like Romagna. Moreover, it's not merely a question of pigmentation: the eastern Po valley tends to have a high frequency of thalassemia and Cypriot-featured folks like the prime example Davide Vallicelli.

thalassemia: http://images.slideplayer.com/33/8220110/slides/slide_2.jpg

In posting the image, I realize the region isn't autosomally equivalent to southern Italy and selective processes there account in large part for the frequency: there's no shortage of mosquitoes in the Po Delta! lol

SouthWestern Balkans: what area did you have in mind? Montenegro-Dalmatia-Neretva or Albania? I'd say the scenario is dubious for the former. Recall the Bronze Age Montenegro sample and consider the region's ruggedness. On the other hand, Albania is more plausible given its harbors and coastal plains favorable to sustaining large settlements of seafaring Minoan-like groups. IIRC, there is elevated J2a there, too.

Sorry for a bit of confusion: like Gioiello, I often include Tuscany under "northern Italy" despite awareness of modern-day genetic variation. I believe that from Beaker times through the Villanovan era Tuscany was in genetic continuity w/Po Valley and/or Liguria, for the most part, as reflected by archaeology. Maybe there is also something to be said about the high freq of R1b-u152 in Tuscany, which isn't really a feature of Umbria or Marche.

Anthro Survey said...

@Alexandros

Yes, the phrase "plus one previously published individual from Diros" is confusing.

Perhaps they made a cardinal error there? Also, why "previously published"? This is something to inquire Lazaridis about. The Diros sample used by Laz in this paper clusters w/out a problem w/Neolithic Anatolian and (northern)Greek samples.

Gökhan said...

Dear David,

Could you add Turkish_Trabzon, Turkish_Kayseri and other Turkish samples into your Fst distances? I would be apprecitiated if you do. I will try to find how can we model Turkish people as Micenian or Minoan ancients via nMonte.

If i miss something please inform me.

AlexDeLarge said...

@1mero171

Sarno et al was a joke of study: now we have the proof that IE languages arrived with the steppe admix in SE Europe: about 21% in Myceneaens and even more in moder SE Europe. A lot of those clowns are not going to sleep tonight.

Anthro Survey said...

@ItalicRoots

Haha You know the funny thing? Due to S. Italy's strong autosomal affinities to the whole Syria/Armenia/(modern)Anatolia area, they suggested that not only Greek, but Italic languages arrived from the East Med area, too. In saying this, they disregarded virtually all evidence linking Italic languages to northern migrations and the widely accepted Celto-Italic linguistic grouping. Talk about an epic fail!

Before I forget, I wanted to correct a major misconception your forum members seem to have. CHG is NOT on a continuum with WHG, SHG, and EHG despite having been Hunter Gatherers. Just like dentists are NOT in the same league with dermatologists and surgeons despite us addressing them as "doctor". Dentists did not go to medical school and have a DDS, not MD after their title. CHG were the eastern counterpart to Natufians. They were a BASAL-RICH group and close cousins to Iran_Neo. Modern middle eastern populations---be they Georgians or Palestinians---are all on a spectrum between CHG/Iran_Neo-like and Natufian/Levant-like, if we ignore extra SSA or steppe in some. The reasons for this are too lengthy to explain in detail here but Chalcolithic era is the likely culprit. Feel free to screenshot this brief-but-elucidating explanation and share on your forum.

AlexDeLarge said...

@Anthrosurvey

That study was so retarded: including all that talking about post Neolitich levantine admixture, despite their SE samples (beside Cypriots) having less of the Red MENA admix than Anatolian farmers!

It actually goes against many dna studies: Paschou et al 2014, Stamatoyannopoulos et al 2017,... and now Lazaridis et al 2017.

Good luck with that stuff!

Davidski said...

@Matt

I had a look at the methods in the paper before I ran my latest tests, and based my outgroups on most of theirs.

At the moment I don't see the need to use more than 16 outgroups. My results are already basically identical to those in the paper.

Anthro Survey said...

@ItalicRoots

The thing is, Lazaridis actually does contend that S. Italy's ancestry cannot be explained by Neolithic farmer DNA alone(2014), but by additional post-Neolithic movements.

I am not going to make confident assertions here as to what those might have been, in what ratio or through what(if any) intermediaries: Minoan and Cypriot-like groups in Bronze Age times or free Hellenized Syrians & Anatolians arriving during Roman rule, etc. We may never tease all of this out adequately. What is clear is that Samnites did not get their language from the East Med, but from migrations further north in the peninsula. I'm gonna go out on a limb and state that there was a sharp genetic contrast b/ween Samnite highlanders, etc. and lowland populations of S. Italy at that time.

In this study, it's abundantly clear that there was CHG-shifted basal-rich post-Neolithic influence from the Middle East into Peloponnese and Crete. I don't see why the same couldn't have occurred in Calabria or East Sicily.

And, btw, areas of North Italy seem to have received some post-Neolithic influence, too, as Simon_W and I discussed, but not to the same degree.

Post-Neolithic basal-rich introgression took place in Iberia as well, of course, but from a "western", CHG-poor source: the Maghreb. This corresponds to their SSA signal. When Mozabites are used to model them, it "eats up" the SSA nicely.

Chad said...

By the way. I did a check on the new Greek sample in Lazaridis et al (2017). It seems that it is less Iranian and HG-like than the Anatolians. The other three, from previous studies, including Rev5, Klei10, and Pal7 apparently are not UDG treated. This isn't the best result, as I'm sure it can be cleaned up, but it will do for showing the obvious. So, using only the new sample, I received the following result.

left pops:
Anatolia_N
Greece_N
WHG
EHG

right pops:
Ust_Ishim
Mota
LaBrana1
MA1
Onge
Karitiana
Levant_N
Iran_N

best coefficients: 0.923 0.050 0.027

std. errors: 0.026 0.031 0.033

fixed pat wt dof chisq tail prob
000 0 5 4.318 0.504624 0.923 0.050 0.027

So, just like the new Peloponnese samples, this sample appears more basal to Anatolians than anything, with Anatolians receiving Balkan HG input. The 3rd edition of the Mathieson et al (2017) paper goes into more detail about how these Greek farmers seem to have an input of 75% from a population basal to the Barcin farmers.

Ric Hern said...

@ Chad

This makes me think that there was a Mesolithic migration of Hunter Gatherers into Anatolia and an adoption of Cattle domestication on their way to North Africa....

Samuel Andrews said...

@Steppe naysayers,

Think about IE languages from an objective position. Just lay out all the facts you know about them. Here's what I know.

-Seven distinct IE languages families are native to Europe; Celtic, Italic, Germanic, Balto-Slavic, Greek, Illyrain, Thracian.

-Only three IE language families are native to Asia; Hittite, Armenian, and Indo Iranian.

-PIE definitely originated less than 10,000 years ago.

Whoever the PIE people were they must have migrated en masse into Europe and only settled in a few places in Asia.

The Sycthian language which belongs to one of the Asian IE languages was spoken by people were genetically European. We can track their ancestors through DNA and archaeology to the Pontic Caspien Steppe in 2800 BC.

People genetically identical to the Sycthian's ancestors migrated en masse into in Europe between 3000 and 2000 BC. They make up 20% of the ancestry of almost every European population.

So is it really just a coincidence the IE Myceaneans were 10-20% Steppe and the non-IE Minoans were 0% Steppe? I mean really!

I know what you naysayers will say ok. "The Hittites, the Hittites."

Big whop, 1 out of like a 1,000 of the literate societies in the Chalcolithic/Bronze age Near East spoke an IE language. If there were literate societies in contemporary LNBA Europe we'd find all kinds of IE languages.

Hittite can easily be explained by a migration from the Pontic Caspien Steppe. Ssaying PIE orignated in the Caucasus but only made big migrations after migrating into the Pontic Caspien Steppe is special pleading.

And to all the Irancentrists saying PIE is from Iran is like saying proto-Semetic is from Europe because of Jews.

Samuel Andrews said...

Edit,
"And to all the Irancentrists saying PIE is from Iran is like saying proto-Semetic is from Europe because of Jews."

Saying PIE is from Iran is like Ashkenazi Jews saying proto-Semetic is from Germany.

Ric Hern said...

@ Chad

Or Mesolithic WHG people were both in the Balkans and Anatolia during the LGM and they were pushed out by the Neolithics with most migrating Northwards into Central Europe and the Steppe...and some in Southeastern Anatolia adopting or inventing cattle domestication and then moved Southwards into North Africa.

Ric Hern said...

@ Chad

Or Mesolithic WHG people were both in the Balkans and Anatolia during the LGM or rather the Oldest Dryas and they were pushed out by the Neolithics with most migrating Northwards into Central Europe and the Steppe...and some in Southeastern Anatolia adopting or inventing cattle domestication and then moved Southwards into North Africa.

AlexDeLarge said...

@Anthrosurvey

You are making a lot of stuff: a brief list of things you got wrong.

1) Sicilians and some Calabrians (South Italy samples from Eurogenes are Calabrians from a small coastal town in province of Catanzaro) do have recent North African admix which is screwing some of those calculators. They have much less of it than Iberians of course.

2) Lazaridis lacked many of those ancient samples including Neolitich/Copper Age/Bronze Anatolians/Greeks/Armenians and Minoans/Myceneans. I've been saying for years that a second Anatolians migration rich in J2 and CHG affected South Eastern Europe thousands of years before the Indo European migrations, which of course was very different from Natufian rich Levantine stuff. I've been also been saying that Cypriots are the closest living group to those Copper/Bronze Age Anatolians, and that pre IE grups in SE Europe were mix of CA Anatolians and EEF. Now look at this study and tell me if I was wrong.

3) IBD analysis from multiple studies fail to find any significant link between Italians and Levantines. There is some sharing with North Europeans, Balkans and North Africans of course, but zero with modern Turks/Middle Easterners. The admixture with North Africa was in both directions because plenty of moslem local converts were expelled to Maghreb during the Middle Ages.

Simon_W said...

@Matt

Europe_LNBA being a broad category its amount of CHG varied considerably; going by
https://docs.google.com/spreadsheets/d/1g-cEJmS9cZ6r-_9p0qj1f2qszxiJbactk89DOd0TlHg/edit#gid=1320636554
Corded Ware_Germany had 18.2% and Bell Beaker Germany only 12.8%.

At any rate, in Table 1 in the paper there are qpAdm models very similar to the one in David's sheet:
Minoan_Odigitria is modeled as 85.6% Anatolia_N + 14.4% CHG.
Minoan_Lasithi is modeled as 84.7% Anatolia_N + 15.2% CHG + 0.1% MA1.
Mycenaeans are modeled as 78.0% Anatolia_N + 17.6% CHG + 4.4% EHG.
The amount of CHG varies in the different models, but this one is closest to David's models in the sheet. And according to the latter, modern Greeks have 13.7% CHG, i.e. less than Mycenaeans and Minoans, not more.

Amanda S said...

There's still a story to be unpicked to explain the common frequency of Y haplogroup E in modern Greeks. We know from the ancient DNA results that this haplogroup has an association with the Southern Levant from before the development of farming. When did men carrying Y haplogroup E start coming to Greece? This paper suggests that this might have been after the Bronze Age but we don't yet have enough data from enough Greek inhabited regions.

Matt said...

@Davidski: At the moment I don't see the need to use more than 16 outgroups. My results are already basically identical to those in the paper.

