PDA

View Full Version : Analysis of Chalcolithic El Portalon samples (Günther at al. 2015)



Maciamo
09-09-15, 13:54
This analysis refers to the Günther at al. (2015) paper (http://www.pnas.org/content/early/2015/09/02/1509851112.full.pdf) already discussed here (http://www.eupedia.com/forum/threads/31551-Ancient-Genomes-suggest-Basque-are-descended-of-Late-Neolithic-Iberians), which tested the genomes of seven pre-Beaker Copper Age and one Middle Bronze Age individuals from El Portalon Cave, near Burgos in northeast Castile, Spain.

One remarkable finding was that the so-called ATP3 individual, who lived circa 3516–3362 BCE, belonged to Y-haplogroup R1b1a2-M269, the main lineage of the contemporaneous Yamna culture (3600-2000 BCE) at the other extremity of Europe.

ATP3 belongs to mtDNA lineage of potential Indo-European origin: K1a2b, which is found mostly in northern Europe and Iran nowadays. ATP20 is U5a1c, a lineage now found mostly around the Baltic (Germany, Poland, Lithuania). All the other maternal lineages are typical of Iberia.

Unfortunately, the paper didn't report the admixtures for ATP3 and ATP20. But I just found that Genetiker ran the admixtures for all samples (https://genetiker.wordpress.com/2015/09/08/analyses-of-copper-and-bronze-age-spanish-genomes/) and that is extremely interesting.

Most samples are predominantly Southern European (ATP17=66.7%, Matojo=64.88%, ATP7=63.4%, ATP16=62.6%, ATP9=51.8%, ATP2=46.8%), except ATP3 and ATP20 who who have considerably less (ATP3=32.47% and ATP20=33.03%) as if they had been recently hybridized and only had half of the Iberian ancestry of other samples.

Both ATP3 and ATP20 have a higher percentage of Northern European and European Hunter-Gatherer ancestry, with an combined average of 20.5% and 23% respectively. Only ATP9 has more (32%).

More fascinating is the fact that ATP4 and ATP20 are the only samples with Veddoid ancestry (i.e. Indian subcontinent, but possibly of IE origin), with a remarkable 28.14% for ATP20 and a much more moderate 3.82% fro ATP3.

Actually ATP20 (2289–2050 BCE) doesn't look very European, with 7.96% of Negroid and 7.76% Northern Amerindian ancestry in addition to its 28% of Veddoid.

ATP3 (3516–3362 BCE) stands out from other samples thanks to its high Northern Middle Eastern ancestry (31.97%) against 0% for ATP20, 11% for ATP17 and between 0% and 8% for other samples. What Genetiker calls Northern Middle Eastern is what we typically referred on this forum as Caucaso-Gedrosian admixture - the same as in the "Armenian-like admixture" in Yamna samples.

With its 32% of Caucaso-Gedrosian, 14% of Northern European ancestry, 6% of European Hunter-Gatherer and 3.8% of Veddoid, it does indeed look as if ATP3 has a bit over half of Steppe ancestry, but with a higher proportion of northern Middle Eastern and Veddoid than Yamna samples. In other words it could be descended to the pre-Indo-European Anatolian R1b-M269, the group of cattle herders that would cross the Caucasus and settle in the Pontic-Caspian Steppe. So could it be an offshoot of cattle herders that directly migrated from Anatolia to Iberia during the Neolithic period. But if so, how did his lineage not get more admixed along the way ? Neolithic farmers all over Europe were overwhelmingly (and often exclusively) Southern-European in admixture.

What baffles me the most is that this individual's ancestors managed to maintain a relatively pure West and South Asian admixture while crossing all Europe at the height of the Neolithic. Could it represent a migration of copper metallurgists from Anatolia to Iberia ? R1b tribes from the Pontic Steppe are thought to have started invading the Balkans several centuries before the Yamna period, from c. 4200 BCE. Copper metallurgy was already well implanted in the Balkans by then, but may have been brought from central Anatolia. An alternative scenario is that R1b-M269 invaders took wives with high Middle Eastern admixture in copper towns in the Balkans, and that their descendants spread metallurgy fairly quickly all the way to Iberia. If that is the case, the Southern-European component might be Balkanic rather than Iberian.

The main problem with the diffusion of copper metallurgists is that copper working is not documented in northern Spain until 3000 to 2500 BCE and this sample dates from c. 3500 BCE (see map of Copper Age diffusion (http://www.eupedia.com/europe/map_diffusion_chalcolithic.shtml)).


In conclusion, it seems pretty clear from the autosomal analysis that ATP3 is ethnically different from other samples from Chalcolithic El Portalon, and is not a native inhabitant of Neolithic Iberia, but a recent Copper Age immigrant, although how he got there remains unclear. A small group (or extended family) of Steppe people with (ox-drawn) carts travelled all the way there, probably after having intermingled with Copper Age immigrants from Anatolia in the Balkans. The maternal north-east European lineage indicates that females also took part in the migration. Even though such events must have been relatively rare and were probably not representative of a large-scale PIE migration.

This could explain how early subclades of R1b-M269 (S116* and DF27) ended up in southern France and Iberia, bypassing the rest of Europe. The bulk of R1b people would only have arrived from the Balkans to Central Europe c. 2800-2500 BCE, and a second wave of R1b-S116 (probably L21) moved into Western Europe from 2500 to 2000 BCE. A third wave (U152) expanded from Central Europe during the Hallstatt and La Tène periods.

A small predominantly male population of Steppe migrants to Iberia in the Copper Age that grew progressively through a founder effect in the paternal line would explain why PIE languages didn't take hold in the peninsula until later Bronze Age migrations (Celts).

An early incursion of Steppe/Balkanic R1b-M269 people to Iberia would also elucidate the sudden appearance of copper technology in the peninsula.

http://cdn.phys.org/newman/csz/news/800/2015/2-ancientgenom.jpg
Image source (http://phys.org/news/2015-09-ancient-genomes-link-early-farmers.html): "Illustration of every day life in the El Portalon cave during the Neolithic and Copper Age"

Maciamo
11-09-15, 12:53
Another surprising thing in Genetiker's runs of K=17 admixtures (https://genetiker.wordpress.com/2015/09/08/analyses-of-copper-and-bronze-age-spanish-genomes/) is that Copper and Bronze Age samples from El Portalon cave all have between 2 and 5% of Bushman-and-Pygmy admixture, except the unusual ATP20 which has 0% but 28% of Veddoid, 8% of Negroid and 7.8% of Amerindian) and ATP2 which has a stunning 28% of sub-Saharan African (14% of Bushman-and-Pygmy + 12.47% Western-Negroid + 1.42% Eastern-Negroid) ! The modern Basques are almost 100% European in admixtures, so how do their presumed ancestors show up with such high