Yeah, sure, of course, you'll proceed however you think best!

I wanted to see if anyone wanted to discuss what they did differently with qpAdm (both than they normally do and your method), what their results were in general for qpAdm and what their results were in the simultations against qpAdm.

@Simon_W: And according to the latter, modern Greeks have 13.7% CHG, i.e. less than Mycenaeans and Minoans, not more.

I guess another one against modern Greeks being much like Mycenaeans+Steppe_EMBA or Mycenaeans+high CHG West Asian populations then.

That method only gives 27.5% Steppe_EMBA as CHG though, and makes up the rest with 15.6% Barcin_N. So that has a low level of CHG in Steppe_EMBA relative to other published qpAdm in the literature.

Ideally, we would be test Mycenaen and Minoan under identical conditions to Greek, e.g. model Greeks with the full ancient West Eurasian All+ and All++ sets used in Lazaridis 2017, or model Mycenaeans with the outgroups plus sparser ancient West Eurasian set Davidskis used in your link (AG3-MA1, Chukchi, Dusun, Igorot, Karitiana, Kosipe, Kostenki14, Lebbo, Levant_Neolithic, Mbuti, Satsurblia, Ust_Ishim, Villabruna).

AlexDeLarge said...

@Amanda S

E subclades in Natufians/Levantines/Eastern Africans and Europeans separated tens of thousands of years ago. Educate yourself.

Aram said...

Anthro Survey

Thanks for that explanation. Now everything falls in place.

Davidski said...

@Gokhan

The latest Fst dataset doesn't have Turkish sub-populations and I can't add them because the marker overlap is only something like 130K SNPs.

But for what's it's worth, Trabzon Turks usually show very similar Fst patterns to Armenians, while the rest of the Turkish groups behave more or less like the Turks in the current Fst matrix.

Anthro Survey said...

@ItalicRoots

here we go again...

1)IIRC it is the Sicilians who have significant North African admixture, stemming from its geographical proximity to NA and events during pre-Roman, Roman, and Fatimid times. Naturally, it corresponds to the elevated SSA signal in Sicily. Calabrese, Campanians, and Puglesi lack it for the most part. Thus, the "NA throwing off calculators" cannot adequately explain why peninsular south Italian populations are modeled well w/Cypriot, Armenian, or Syrian populations. Nor can it explain the MENA scores they get in 23andMe---otherwise absent in Occitania and Iberia. There is a clear trend across various independent venues.

2)"second Anatolians migration rich in J2 and CHG affected South Eastern Europe thousands of years before the Indo-European migrations, which of course was very different from Natufian rich Levantine stuff"

Isn't it what I had stated? Btw, not thousandS, but a millenium at most. Also, we should not assume it to have taken place as a single, discrete migration.

It seems you're not hearing me when it comes to modern-day genetic variation of the Middle East. Again, CHG was just as basal-rich as "Natufian stuff". That isn't to say they weren't very different because they were. The differences were mainly owing to ANE in CHG and Iran, as opp to Villabruna-like UHG in Natufians. It's also likely due to geographic isolation of the earliest basal-rich groups(ancestral to them) over a long period of time.

To restate my earlier comments more eloquently: there was a significant reduction of this initial bipolar genetic variation in the middle east, come Chalcolithic times. It wasn't as simple as Natufian-like Levantine horizon clashing with a CHG/Iranian horizon: there were Iran_Chalco-like intermediaries playing a role in creating the smooth, modern-day MENA continuum we see today.

In other words, Syrians are not and were not, by Roman times, "Natufian stuff", but a CHG-heavy population not terribly dissimilar to folks north of them in Armenia and Anatolia. Syrians and Georgians aren't that diff at all, dude. The best proxies for Natufians out there are Horn Africans and Maghrebis, in that order, if we subtract SSA out.

There is more J2 in S. Italy, but J1 is quite present in some areas. Moreover, there's no reason to think north Levantine groups couldn't have been J2, as it's a common Hg there. There's also no reason to think it could only have been male-mediated in the long run.

**Btw, in the case of Anatolia, not only was there an increase in CHG ancestry, but also of Levant_Neo-like DNA, to some extent, during and after Chalcolithic times. Again, the overarching theme here is a decrease in the genetic variation of the Middle East---various ancestral lineages pretty much blended over time with mediator populations.

Samuel Andrews said...

@italicroots,

At some point all European yHG E V13 lineages were in the Middle East. Other E clades in the Middle East could have replaced the E V13 that was there. But anyways there is E V13 in the Middle East which could be left overs from the population that gave E V13 to Europe.

ArchTypical European mtDNA T2b, J1c, K1a, and H1 all originated in EEFs from Turkey even though those mHGs are almost never found in the Near East today. What happened was EEF's relatives back home in Turkey got replaced. Just like how R1b P312's relatives back home in Russia got replaced.

Matt said...

@ Simon_W, despite last post, have to say, re: Mycenaeans and modern Greeks, sharing by f4 stats shows some increase in sharing with CHG for all modern Greeks relative to Mycenaeans:

http://www.nature.com/nature/journal/vaop/ncurrent/fig_tab/nature23310_SF8.html?foxtrotcallback=true

I would think that simply comes through Steppe related ancestry, as that's the consistent with the rest of the signal. There could be some mechanism where significantly modern Greeks have less CHG ancestry than Mycenaeans but somehow have more relatedness to CHG via EHG or something, but I don't think that would work. Either way, best to test both via the same method in the same environment to compare directly.

Off topic of my reply to Simon, interesting that Crete_Armenoi is one of the strongest f4 stat differences between modern Greeks and ancient Mycenaeans (though the error bars are large on this one).

Simon_W said...

@Amanda S

The Balkanic E-V13 peaks in the Kosovo and on Crete E1b is far less frequent than J2a (about 34% vs. 11%). So kind of expectable that Minoans turned out as J2a dominated.

Simon_W said...

@Anthro Survey

On anthrogenica I opened a thread on Italian prehistory and early history,
http://www.anthrogenica.com/showthread.php?10654-The-peoples-of-ancient-Italy-(and-their-origins)
and there someone posted a few admixture calculator results from the eastern Emilia-Romagna.

An individual from Ferrara (in spite of being near the Po delta) wasn't southeastern shifted at all: 75.1% Tuscan + 24.9% West_German @ 2.08

On the other hand, an individual from Cesena was: 71.2% Tuscan + 28.8% South_Italian @ 1.81

Even more so than an individual from Rimini: 83.2% Tuscan + 16.8% South_Italian @ 2.01

And having just one grandparent from Cesena I scored 12% Mediterranean Islander (Sicilian/Maltese/Cypriot-based component) on DNA.Land.

And a genetic cousin with parents from Cesena and Oristano (in the middle of the Sardinian west coast) got approximated as:
1 French_Jew_ + Kosovar_ + Kosovar_ + Sardinian_ @ 1.363850
2 Bulgarian_ + Italian_Abruzzo_ + Sardinian_ + Sicilian_East_ @ 1.407580
with MDLP K23b. (Apparently Oristano is somewhat Balkan-shifted, probably from Byzantine influence.)

As for your question what area I had in mind when saying Southwestern Balkans: I'm not sure about the exact affinities of Gaudo, but the affinities of the South Italian Laterza culture were definitely to the north of Albania. But then again Laterza didn't spread to Tuscany. You mentioned the Bronze Age Montenegro sample; its position in one of David's PCAs seemed Iberian-like; but on the other hand I've noticed that in the ADMIXTURE analysis of the Allentoft et al. 2015 paper where this sample was published it does have a lot of Caucasus-affinity, apparently a similar amount as modern Greeks. (But having just seen in my discussion with Matt how misleading ADMIXTURE output may be with Caucasus-affinity, I'm cautious.)

Yes, the high incidence of R1b-U152 in Tuscany is like in northern Italy too, and hardly a legacy of the Etruscans, but associated with broadly Italo-Celtic people from central Europe.

Gioiello said...

@ Samuel Andrews
"ArchTypical European mtDNA T2b, J1c, K1a, and H1 all originated in EEFs from Turkey even though those mHGs are almost never found in the Near East today".

Do you continue to eat shit also after breakfast? K1 and much more belongs to European hunter-gatherers as also Mathieson said (with Y R1) and much other.

"even though those mHGs are almost never found in the Near East today"
Ahahahahahahahahahah

Gioiello said...

@ Simon_W
"Yes, the high incidence of R1b-U152 in Tuscany is like in northern Italy too, and hardly a legacy of the Etruscans, but associated with broadly Italo-Celtic people from central Europe".

Did you get breakfast with Samuel Andrews this morning?
Why do you speak of things you don't know at all?
Northern Italy (but above Tuscany/Liguria and nearby) gets the highest frequency of R-U152 all over the world, and meanwhile you take milk and coffee (with much chocolate) with Shemuel, we (Marco Grassi and me) are testing people in this zone, and we know what happened, not you.

Simon_W said...

@ Matt

"re: Mycenaeans and modern Greeks, sharing by f4 stats shows some increase in sharing with CHG for all modern Greeks relative to Mycenaeans:

http://www.nature.com/nature/journal/vaop/ncurrent/fig_tab/nature23310_SF8.html?foxtrotcallback=true"

Yes, interesting, good point. So maybe the ADMIXTURE output wasn't that wrong after all.

Rob said...

@ Simon_W

"Yes, Steppe_EMBA would do that, but there was no Steppe_EMBA population that postdates the Mycenaeans. And admixture from Europe_LNBA for instance wouldn't suffice to explain the extra-CHG in modern Greeks."

Aren't you forgetting the Slavic migrations ?
They'd be coming from a forest-steppe homeland and be still very steppe genetically , albeit with modest EEF admixture over time.

Simon_W said...

@Gioiello

"Northern Italy (but above Tuscany/Liguria and nearby) gets the highest frequency of R-U152 all over the world"

I don't think so:
http://cache.eupedia.com/images/content/Haplogroup-R1b-S28.gif

And obviously the area where a haplogroup peaks isn't necessarly the area where it originated.

Simon_W said...

@Rob

Yes, we've cleared that up with Matt; the extra CHG in Greeks seems to come from steppe admixed pops, which in this case would mean from the Slavs most of all.

Gioiello said...

@ Simon _W

Don't believe Maciamo. He continues to say that R-V88 came from Middle East after ten years that I demonstrated that it came from the Italian Refugium and now also peer reviewed papers begin to say that.
You, as all the others, use frequency when it satisfy your Germanism, and don't use it when it contradicts your "Blut". We have much more than frequency, and are waiting that your friends tests Italy at least how much Krautenland.

Rob said...

@ AnthroS

"This Levant-shifted population, in turn, would have received CHG-rich admixture and later (Balkan?) steppe admixture to form Mycenaneans."

We need to look at this some more. We have two LN Greek samples, and one (Klei) packs more levant shift, whilst the other significantl less. The latter is a better source for Minoans, whilst the former is better for the pre-IE substratum for Myceneans (somewhat paradoxicaly).
Also, the BA south Anatolians are Levant shifted.

Samuel Andrews said...

@Gioiello,
"yHG R1b is from Italy"
"Indo European languages are from Italy"
"yHG J is from Italy. It's recent in the Middle East"

I've had it up to here with your bs.

https://www.youtube.com/watch?v=e1DDf8sWrGE

Gioiello said...

@ Samuel Andrews
"@Gioiello,
"yHG R1b is from Italy"
"Indo European languages are from Italy"
"yHG J is from Italy. It's recent in the Middle East"

I've had it up to here with your bs.

https://www.youtube.com/watch?v=e1DDf8sWrGE"

Those aren't Italians, but what Simon_W, perhaps also AnthroSurvey and many others think Italians are: i.e. Arabs converted. Certainly they have nothing to do with Tuscans.
R-V88 didn't come from Middle East but from Italy or at lest Western Europe: 1 to 0.
We'll see the rest. Soon I think.
That J is recent in Middle East is proved now: I said not more than 5000 years, now also your compatriot Ted says that it came with the 4200 ya event from Eastern Mesopotamia.
The Arab line not older that 3200 years I said that might have come from Europe. I wouldn't exclude Sardinian mercenaries. We'll see. next. And much other.

Simon_W said...

@Gioiello

You're living in a La La Land. But it's always fun to read your deluded, mad comments.

There are three yDNA samples from Chalcolithic to EBA Remedello di Sotto. Do you know where Remedello di Sotto lies? It's right in the area of maximum R1b-U152 frequency of northern Italy. And all three of them had I2a and none had R1b-U152. How come? The only sensible conclusion is, that R1b-U152 was far less common in this area back then than it is now. In other words: The frequency has changed considerably. And no, I don't think it's wise to believe that R1b-U152 was common in Remedello di Sopra.

The earliest attestations of R1b-U152 so far is from the Bell Beaker culture of Hungary (sample I2365, dated to 2465 - 2205 BC), France (sample I3875, dated to 2459 - 2242 BC), and Bavaria (samples I3589, I3597 and E09569, dated to 2397 - 2149 BC). Moreover, nearly all Bell Beaker R1b samples outside Iberia whose sub-branch of R1b could be determined, belonged to R1b-S116, the parental clade of R1b-U152, while one belonged to R1b-U106, the brother clade. So this makes it clear that the non-Iberian Bell Beaker people had a crucial role in the origin of R1b-U152. It's quite possible that the Bell Beaker individual from Parma also had R1b-U152, at least we know he had R1b-S116. But the thing is: The non-Iberian Bell Beakers had zero evidence for Italian, Remedello-like admixture and, in the words of Olalde et al.: "the great majority of Beaker Complex individuals outside of Iberia derive a large portion of their ancestry from Steppe populations".

And please note that my "Blut" is just 9/16 German; it's also 1/4 Italian and 3/16 Swiss.

Simon_W said...

@Gioiello

"That J is recent in Middle East is proved now: I said not more than 5000 years, now also your compatriot Ted says that it came with the 4200 ya event from Eastern Mesopotamia."

Ridiculous. Satsurblia in Georgia already had J1 at 11430-11180 BC. Another migrant from Italy, lol?

AlexDeLarge said...

@Anthrosurvey

The haplo J1 is of course a mesolitich Caspian/Caucasian thing. I dont even see how this is supposed to be an argument.

A few questions for you "Lebanese slaves" crowd.

1) The absence of post Mycenaean Levantine shift in modern SE Europeans.

2) The absence of Levantine/Natufian E1B1B subclades in both ancient and recent South Eastern Europe.

3) The lack of IBD sharing between modern SE Europeans and modern Middle Easterners.

Rob said...

@ Gio

M269 Looks like its from Black Sea:

L151 centred on NW Europe
L51 generally western Europe
Z2013 Balkans - Caucasus- Near East.
L23 NW Black Sea
M73 NEE - EE

V88 South Europe

Gioiello said...

@ Simon_W

"@Gioiello

"That J is recent in Middle East is proved now: I said not more than 5000 years, now also your compatriot Ted says that it came with the 4200 ya event from Eastern Mesopotamia."

Ridiculous. Satsurblia in Georgia already had J1 at 11430-11180 BC. Another migrant from Italy, lol?"

Ridiculous? Satsurblia is Caucasus, and not Middle East, i.e. Palestine and Arabia, where Semites thought that their hg J (1 and ") was come. Caucasus and Iran (and also Anatolia) were other places than "Semite Middle East". Also the nationalist Ted recognized that. And you not? Who are you for saying and thinking that?

Gioiello said...

@ Rob
"@ Gio

M269 Looks like its from Black Sea:

L151 centred on NW Europe
L51 generally western Europe
Z2013 Balkans - Caucasus- Near East.
L23 NW Black Sea
M73 NEE - EE

V88 South Europe"
Rob, you are a Balkanic, I am Italian. We spoke and wrote each other a lot in the past. I always said to you that I considered Italy and Balkans closely linked. Even though I think that R1b (and possible R1a and much other) were in Italy till the Younger Dryas, after of course migrated in many directions, and of course the Balkans are the closest place. In fact the Villabrunas have been found in the Balkans after the Youger Dryas from Mathieson (who doesn't seem in line with all the harvardians and stanfordians and kurganists and levantinists). I am taking into account from the beginning that also my R1b1a2-L23-Z2110-FGC24408 may have come from Caucasus even in medieval timnes with the invasions... but look at how many Italians are more and more in the Z2110 subclade, and not only in R-V88, R-L389, but also the oldests R-M73 (xM478), not counting R-M269 (we'll see the meaning of the R-M269-PF7562 (x PF7563) found above all in the Eastern Anatolia and Caucasus), not speaking of R-L51-PF7589 (practically at 0% Eastward Italy). I have nothing against the Balkans, but I fought against an origin of R1b from Middle East, and also the kurgan theory isn't convincing to me as that IE language of the centum stock have come from there. R1a, satem languages, Indo-Arians yes...

Carlos Aramayo said...

@ Ariel and Alexandros

I was thinking of D. Anthony`s split of Anatolian around 4000 BC, but maybe, as you Ariel comment, can be too high for IE arrival to Greece. So the first migration to Aegean (not necessarily of Anatolian speakers) but Indo-European indeed, could be around 2900 BC affecting Minoans, and the second around 1700 BC at the beginning of Mycenaean times.

Chad said...

David,

I'm curious if you might be able to doctor your PCA in a way that moves Anatolia to a position between Iran, Natufians, and WHG. I'm not sure if it is a weighting issue or what not. I am just curious if something like that might move other samples around and change the picture of admixture events for us to get a clearer picture for modeling. For instance, using qpGraph, I got the following results for Anatolians:

44% WHG
35.84% Natufian
20.16% Iran_N

Here is the worst fstat:

worst f-stat: Nat Ira Nat Ira 0.499893 0.499936 0.000043 0.003729 0.011

Here is that notation:
root Mbuti
label Anatolia_N Anatolia_N
label Natufian Natufian
label WHG WHG
label Iran_N Iran_N

edge A Mbuti A
edge pIran A pIran
edge Iran_N pIran Iran_N
edge pHunter A pHunter
edge pWHG pHunter pWHG
edge WHG pWHG WHG
edge HG pWHG HG
edge pNatufian pHunter pNatufian
edge Natufian pNatufian Natufian
admix pAnatolia pIran pNatufian 50 50
admix pAnatolia2 pAnatolia HG 50 50
edge Anatolia_N pAnatolia2 Anatolia_N

Here, I've used the 13 Mbuti 5 Iran_N, 6 Natufian, and 24 Anatolians from Lazaridis et al (2016). WHG is Bichon, Villabruna, and Loschbour.

Open Genomes said...

FYI, Gedmatch IDs for previously uploaded Neolithic individuals, with various depths of filtering:


M063398 Bar31 Anatolian Neolithic, Barcin, Turkey, 6419-6238 calBCE
M220828 AH2 Early PPN Tepe Abdul Hosein Iran 8205-7756 calBCE (10215-976)
M249214 GD13A-I1290 Early Neolithic Ganj Dareh Iran 8179-7613 calBCE
M392829 WC1 Wezmeh Cave, Iran 7455-7082 calBCE (9465-9092 ybp)
M417000 Bon002 Early PPN Central Anatolia Boncuklu 8279-7977 BCE)
M423599 I0867 Levant PPNB Motza Israel 7300-6750 BCE (9310-8760 BP)
M510029 tep002 Tepecik-Çiftlik (level 5) c. 6500 BCE (> 8500 BP)
M572712 Klei10 L. Greek Neolithic, Kleitos, 4230–3995 BCE (6240-6005 BP)
M608390 Hotu IIIb Hotu Cave Iran Mesolithic 9100-8600 BCE)
M658611 Rev5 Greek Neolithic Revenia 6438–6264 BCE (8448-8274 BP)
M701826 Bon002 depth>=2 Early PPN Central Anatolia Boncuklu 8279-7977 BCE
M711494 Bar8 Anatolian Neolithic, Barcin, 6212-6030 BCE (8222-8040 ybp)

batman said...

@ Salden,

How can aDNA from IA explain "the last 10.000 years"?

AlexDeLarge said...

@Samuel Andrews

You "Lebanese slaves" crowd are becoming even more annoying than Afrocentrists editing Egypt wikipedia pages for "Muh black Egypt".

"The small presence of E-V13 in the Near East could be better explained by the extremely long Greek presence in the eastern Mediterranean from the time of Alexander the Great until the end of the Byzantine domination over the region during the Middle Ages. It would be unthinkable that over 1,500 years of Hellenisation and Byzantine rule in Anatolia and the Levant didn't leave any genetic trace. In Anatolia, E-V13 is found mostly in the western third of the country, the region that used to belong to ancient Greece. The absence of E-V13 from Central Anatolia does not concord with a diffusion linked to Neolithic agriculture. There is clearly a radiation from the Greece (where E-V13 makes up approximately 30% of the paternal lineages) to the East Mediterranean (where the frequency drops to under 5%)."

http://www.eupedia.com/europe/Haplogroup_E1b1b_Y-DNA.shtml#V13_origins

Rob said...

@ Carlos Ara

"the second around 1700 BC at the beginning of Mycenaean times."

Quite improbable. A migration from the steppe to Greece c. 1700 BC would mean the Mycenaeans spoke Proto-Iranian

Davidski said...

@Rob

Quite improbable. A migration from the steppe to Greece c. 1700 BC would mean the Mycenaeans spoke Proto-Iranian

It would just mean that Proto-Greek was spoken somewhere north/northeast of Greece at that time.

Rob said...

Possible
But would migration event occurred into Greece in 1700 BC ?
Mallory summarised 4
But the two major/ verified ones are 3500 BC and 2200 BC

Rob said...

I'll re- look at the origins of Shaft graves .....

Davidski said...

The Sintashta-like individual from MBA Bulgaria is dated to around that time, but it's unlikely that he was the first of his kind in the region. He may have been one of the stragglers at the tail end of a wave from the steppe that started 500 years earlier.

Rob said...

The mycenean shaft graves are thought to develop from middle Helladic cists and tholoi
So the only marker for 1700 BC is chariot gear; but that's implausible linguistically; and it probably represents an
Transnational elite fashion. It's found in non-IE contexts too (Egypt, etc)..

Rob said...

Maybes
Wasn't this paper supposed to have data from northern Greece / Macedonia too, and down to Iron Age ??

Davidski said...

Yes, but ancient Egyptians don't have Sintashta-like ancestry.

Rob said...

1700 Bc represents the flowering of development, not a post-Blietzkrieg era.
Hence the elaboration of shaft graves, adoption of Status items like chariots and expansion of trade networks as far as Baltic and Scandinavia

Rob said...

Are you suggesting Sintashta spoke Greek ?

Davidski said...

Are you suggesting Sintashta spoke Greek?

No, I'm saying that a steppe population very similar to Sintashta spoke proto-Greek and were the founders of the Mycenaean dynasty.

Based on the presence of Bulgaria_MLBA I2163 in the region in 1750-1625 calBC, they may have arrived in Greece either side of 2000 BC.

Rob said...

@ Dave

"No, I'm saying that a steppe population very similar to Sintashta spoke proto-Greek and were the founders of the Mycenaean dynasty.

Based on the presence of Bulgaria_MLBA I2163 in the region in 1750-1625 calBC, they may have arrived in Greece either side of 2000 BC."

This is not "a steppe population similar to Sintashta in Bulgaria".
I2163 is an individual of the Srubnaya culture (or some such) who made it to the steppe zone of Bulgaria, as per the documented Iranic - Thracian contacts and adstratum influences. But any claim to connect this with proto-Greek as a whole seems like an epic s-t-r-e-t-c-h.

Ric Hern said...

Well the Multi-Cordoned Ware Culture developed out of the late Catacomb Culture and probably mixed with Sintashta like people from further East.

Davidski said...

Then how did random Mycenaeans end up with ~20% of this type of admixture? I could understand samples from royal tombs showing minor foreign input from contemporary far away cultures, but random commoners? Those Iranics must've got around.

Ric Hern said...

Maybe the Steppe Cultures had pockets/clans who looked very similar over a very broad range. So maybe many samples is needed over a very broad range and maybe Cultures was less defined/ connected to certain ancestry than previous much less mobile groups ? The chariot certainly made a huge difference...

Ric Hern said...

Maybe the Steppe Cultures had pockets/clans who looked very similar over a very broad range from +-2000 bC onwards. So maybe many samples is needed over a very broad range and maybe Cultures was less definately connected to certain ancestry than previous much less mobile groups ? The chariot certainly made a huge difference...Maybe this fluidity led to the formation of the East West trade that we see with the later Scythians.

Rob said...

Dave
You're getting 20% Sintashta admixture because you modelled it that way
Minoans were from and developed in Crete, not Central Greece. Using them in your modelling of Mycenaeans is therefore a false premise. You did not use Kumtepe IV, nor did you use any north Balkan late copper / EBA groups. As a result your result has to compensate for the required "Northern input" by coming up with 20% Sintashta.

Davidski said...

My result is fine, because the Peloponnese Late Neolithic samples from Mathieson 2017 will cluster with Minoans, or even slightly south of them.

Rob said...

Where did all the R1a-Z93 disappear to Dave ?
They came, they saw, they conquered, then vanished

Davidski said...

Lots of royal usurpations and founder effects in the Aegean since then, but it's still there on the mainland and even in Crete, even after all of these millenia.

http://www.nature.com/ejhg/journal/v23/n1/suppinfo/ejhg201450s1.html?url=/ejhg/journal/v23/n1/abs/ejhg201450a.html

Rob said...

Haha. The Byzantines loved a good old uncle - nephew usurpation.

Matt said...

Not totally on topic, but talking about mentions of Sintashta interactions with Europe and thinking about the degree to which Mycenaeans could've either a) a lot of a low Steppe_EMBA populations or b) a little of a high Steppe_EMBA population, does get me thinking more about how Sintashta / Andronovo themselves formed.

75:25 Europe_MN:Steppe_EMBA so... scenarios could be:

* 100% Corded_Ware replacement
* 50:50 Bell_Beaker:Steppe_EMBA
* 32:68 Balkans_Bronze_Age:Steppe_EMBA, assuming Balkans BA roughly 20:80 Steppe:EuropeanMN
(and very unlikely and the timing is probably totally wrong but for the sakes of inclusion - 29:71 Peloponnese Mycenaean:Steppe_EMBA, assuming Peloponnese Mycenaean 13:87 Steppe:Anatolian).

I think some of the older qpAdm models showed pretty low WHG fractions in them, so on that grounds the Balkans Bronze Age / southeastern type ancestry might make more sense.

On the other hand those have aged pretty quickly and may suffered from the limited selection of pRight. Using something more like the 16 used by Davidski or All+ or the All++ 30, or maybe the simulations approach might be able to tell us more in future?

Aram said...

Bingo
As I was suspecting the J1a from BA Anatolia is positive Z1828+. This mean it came somewhere from East Turkey which is also confirmed by autosomes if You compare BA Anatolia to South Anatolian Neolithic (Tepecik) and not to NW Anatolia Neolithic/Chalcholithic.
The Mycenean is not confirmed to be J2-M67. But it is just matter of time to see there a J2-M67.
---
Sample Population Date BC Haplogroup
I9130 Minoan 2900–1900 G2a2b2a-CTS946
I0070 Minoan 2000–1700 J2a1d-M319*
I0073 Minoan 2000–1700 J2a1d-M319*
I9041 Mycenaean 1700–1200 J2a1-L26(xabcdeghi)
I2495 Anatolian 2558–2295 J1a2b-Z1828

----------

And another amazing news is that U3b3 was found in Minoan context. We have two U3b3 from Urartian context also. ( Margaryan & Derenko ).
Also another H13a1 is a Minoan. H13 and U3 are known to have high diversity/frequence around Caucasus. U3 was practically absent from Euro - Anatolian Neolithic.

Gioiello said...

@ Aram
"Bingo
As I was suspecting the J1a from BA Anatolia is positive Z1828+. This mean it came somewhere from East Turkey"

The sample is positive for 23 SNPs out of 115:
J-Z1828 Y3475/ZS2787 * Y4412/FGC14010/Z18392 * Y4404/FGC14024/Z18399+112 SNPs 16600 ybp, TMRCA CI 95% 9100 7200 ybp" class="age"formed 18500 ybp, TMRCA 8100 ybp
and the subclade present in Caucasus and India is downstream Z1842 which this sample is certainly negative for, whereas Z18463 is above all European. L620 is very old in Europe, and we'll see if Portuguese Ricardo Costa de Oliveira descends from Iranians (as he thinks) or Europeans (as I think). We are waiting for a Big Y test upon an Iranian. Thus, before exulting, study genetics and all the rest.

Aram said...

Gioiello

You need to be more attentive. I also predicted that it belongs to European branch of Z1828.
http://eurogenes.blogspot.com/2017/08/steppe-admixture-in-mycenaeans.html?showComment=1501751105962&m=0#c6897423329594136032

The Euro branch is the brother of Caucasian branch.So...


Aram said...

The oldest J1 is from Georgia. How many paleo J1 You have from Europe?

Gioiello said...

@ Aram
This is what you wrote above;
"That Anatolia BA J1a is most probably a J1-Z1828 the European branch. The simultanous presence of CHG and Levant means it came from a place which is between Caucasus and Levant. Needless to say what is the name of that place.
So the Out of Mesopotamia is oversimplification! It can be true for J1 P58, but not for other branches like Z1828 who are rooted in North Near East [...] This is the European branch of J1a which has very little to do with Mesopotamia.
https://www.yfull.com/tree/J-Z18463/
Btw on Molgen some commentators predicted it's presence in BA Anatolia a while ago [...] Oops this particular sample seems to be negative to Z1828 but it doesn't mean much because other samples could be positive. Also it is not mentioned to be positive for P58".

I am seeing that you follow Molgen in Russian. It is a pity that they closed eng.molgen, where I wrote more than 2000 letters with infinite researches and data. I should study Russian better for following Molgen, but...
I have nothing against you or Armenia or Caucasus: I am inquiring if also my R1b1a2-L23-Z2110-FGC24408 may have come from there (or if happened the other way around). I dislike biased positions. Perhaps you know that I support that hg. J was amongst the Western European hunter-gatherers in spite of Satsurblia 13000 years ago. We'll see. I am waiting that these harvardians test the Villabrunas beyond the R1b1 found so far.

Alexandros said...

@Rob

I do not see anything wrong with modelling Mycenaeans simply as Minoans + a series of different Steppe-derived populations. I see your point of Minoans not being indigenous to the regions resided by the Mycenaeans, but with the lack of pre-Mycenaean BA samples from those regions, Minoans seem like a reasonable proxy. Generally, I would not expect BA Southern Greeks to be much different from Minoans, or even from people from the great BA Cycladic civilization further to the east. You can try to alternatively model Mycenaeans using Anatolia_BA or Levant_BA, but you will notice that the model fit is much poorer, even in the presence of Greek_N or Anatolia_N.

Alberto said...

@Alexandros

I do not see anything wrong with modelling Mycenaeans simply as Minoans + a series of different Steppe-derived populations.

But why steppe-derived? Did you miss the samples from the latest Mathieson et al. paper about the Balkans? Did you miss the Kumtepe samples? And why ignore the south Caucasus ones?

I also don't understand what's the point of the Sintashta model. Apart from showing that Minoans share a good amount of drift with Mycenaeans but the latter plot to the north east of them, so adding a population in that direction like Sintashta makes for a decent 2-way statistical admixture model, I don't think you can read much more into it.

Rob has already pointed out all the problems with that (linguistic and more). I'd add that the ethnogenesis of the Mycenaean population happened around 2400-2000 in the SE Balkans probably*, which makes it chronologically impossible. Plus there's the question of R1a-Z93 too. With a 21% Sintashta-like admixture of mostly male elite, you should expect some 40-50% R1a-Z93 in Mycenaneans, something which is totally unrealistic. Common sense says that the lineages should be a mix of Balkans and West Asian, which could include some R1b-Z2103 that was in the Balkans early enough, but that's it.

* Archaeologically, I'd recommend this paper (IIRC first recommended by Rob himself) by Heyd et al. about the processes taking place in the relevant area to see that indeed aDNA is aligning with archaeology quite nicely:

https://www.academia.edu/28958137/Kanl%C4%B1ge%C3%A7it-Selimpa%C5%9Fa-Mikhalich_and_the_Question_of_Anatolian_Colonies_in_Early_Bronze_Age_Southeast_Europe

Alberto said...

OT, but possibly of interest to Y-DNA savvy people:

Analysis of the R1b-DF27 haplogroup shows that a large fraction of Iberian Y-chromosome lineages originated recently in situ

https://www.nature.com/articles/s41598-017-07710-x

Interesting that apart from the time of expansion of R1b-Df27 (3500-3000 y.a) they could also detect the more recent expansions of other HGs that happened all at the same time, more or less coinciding with the Germanic-Alanic invasions.

Davidski said...

@Alberto

The Sintashta+Minoan_Lasithi or Steppe_MLBA+Minoan_Lasithi model is very solid statistically, and it has backing from direct evidence from ancient DNA, because in Mathieson 2017 we have a Sintashta-like male in MBA Bulgaria and Late Neolithic samples from mainland Greece that resemble Minoans.

So if you'd like to challenge this model, then by all means, but make sure that your alternative model is also statistically very sound and has support from ancient DNA.

How many Armenian_EBA-like samples are you seeing in the Bronze Age Balkan data, if that's your preferred angle?

Anthro Survey said...

@Simon_W

One other Rumagnol acquaintance of ours is essentially modeled as mostly Tuscan+Bergamo(he's 3/4 Rumagnol, 1/4 Ferrarese). Thanks for sharing those, btw. It satiated my curiosity and now I can see his result is quite normal.

As for why the Ferrarese guy you shared(and the sample from Fiorito et al) differs? A simple theory: dilution of the original stock by settlers from Lombardy. It's noteworthy, after all, that Ferrarese folks don't speak Rumagnol language, but Lombard("Emilian"/southern variety). Just as noteworthy is that Ferrara area wasn't as enfranchised during Roman times, whereas Rumagna's importance led to greater urbanization and optimized population growth.

Think of how R1b was practically replaced by R1a in Ukraine/Romania and is now found in high %s in W.Europe. In the same way, Minoan-like DNA could have radiated outwards from the Delta epicenter, only to be diluted at its original source later.

To test this idea further, we should get an idea of how Polesine/Rovigo and Padua score. If you come across any, let me know. The closest I've come to was finding someone of 50% Polesine ancestry.

As for the MontenegroBA sample---does it show steppe affinity or just CHG-like? If the latter, then it's actually not impossible. This would merely imply a modeling scheme of Chalcolithic_Iberia-like+Basal rich_CHG if we go by simple PCA geometry and assume negligible steppe ancestry. But you're right----Admixture output should be approached with caution!

Re E-v13 in Kosovo: classic founder effect. Ottoman-era. Dienekes devoted 1 or 2 blog posts to this. Essentially, a handful of Gheg settlers from the mountains rapidly expanded over 4-5 centuries. This process was facilitated by extensive privileges granted by the Sublime Porte and an efflux of Serbs. A large number evacuated in the 1600s because they cooperated in an Austrian-led campaign that nearly removed Ottoman power from the Balkans. I'll should mention that a Tuscan general was in charge for much of it, so Gioiello can sleep well tonight. :D
Ghegs from the Albanian highlands likewise have high %s of it, but not like in Kosovo and more on par w/Thessaly instead.

Anthro Survey said...

@Rob

Paradoxical indeed.

What do you think about the idea of a Cilicia-SouthAnatolia-greece coastal corridor Aram and I have entertained and its role as a pipeline for more levant-shifted dna?

Palacista said...

@Batman,
The Elamite-Dravidian connection is at best speculation, a connection to Afroasiatic is fantasy. That J2 was found in a context that was known to be pre IE.

batman said...

1Mero171 said...

"There was a large migration spreading the ancestry CHG/Iran and Ydna J in the eastern Mediterranean during the Bronze Age and this needs to be further investigated. Note that the Egyptians Schuenemann et al. 2017), Levantines (Lazaridis et al., 2016), Anatolians, and Caanites (Haber et al., 2017) were found to possess this mixture from the Bronze Age. Now we have the southern Greeks as well. The shift of Minoans and Myceneans on PCA with respect to CHG/Iran may be explained by the mix with people from this massive migration. Mathieson et al. (2017) also noticed this shift in some samples from the Balkans."

There are historical records describing a wave "ex oriente" into Mesopotamia around 2.300 BC, known as the 'Akkadian invasion' - that giave rise to the first known military weapons and armies able to conquer larger populations.

Facing a Sumerian population of farmers and traders the self-appointed masters of war could cut through Mesopotamia and conquer eastern Anatolia as well as Syria and the Levant - from where they could even reach the Greek islands, as far west as Crete (Keptara/Keptiu).

https://en.wikipedia.org/wiki/Akkadian_Empire

Due to the linguistic connection between the elamo-dravidian and the semittic languages it's been theorized that the origin of the Akkadians were "Indo-Iranian" and/or 'Indo-Aryan'. Following the spread of y-dna J2/J1 one may find the same migrational wave, from the indo-iranian area east of the Suleiman mountains. Before this intrusion it seems that the old, helladic y-dna G2/G1 was the predominant y-line between Crete and Tigris.

After the Akkadian Wars and the Thera eruption (3.700+ BP) the West-Minoan/East-Minoan capitols on Crete was abandoned. Seemingly the rulers of the G2-dynasties - pictured were moved to the safety of the northern mainlands of the Med - initiating the (later) capitols of Rome and istanbul, with the royal lines of Jupiter and Zevs were as (indigenous) rulers of the Western and the Eastern Meds, respectively. Untill the Trojan wars and the Phoenician and Venetian take-overs...

After devastating wars the Akkadian invasions were halted, allowing a period of "repair and restoration" of social life, economical production, common laws and civil order. The re-constitution of Hamurabi in Mesopotamia appears simultanoiusly with the 'Minoan-Mycenean' rennesance that reached Pelepones and Crete from the Greek mainland, around 1.600 BC.

https://www.nature.com/articles/ncomms2871

The 'Mionan rennesance' were obviously accompanied with some (peaceful?) immigrations from the north - such as Balkan and northern Europe - constituting the Mycenean period. This later influx may explain the appearance of y-dna I2/I1 and R1 in southern Greece - as well as the striking connections between the Mediterranean, the Balkan and the Baltic bronze-ages.

https://jarnaes.wordpress.com/1-minoan-crete-linear-a/

Consequently we may explain the large intrusion of J2/J1 into BA Greece - as a consequence of a "Indo-Arian' intrusion into the old, Greco-Arian dynasties of G2/G1. - with the Akkadian conquest of Asia Minor, known to have reached Crete ('Kaptara') already during the regime of Shar-rukhin. When the cultural rennesance and Mycenean Greece is constituted there were new European y-lines represented - reinforcing the European origins of the Minoan culture.

Rob said...

@ Alexandros

"I do not see anything wrong with modelling Mycenaeans simply as Minoans + a series of different Steppe-derived populations"

Theoretically it isn't incorrect, but it's more of a steppe admixture (perhaps via a couple of sources) rather than derivation.

Alexandros said...

@Rob
Sure, I see. No problem with that. I am not suggesting it is Sintashta per se. Could be a population related to them. I also theoretically do see value in an Armenian influx (the authors themselves put these two hypotheses as predominating), but I tend to believe more the steppe-derived ancestry.

@Alberto
"With a 21% Sintashta-like admixture of mostly male elite, you should expect some 40-50% R1a-Z93 in Mycenaneans"

I do not fully get your reasoning here. Can you please elaborate a bit more? Generally, it is way too early to say anything about Mycenaean Y-haplogroups. The authors provide data just for a single male, who happened to be J2a! Basically, the only thing we know at this point in time is that J2 was present in Mycenaeans, nothing more nothing less. I would not be surprised if we see R1a, G2a, or even E-V13 in subsequent samples. I wouldn't put my hand in fire for the latter though..

Thanks for the book suggestion btw, I will surely go through it.

postneo said...

"I do not see anything wrong with modelling Mycenaeans simply as Minoans + a series of different Steppe-derived populations"


Nothing wrong unless you take it in to account.


Minoans are an island population and will always show a distance from the mainland at any point in time.
From the PCA we that Over 3000 years Greeks have slowly become more steppe/mainland/Balkan like vs the initial Anatolian like founding populations on the islands. Modern Greeks are more sintashta vs Mycenaeans.

But thru all this they remain Greek speaking. The Macedonians were surely more "steppe" than athenians but they were still carriers of Greek culture.

Davidski said...

@postneo

Take a look where pre-Mycenaean Late Neolithic Peloponnese (mainland Greek) samples cluster.

http://eurogenes.blogspot.com.au/2017/05/steppe-invaders-in-bronze-age-balkans.html

So in your opinion how did the Peloponnese Mycenaean genotype form, considering that it's shifted by about 20% towards the steppe?

Rob said...

@ Alexandros

"Sure, I see. No problem with that. I am not suggesting it is Sintashta per se. Could be a population related to them. I also theoretically do see value in an Armenian influx (the authors themselves put these two hypotheses as predominating), but I tend to believe more the steppe-derived ancestry."

Yet that's my main point - I suspect there's both. Not that I thought one was more important than other.

However, I do note that Myceneans have little if any direct WHG

Mycenaean
Barcin_Neolithic:I1099 68.7 %
Kotias:KK1 19 %
Levant_Neolithic:I1704 5.7 %
Karelia_HG 5.2 %
AfontovaGora3:I9050.damage 1.4 %
Villabruna:I9030 0 %


Yet Sintashta has very high WHG ratios.

Sintashta:RISE395
Kotias:KK1 30 %
Villabruna:I9030 28.4 %
Barcin_Neolithic:I1099 20.8 %
Karelia_HG 20.8 %


Kumtepe_LN:kum4
Barcin_Neolithic:I1099 46.6 %
Kotias:KK1 19.7 %
Karelia_HG 18.3 %
Levant_Neolithic:I1704 12.9 %
Yoruba 2.1 %
AfontovaGora3:I9050.damage 0.4 %
Villabruna:I9030 0 %

If Sintashta is the source of ANE admixture in Myceneans, there should be WHG associated with, which there isn't. Instead it seems that Myceneans and Kumtepe IV received it via a source which was exclusively CHG & EHG (+/- ANF), and no WHG, thus not via Sintashta, CWC or Yamnaya, if such a hypothetical population existed.
But we do start to see such a combination admixing c. 4200 - 3600 BC around the place: Varna, Khvalynsk, Dereivka, Kumtepe IV, ..

Davidski said...

@Rob

However, I do note that Myceneans have little if any direct WHG

Barcin is eating up the WHG in your model, because you're lacking the more relevant basal Peloponnese Neolithic samples.

Alberto said...

@Davidski

But honestly I don't know if you are serious about the Sintashta thing. From all the steppe groups, Yamnaya (or pre-Yamnaya) are the only likely ones to have had some input into Mycenaean Greeks.

Surely I can get better fits with Global 10 data, but unfortunately we don't have the Mathieson samples yet, so I still need to add some Neolithic Europe from further away:

Mycenaean
"Minoan_Lasithi" 45
"Kumtepe_LN" 23.65
"Esperstedt_MN" 11.6
"Anatolia_ChL" 8.3
"Yamnaya_Samara" 6.95
"Jordan_EBA" 4.5
"Sintashta" 0

Which is more realistic nevertheless.


@Alexandros

I don't know if you have followed other ancient DNA papers from the BA, but the pattern we see is that the R1 expansions have a much higher impact n the Y-chromosome than in the autosomes. Double or more. That's what I meant by saying that if Sintashta people had a 21% impact in the autosomes, we should expect some 40-50% (or more) in the Y-chromosome. Which for Sintashta happens to be R1a-Z93, very rare in Europe. And it's just too late to have anything to do with the genesis of the Mycenaean culture or language.

Indeed, we only have one Y sample, plus then we have other samples from the Balkans. And we have the autosomes. And modern DNA. So put all together and you're right to expect things like J2 (confirmed), J1, E-V13, I2a, G2a or R1b-Z2103. There is R1a-Z283 in Greece today, but I guess we all suspect it to be from Slavs.

Davidski said...

@Alberto

Yes, I'm very serious about the Sintashta thing, although let me just reiterate that this was a culture closely related to Sintashta and not Sintashta per se.

Let me also reiterate that I'm looking at statistical fits based on raw data, not PCA data. So I don't agree with you that Sintashta isn't a good fit, because I can see that it is.

And I don't have a problem with the low frequency of R1a-Z93 in Greece today. There may have been a time when R1a-Z93 did reach high frequencies among Mycenaeans, and then almost disappeared.

What happened to the Corded Ware R1a-L664 in Bavaria, because there's not much of it around now? In fact, in Poland almost all of the R1a can be traced to the Slavic expansions, and not directly to Corded Ware.

Steven said...

Could the northern element in Cretan_Armenoi (Mycenaean) be related to I0115, Esperstedt, Germany; ESP3, 1931-1780 BC? This sample carries a lot of eastern European DNA and east Germany has a lot of I2a2 and R1a. Could the southern element in Cretan_Armenoi be related to Natufian? The Cretan_Armenoi has a lot of southwest Asia and north Africa on Dodecad calculators as well as eastern European, but no Gedrosia. It doesn't look like it carries any Yamnaya.

Jaydeep said...

In support of arguments put forth by Rob & Alberto, I would like to make a few observations:-

The argument for a steppe input into Myceneaeans, is based on the modelling by Lazaridis et al of Mycenaeans as a mixture of Minoans from Lasithi and either one of steppe_MLBA, European_LNBA & steppe_EMBA.

However, if this modelling was correct, then the Mycenaeans should share more alleles with either of these 3 groups in comparison to the Minoans. However, as Extended figure 2.D shows, this simply is not the case.

Moreover, as Rob has noted, the Mycenaeans are not descended from the Minoans. Therefore this is one more reason to doubt this model of Mycenaeans as a mixture of Minoans & European_LNBA/steppe_MLBA/EMBA.

-----------------------

The authors have also tried to model the Mycenaeans as a mixture of Anatolian Neolithic & the steppe populations but it does not work because, in their own words Mycenaeans have more Iran-related than EHG-related ancestry . A two way model of initial Iranian related ancestry and a later steppe related ancestry might help to resolve this and this is why the model of Minoans + steppe groups works. But as I have pointed out, there is no excess of allele sharing between Mycenaeans & any of the steppe admixed European groups in relation to the Minoans. Hence this model is also a failure.

-----------------

The most sensible model, therefore, is of modelling Mycenaeans as a mixture of Anatolian_N + Armenian_Chl/MLBA, as this model also eliminates the need to posit multiple admixtures.

Armenian_Chl was modelled by the Lazaridis team of 2016, as a mixture of Anatolian_N + Iran_N + 29 % EHG. There is no evidence that this EHG admixture necessarily comes from the steppe. But it could have come from the East. Iran_Hotu, at about 10,000 BP, can already be modelled as Iran_N + 9 % EHG. This suggests that already at the start of the Holocene, populations towards South Central Asia may have been Iran_N like but with more EHG/ANE type ancestry.

A movement from South Central Asia can therefore explain the EHG percentages in Armenia_Chl. It may also be worthwhile to note that all 3 samples of Armenia_Chl were ydna L1a, a HG most common today in South Asia. Further, in the 2015 Haak et al paper, which first postulated a Near Eastern Armenian-like admixture into the Yamnaya, the best model for admixture into the Armenians was LBK_EN + Sindhi (Z score - -15.7)

Hence, in most likelihood, Armenia_Chl did receive admixture from a population from South Central Asia which could then have passed on the Mycenaeans. It may also be noted that in the Extended Figure 2.D, Iran_Hotu shares slightly more alleles with the Mycenaeans than with the Minoans. This may not stand out as significant on its own but taking it together with the other factors, suggests that the extra steppe/EHG affinity of the Mycenaeans could be a result of a eastern admixture, from the east of Iran_N i.e. South Central Asia.

-------------------

Lest one wonders, there is archaeological evidence of Central Asian influence in the Maykop culture. This is a matter of record.

Read here :- https://www.dropbox.com/s/gzn50m89zkhurmt/ivanova_2013a.pdf?dl=0

Go through the chapter 4 - The Valley of the Lower Kuban.

If you're interested in the linguistic argument, read the work of Joanna Nichols -

https://www.scribd.com/doc/150388032/The-Epicentre-of-the-Indo-European-Linguistic-Spread-Useful




Matt said...

Looking at Global 10, looks like models that would work OK for Sintashta:

Bell Beaker Germany:Yamnaya Samara - 50:50
Corded Ware Germany:Mycenaean - 90:10

Former seems more likely.

Models with Yamnaya Samara:Mycenaean don't look to work for Sintashta. Looks like too CHG shift.

Mycenaean seem like they model with Minoan OK with either Sintashta or Yamnaya. Looks like cannot be modeled with Minoan and Europe LNBA, and that this would need an exaggeration on GreekLN->Minoan cline to work.

Crete_Armenoi looks like she combines either Mycenaean:EuropeMN / Minoan:Bell Beaker Germany or Yamnaya:GreekLN. Mycenaean/Minoan with Yamnaya or Sintashta doesn't work at all.

Modern Greeks look to model OK with Bosniak / Bulgarian plus Mycenaean. Models with Polish / Ukrainian+Mycenaean look slightly less functional.

Chad said...

Greek_N, the new one, plus Anatolia BA, and Euro LNBA is the way to go. The West Asian stuff came early on. Well before Mycenaeans. Greece became very much influenced by Anatolia BA, in the early Helladic. Mark my words. We'll get samples that show Greek is from the North.

Chad said...

Honestly, I wouldn't be surprised if pre-Mycenaean people were closer to Anatolia_BA than Minoans.

Anthro Survey said...

Jaydeep,

When two populations consist of relatively similar ancestral components, by way more recent ancestral nodes no less, we should not expect any groundbreaking results from d-stats. It's one thing with (Mycenanean, Minoan; Yoruba, Chimp) or (Loschbur, Gokhem; Natufian, Dinka), but quite another when X and Y are Minoans and Mycenaneans. And, not to be Captain Obvious, but if Mycenanean do have steppe and, in turn, EHG(which I'm 90% sure of), we can both agree it's not much. Throw in a bit of drift to all of this and you're in for results that downplay the extent of what transpired.

Mycenaneans modeling as Neolithic Anatolians and Armenia_MLBA can be interpreted as an averaging out of steppe-admixed Balkaners and, say, Anatolia_Chlc ancestries on top of the Neolithic substrate. This is a much, much more parsimonious scenario as I'm sure you realize. You can visualize this idea with some rudimentary vector math on the PCA, too.

So, if this tri-ancestral scenario applies, if the steppe:CHG_rich ratio was pretty even and if the CHG_rich ancestry had pseudo-EHG affinity like Armenia_MLBA, do you honestly think formal stats would be screaming?

Besides, look at Crete_Armenoi again. Try as you might, but you can't explain it all away with Armenia_MLBA. Clear Bronze Age European affinities there and huge difference w/Minoan samples. Now, if Crete was affected thus, what would make anyone think these steppe-carries would have skipped over Greece like grasshoppers? lol
_____________________________________________________________________
"This suggests that already at the start of the Holocene, populations towards South Central Asia may have been Iran_N like but with more EHG/ANE type ancestry."

And yet both CHG and Armenia show more of this pseudo-EHG ancestry than Iran_n despite the latter being positioned in a more SC-wardly direction. Btw, I say pseudo-EHG because this is due to independent combination of ANE and Villabruna-like ancestries in the region, as opposed to an actual influx of EHGs from Belarus or smth.

As for Hg L? Should we make the same argument about R-v88 coming out of Africa using this logic? V-88 originated in either the mid-east, NA or Europe, areas shaded with the lightest hue of red on its frequency map. Likewise, consider how R1a almost entirely replaced R1b in Ukraine/Romania, its former abode and source region(for relevant subclades). In the same way, L's origins probably lie somewhere in Iran/Caucasus and its high frequency in Greater India attests to a Neolithic expansion from Iran. Its low frequency in modern Iran/Caucasus indicates replacement by J2, J1, etc. at sometime after the neolithic.

Re/ Sindhis---They are the "Sardinians of South Asia" when it comes to (iran-related)Neolithic ancestry and not particularly ASI-rich. Combine this with some western Basal-rich component like LBK/Anatolia and voila----Armenians. Grade school math.

So far, all available genetic and archaeological evidence points to South Asia being a genetic SINK, not source. It doesn't sit well with some, I get it, but there is no shame in accepting this, for it doesn't diminish India one bit.

Rob said...

Dave
Are the peloponesse samples labeled "Greece EN " in your data ?

Davidski said...

I don't have them in my data because they haven't been released yet. They'll be published along with Mathieson 2017.

I'm pretty sure they were sampled to prove directly that a Minoan-like population lived on the Greek mainland prior to the Mycenaeans, but they didn't make the deadline for the Mycenaean paper.

Chad said...

They're shifted toward Levant_N from Anatolian_N.

mike said...

Where does Z2103 in Greece fit into your model David?

Chad said...

Hmm. Those Greek farmers do look like something from maybe SC/SE Anatolia. Funny thing is, Mycenaeans look like they'll get a good upswing in Europe LNBA/Steppe EBA with those.

Davidski said...

@mike

There are probably layers of steppe influence in Greece, but I think that the most important wave from the steppe arrived in Greece around 2000BC, and it brought both Z93 and Z2103.

Steven said...

@Davidski

Why is Z93 not found in modern Greece?

Davidski said...

@Steven

Z93 is found in Greece, it's even found on Crete.

Arza said...

@ Davidski

I have two questions:

1. Can you rerun qpAdm test with Corded Ware samples restricted to RISE00, I1532, I1542 and RISE431?

2. Do such high std. errors disqualify the model?

chisq: 2.294
tail: 0.999513477
best coefficients:
0.635 0.116 0.249
std. errors:
0.208 0.087 0.255

Davidski said...

@Arza

Yeah, it looks like you've got a high correlation between refs 1 and 3 which is screwing up the model.

Here's that CWC model...

https://drive.google.com/file/d/0B8XSV9HEoqpFbThZOFFDcm54U00/view?usp=sharing

Chad said...

With where those Peloponnese samples sit, they'll change a lot of stuff. I can't wait to get my hands on them.

Rob said...

@ Matt

RE: Sintashta

I thought this is interesting.

Sintashta:RISE395
Armenia_Chalcolithic:I1631 46.6 %
Latvia_HG:ZVEJ32 26.8 %
Ukraine_HG1:StPet2 20.55 %
Kotias:KK1 6.05 %
Loschbour:Loschbour 0 %
Hungary_HG:I1507 0 %
Samara_Eneolithic:I0434 0 %

I think the Majkop genomes will be interesting.

Rob said...

@ Davidski

"Z93 is found in Greece, it's even found on Crete."

It is intersting that from a sample of 171 mainland Greeks, there was zero Z93
Pooling the data from 1000 Balkan-ers and Crete, there were 5.

Rob said...

A slightly zoomed in PCA of Greek , Anatolia and NW Asian samples
http://imgur.com/8M43rbj


Some broad patterns:

1) South Anatolia : the Tepecik to Bronze Age shift is toward Caucasus

2) Greece: if we take LN as a starting point, the Myceneans are shifted as if toward Maros (ie a Balkan groups with some steppe admixture), as well as Caucasus-ward. The latter Mycenean plot quite close to Kumtepe IV.

3) West Anatolia : The barcin Neolithic - Chalcolithic - Kum 4 sequence. : Again Caucasus -ward, broadly. BUT if we take our starting point at barcin Chalcolithic and Kumtepe 6, the shift of Kum 4 is toward something toward a steppe-admixed Balkan EBA.

Steven said...

@Rob

I agree. I looked at the R1a haplogroup project on FTDNA and only saw one Greek who was Z93 positive (there are over 6,000 members of the project).

Davidski said...

@Rob

Maykop has nothing to do with Sintashta.

And for Mycenaeans the only reference samples that really matter are Peloponnese Neolithic, Anatolia_BA and Bulgaria_MLBA (or perhaps Europe_LNBA).

@Steven

Look here.

http://www.nature.com/ejhg/journal/v23/n1/suppinfo/ejhg201450s1.html?url=/ejhg/journal/v23/n1/abs/ejhg201450a.html&foxtrotcallback=true

Arza said...

@ Davidski

Thank you!

I see now that you removed not only few samples from Anatolia_N but also I0124 (I presume) from EHG - due to this I had a worse fit and slightly different proportions.

high correlation between refs 1 and 3 which is screwing up the model.

Minoan_Lasithi and Anatolia_BA. The middle one was CWC (this time RISE431, I1540, RISE436). In another set with Greece_N (slightly smaller, but still too high std. errors) again the "RISE00-set" was preferred, just like in the 2-way model.

The logic behind this:
https://s6.postimg.org/5wor22kc1/CWCaean.png
1. LNBA and MLBA is preferred over EMBA - source "below" Yamna
2. Sintashta over Srubnaya - source "below" Srubnaya
3. check if something can produce better a fit than Sintashta

If you're bored enough you can play like this.

In a 2-way model "RISE00-set"+I1540+RISE436 had worse fit, so it was too far away.
3 overlapping Sintashta samples also had a worse fit.
Same with 6 overlapping Unetice samples:
I0115 F Europe_LNBA
I0803 F Europe_LNBA
I0164 F Europe_LNBA
RISE109.SG F Europe_LNBA
RISE150.SG F Europe_LNBA
RISE154.SG F Europe_LNBA

Now I wonder how would PL_N17 work. RISE598 should also plot somewhere in the neighbourhood.

Here is something interesting (that would be also valid for Sintashta):
https://s6.postimg.org/snxtoh5dd/myc.png
Moderns in the Balkans seem to form a 3-part zig-zag. Two-way model between Minoan and Sintashta/CWC can create the middle part of it, lets say a Mycenaeano-Greaco-Macedono-Thracian continuum. Add Slavs from the left, Cypriots from the right, and voilà, modern distribution.

And at the end something hilarious:
the model 79%Minoan_Lasithi+21%Europe_LNBA tends to share more drift with Mycenaeans (at the |Z|>2 level)

MinMyc.ind
RISE568.SG F Europe_LNBA
RISE569.SG F Europe_LNBA

Ooops, they did it again...

Rob said...

@ Davidski

" for Mycenaeans the only reference samples that really matter are Peloponnese Neolithic, Anatolia_BA and Bulgaria_MLBA (or perhaps Europe_LNBA)."

Utter nonsense.
Yeah let's pick and choose one sample from 7000 BC which will maximise the otherwise virtually non-existent admixture from Sintashta.
Come on- dig up some Z93. Uniparentals don't lie, but people can come up with also sorts of crap, ahistoric models

Samuel Andrews said...

Rob just likes to do his best to debunk anything that has to do with Steppe migration and a Steppe origin for Indo European languages. That's why he puts his support behind any data that suggests an alternative scenario.

I think he might do this because he thinks us Steppe-supporters are big fat racists. It's that or something else emotional and unrelated to logic. I'm not even sure he believes half the things he posts.

He's been dis proven before, he'll be dis proven again.

Rob said...

I love it when the lobotomised try to psychology

Davidski said...

@Rob

Here are the dates for the Peloponnese_Neolilthic samples.

5479-5338 calBCE
5000-3200 BCE
5000-3200 BCE
5000-3200 BCE
4043-3947 calBCE

postneo said...

@David

Bulgaria is closer to the steppe than Mycenia so we should expect even higher steppe contribution.

10-20 % higher steppe higher steppe and a gradually increasing steppe component is only to be expected as you go closer to the steppe.

What we need is temporal sampling of the same area that has little steppe and gets a sudden temporal pulse.
Thats is what is needed to prove the chariot borne version of IE spread.

Samuel Andrews said...

I passed psychology class so I think I know what I'm talking about. You're a classic example a stupidity.

Davidski said...

@postneo

Bulgaria_MLBA I2613 is 100% steppe, because he's starkly different from most other Balkan samples dating to similar periods, but indistinguishable in terms of genome-wide DNA from contemporary samples as far as the Urals and even Altai (Andronovo, Sintashta etc.). He also belongs to Y-HG R1a-Z93 and mtDNA U5a.

So he's obviously a new arrival, or a descendant of recent arrivals, in the Balkans from the steppe.

Do you now finally understand that your isolation by distance argument doesn't work and can you find it in yourself to move on from it?

postneo said...

@ Sam Andrews

"I think he might do this because he thinks us Steppe-supporters are big fat racists"

No because your reading of data is one sided. You just said that Asia has only Hittite and Indo-Iranian and Armenian

You ignored Tocharian for starters. Also by clubbing Indic and Iranic you are ignoring diversity. An artificial scenario created by linguists by ignoring modern languages in Asia as a diversity measure whereas they count 2000 younger languages in Europe. Its not their fault but an artifact of data preservation.

Dardic and Sinhala are hugely diverged languages for example but they are ignored because of the primacy of Vedic. This is probably the case with Avestan as well. in Europe no counterparts to Vedic or Avesta exist except for Mycenean so for the rest they just pick up highly diverged daughters.

This has been noted by other scholars... At the the end of the day this sort of stuff is fuzzy and can be argued either way

Samuel Andrews said...

@postneo,

"This has been noted by other scholars... At the the end of the day this sort of stuff is fuzzy and can be argued either way"

To me it isn't fuzzy because the Anatolian EEF migration in 5000 BC was the last Asian migration that affected all of Europe.

There was no migration directly from anywhere in Asia into all of Europe after 5000 BC. There's no trace of Chalcolithic South Asian or Anatolian or Mesoptamian or Caucasian ancestry in (North) Europe). Instead we see a large scale migration directly from the Pontic Caspien Steppe between 2800 and 2300 BC, that carried loads of Caucasus ancestry, into Europe.

If PIE originated in Asia there are two explanations for the ancient DNA data.

1. The Caucasus ancestry in Steooe_EMBA came with PIE language. Then IE spread out of the Steppe into Europe and into Asia with Indo Iranian.

2, IE languages did spread directly from somewhere in Asia into Europe but made no genetic impact.

I say just go for the simple answer; PIE originated and spread out of the Steppe. The extra layer of Iran Neo in Sycthians is interesting. One could argue that's from their Iranic ancestors. I just don't think so though, it looks it has an independent origin we're not familiar with.

Speaking of Tocherian, the guy on the left was Tocherian. He looks like lots of people I see at Wrigley Field. He isn't a coincidence, I think he strongly indicates Tocherian is also from the Steppe.

https://upload.wikimedia.org/wikipedia/commons/b/b8/Central_Asian_Buddhist_Monks.jpeg

Rob said...

@ Dave

Okay, it'll be good when we get the Pelopon Neol included too.


@ Post-Neo

As you said, what matters is language diversity and structure of the *tree. So it doesn't matter if 99% of languages are in location 'X", what matters is the splits.
And the earliest splits are (1) Anatolian (2) Tocharian (3) The Greco-Armenian node (4) Indo-Aryan (5) the north European IE languages.
Some would argue that PIEs were just sitting on the steppe sending out periodic migrations to account for this, but I'm somewhat skeptical.

Samuel Andrews said...

@Rob,
"I realise you might have a rather big chip on your shoulder and need to construct some grandiose ancient history around you to compensate"

See this is what I was talking about. You assume Steppe supporters are losers who need awesome ancestors to feel good about themselves. It's a ridiculous assumption. I'm not the mentally disturbed weirdo you think I am.

"So do try to hold it together and not get so triggered every time I point out the holes in some not so well thought out ideas"

I get triggered mainly because you've proven to be a jerk. Notice I don't yell at Gioiello very much even though his ideas are even more out there. It's because Gioiello hasn't ever put his nose up to other posters.

"You do realise the Balkan Eneolithic cultures Im referring to are around the Black sea, right ?"

Balkan Eneolithic *outliers*. Corded Ware and Srubnaya branched out from the same earlier R1a M417 nation. Considering Corded Ware and Srubnaya were only about 20% EEF this ancestor could not have lived anywhere in the Balkans.

Now lets look at the R1b P312 nation who delivered the other big Steppe migration into Late Neolithic Europe. Chances are they descend from the same much earlier R1b L23 nation that Yamnaya also descended from. R1b L23 along with no EEF ancestry in Yamnaya confirms the R1b P312 nation's EHG/CHG is from the Steppe.

The R1b P312 and R1a M417 nations covered everything in Europe north of Italy, south of Lappland, and west of Russia by 2300 BC. Eventually R1b P312 reached Spain and Italy meaning those places also got their Steppe straight from the Steppe. The only exceptions could be in Southeast Europe. That's only place where 'Steppe' ancestry could derive from migrations which predate 3000 BC.

Aram said...

Gioiello

Theoretically J hg can be present in Paleolithic Europe especially in the South Europe but I think they will be dead end.
As for Your Y dna. I said my opinion. It has European origin. Don't worry for that. The Armenian who is close to You is migrant from Europe.

Anthro Survey

I am currently rereading Kilinc paper about Tepecik. They are speaking about close affinity of Otzi Iceman and Tepecik. This is an interesting story line about South Anatolia and South Europe. I will share my opinion when I finish it.

Gioiello said...

@ Samuel Andrews
"I get triggered mainly because you've proven to be a jerk. Notice I don't yell at Gioiello very much even though his ideas are even more out there. It's because Gioiello hasn't ever put his nose up to other posters".
I did hypothesis, based upon a deep study of the haplotypes long before that we had SNPs as to now, foreseeing experiments which confirmed or disproved that. All in line with the scientific method at least from Galileo Galilei.
Even about my hg. R1b1a2-L23-Z2110-FGC24408 I offered to pay YFull for the test of an Arab linked to me with only that old SNPs and about the Armenian Melik I asked a deep test forundertsanding when our haplotyoes separated, because from that anser we would be able in answering where my hg was orinated from.
"Now lets look at the R1b P312 nation who delivered the other big Steppe migration into Late Neolithic Europe. Chances are they descend from the same much earlier R1b L23 nation that Yamnaya also descended from. R1b L23 along with no EEF ancestry in Yamnaya confirms the R1b P312 nation's EHG/CHG is from the Steppe.

The R1b P312 and R1a M417 nations covered everything in Europe north of Italy, south of Lappland, and west of Russia by 2300 BC. Eventually R1b P312 reached Spain and Italy meaning those places also got their Steppe straight from the Steppe. The only exceptions could be in Southeast Europe. That's only place where 'Steppe' ancestry could derive from migrations which predate 3000 BC".

All what you say depends from answereing my question above, because my haplotype is 6000 years old as to YFull (even 7000 or 8000 as to me).

@ Aram
"Gioiello
Theoretically J hg can be present in Paleolithic Europe especially in the South Europe but I think they will be dead end.
As for Your Y dna. I said my opinion. It has European origin. Don't worry for that. The Armenian who is close to You is migrant from Europe".

If that will be demonstrated, being my subclade Z2110 with R-L51-PF7589 the oldest survived so far, this would be the definitive proof that my theory is right, the only right.

Anthro Survey said...

@Samuel Andrews

"The extra layer of Iran Neo in Sycthians is interesting. One could argue that's from their Iranic ancestors. I just don't think so though, it looks it has an independent origin we're not familiar with."

The extra layer most likely stems from Transoxiana or Bactria, whereby steppe nomads would have encountered relatively densely populated urban cultures populated by Iran_Neo-like people. People tend to overlook this for some reason despite Sarianidi's fruitful lifelong archaeological work there. It doesn't come from the Iranian plateau---I can almost guarantee that. You know, it's quite ironic when you think about it: the political entity we call Iran played no role in the initial Iranic ethnogenesis. It mirrors how this area, which many people call "Persia", played only a secondary role in the development of New Persian language, a product of literary revival in areas like Balkh and Marv during the early Islamic period.

"Speaking of Tocherian, the guy on the left was Tocherian. He looks like lots of people I see at Wrigley Field. He isn't a coincidence, I think he strongly indicates Tocherian is also from the Steppe."

The guy could just as well be Sogdian. We can't be sure. Needless to say, most Sogdians would have been brunette and/or sported West Asian features(like, say, Ahmad Shah Massoud) but the those minority types you'll encounter among Tajiks, Pamiris, Yaghnobis, Pashtuns, Dardics, Nuristanis, etc. scream steppe. No doubt about that.

Samuel Andrews said...

@Anthro Survey,
"The extra layer most likely stems from Transoxiana or Bactria, whereby steppe nomads would have encountered relatively densely populated urban cultures populated by Iran_Neo-like people."

That would make a lot of sense.

"The guy could just as well be Sogdian. We can't be sure. Needless to say, most Sogdians would have been brunette and/or sported West Asian features(like, say, Ahmad Shah Massoud) but the those minority types you'll encounter among Tajiks, Pamiris, Yaghnobis, Pashtuns, Dardics, Nuristanis, etc. scream steppe. No doubt about that."

I do think it is possible Tocherians were an IE BA Steppe groups that found a way to remain genetically isolated. They could have even been R1a Z93 but not Indo Iranian. On other hand Sycthian DNA, the place they lived, the genetic makeup of people there today, makes me think there's no way they didn't have a lot of Iran Neo and East Asian stuff.

Wikipedia has two more similar paintings of Tocherian redheads.

https://upload.wikimedia.org/wikipedia/commons/6/62/Donor_figures_from_Kizil_Caves.PNG

https://s-media-cache-ak0.pinimg.com/originals/8b/47/bd/8b47bdb09701be3fa2884417b9c45cba.jpg

One could argue the hair color in those paintings isn't supposed to reflect reality. Modern Uyghur's non-Eastern Asian ancestry has hardly any EHG, it's loaded with CHG/Iran Neo. Then again the same is true for Turkic speakers in areas where Sycthians used to live.

Anonymous said...

Mówiono tu o wieku R1b-BY611 że bardzo młody.
Nie do końca to prawda. Proszę spojrzeć na raport
http://www.projectory.de/genetics/str/2017-05-21/y-str-report.pdf
Ten raport został przychylnie przejęty przez ważne osoby.
Co ciekawe różnice dotyczą też grupa BY593.
Wydaje się że wiek BY611 jak i BY593 wysterowano ręczne by wiązało się to z „inwazją Słowian”.
Co do Germańskiej lina BY250 jest znaczne lepiej .

Rob said...

@ Sam

""You do realise the Balkan Eneolithic cultures Im referring to are around the Black sea, right ?"

Balkan Eneolithic *outliers*. Corded Ware and Srubnaya branched out from the same earlier R1a M417 nation. Considering Corded Ware and Srubnaya were only about 20% EEF this ancestor could not have lived anywhere in the Balkans"

See, you dont have a clue what you're talking about, you can't differentiate different phenomena and different regions.
I have never stated Corded Ware comes from Balkans. I struggle to discern if you're intentionally strawmanning/ trolling, or just need help.

"Now lets look at the R1b P312 nation who delivered the other big Steppe migration into Late Neolithic Europe. Chances are they descend from the same much earlier R1b L23 nation that Yamnaya also descended from. R1b L23 along with no EEF ancestry in Yamnaya confirms the R1b P312 nation's EHG/CHG is from the Steppe. "

Have i stated anything to contrary. ? Is this thread about Bell Beaker all of a sudden ?
We were talking about the source and nature of steppe admixture in Greece and the Balkans, not England.

Do you know how much R1a and R1b we have found so far in Greece and the Balkans ? 1 out of 24.

"The R1b P312 and R1a M417 nations covered everything in Europe north of Italy, south of Lappland, and west of Russia by 2300 BC."

Incorrect. They migrated throughout Europe and settled certain regions, but did not "cover" it.. We have late TRB and GAC which existed at the same time CWC/ BB, and they're non -R1
We have Pitted Ware culture in Scandinavia which was I.
The Late Neolithic cultures in France which existed until 2200BC were probably I & G.
We have Unetice culture which is I2.
We have later Bronze Age and Iron Age groups from Poland which are all I2, I1 and G.
Where did they all come from ? Idaho ?

The expansive range of modern R1a in eastern Europe and Balkans is from a very recent expansion of Slavs. There's so much you don't know, and you never will because you've got a bad attitude and poor aptitude.

" I'm not the mentally disturbed weirdo you think I am"
Demonstrate it.

Anthro Survey said...

@Sam

Yeah, the geography of the Tarim and greater Xinjiang region could have allowed the Tocharians to maintain that kind of isolation.

As for CHG in Turkics--whilst it holds for Uighurs, as you say, it's not the case across the board. Steppe ancestry takes up a huge chunk of Kazakh DNA. My guess is that Uighurs have a mainly sedentary ancestral subtrate, while Kazakhs are largely descended from the nomadic Scythians who were still primarily steppic in addition to some extra CHG.

It's also a pretty good bet that Seljuks and later Oghuz migrants to Anatolia triggered by Mongol invasions were steppe-rich. Central Asian ancestry in Turks has been clocked at ~20%. Dienekes' methods estimate 10% of it to be ENA. The remaining 7-10% is probably steppic and accounts for the diff in steppe ancestry b/ween them and groups like Armenians, Pontics, Laz, and Capaddocian Greeks.

Gioiello said...

Mówiono tu o wieku R1b-BY611 że bardzo młody.
Nie do końca to prawda. Proszę spojrzeć na raport
http://www.projectory.de/genetics/str/2017-05-21/y-str-report.pdf
Ten raport został przychylnie przejęty przez ważne osoby.
Co ciekawe różnice dotyczą też grupa BY593.
Wydaje się że wiek BY611 jak i BY593 wysterowano ręczne by wiązało się to z „inwazją Słowian”.
Co do Germańskiej lina BY250 jest znaczne lepiej .

BY611 formed 4800 years ago.
The oldest sample survived is ERS256992 ITA from Sardinia separated 3200 years ago.
Two are the subclades;
R-Y30192, separated 3200 years ago from the sardinian sample and has a survived sample in YF08067 ITA from Avellino province. A more recent subclade separated 1050 years ago is present in Iberia (YF09270 and YF01929.
The other subclade separated 3200 years agp has a survived sample in YF006082 from USA and the subclade R-Z2705 whose ancestor lived 2600 ya but the survoved samples are less than 900 years ago, It is the "Balkan cluster" I discovered many years ago.
Ask Samuel Andrews but also Ted and any other where the origin very likely was.
Also about this argument I wrote infinite letters, about Rizzo from SMGF, about Bisaccia who is very likely the sample from Avellino, not belonging to the "Balkan cluster" but ancestor of it etc etc ...

Samuel Andrews said...

@Rob,

Throughout this summer you've been arguing Balkan Chalcolithic *outliers* are the source of the Steppe in LNBA Europe. Your 'new' theory that the Myceanean's Steppe is from the Balkan Chalcolithic *outliers* is from the same retarded ideas. That's why I pointed out that the two biggest Steppe migrations into Europe(R1a-M417, R1b-P312) most definitely originated in the Steppe not the Balkans.

I guess you've decided to retreat from your previous stance which argued all Steppe ancestry in Europe is from the Balkan Chalcolithic and are now arguing just arguing some Steppe ancestry is from the Balkan Chalcolithic.

"Incorrect. They migrated throughout Europe and settled certain regions, but did not "cover" it.. "

Ok, mr. perfectionists they didn't cover everything but nonetheless they took over a huge chunk of Europe within a couple hundred years. That you can't deny. And you also can't deny they are the main source of Steppe ancestry in modern Europe(all of R1b P312 rich western Europe, all of R1a M417 rich Eastern Europe). Thus, the premise of your retarded theory about Balkan Chalcolithic outliers being the source of Steppe ancestry in Europe is DEAD WRONG.

"The Late Neolithic cultures in France which existed until 2200BC were probably I & G.
We have Unetice culture which is I2.
We have later Bronze Age and Iron Age groups from Poland which are all I2, I1 and G.
Where did they all come from ? Idaho ?"

Just because their Y DNA isn't R1 doesn't mean they likely come from the Balkan Chalcolithic. This is so typical of you. When the evidence doesn't absolutely prove a premise of the pro-Steppe crowd you parade every alternative theory you can think of and act as if the data suggests it's correct.

The I1-rich population in Iron age Poland were Goths. They were recent arrivals from Scandinavia. That's why they had so much I1. Btw, there was a recent Polish study which found most Wielbark people had I1 not I2 so I'm right about them being mostly I1.

They aren't representative of Bronze age Polish which could have been R1a rich. We already know the Baltic BA was R1a and mostly Corded Ware. Also, the only BA genome from East Germany belongs to R1a Z282.

The famous Iron age peoples of Eastern Europe; Germans, Sycthians, Huns, etc. were recent immigrants and don't represent the native peoples. Corded Ware's R1a didn't only remain in a tiny isolated proto-Slavic community for 3,000 years and then expanded out of no where.

Unknown said...

@David
Look here.

"http://www.nature.com/ejhg/journal/v23/n1/suppinfo/ejhg201450s1.html?url=/ejhg/journal/v23/n1/abs/ejhg201450a.html&foxtrotcallback=true"

Having been a coauthor of the paper, we found 0.6% Z93 in mainland Greece and 1% Z93 in Crete. A bit underwhelming and less Z93 in the mainland. My guess is that these are relic Persians/Anatolians. A better bet would be R1b-Z2103 for an I-E/Greek connection.

Davidski said...

@Roy

It's just modern Y-DNA. Hardly relevant because of, for instance, the LBA collapse, a couple of dark ages, you name it.

Ancient DNA has already pinpointed a Sintashta-like sample with Z93 in the Bronze Age southern Balkans, and I bet that he'll be a good fit for the steppe ancestry in Mycenaeans. That's all that counts.

Gioiello said...

@ Roy King

"@David
Look here.

"http://www.nature.com/ejhg/journal/v23/n1/suppinfo/ejhg201450s1.html?url=/ejhg/journal/v23/n1/abs/ejhg201450a.html&foxtrotcallback=true"

Having been a coauthor of the paper, we found 0.6% Z93 in mainland Greece and 1% Z93 in Crete. A bit underwhelming and less Z93 in the mainland. My guess is that these are relic Persians/Anatolians. A better bet would be R1b-Z2103 for an I-E/Greek connection".

But where are all these R-L23 in Greece? They are above all R-L23-CTS9219 "Balkan cluster" (my discovery) which isn't older than 900 years in the Balkans and the oldest subclades are all in Italy or Western Europe.
Also the unique R-L389 (Diamantikos) has YCAII=18-22, which is 1 of the 4 haplotypes present in Italy etc etc.

«Oldest ‹Older   201 – 400 of 436   Newer› Newest»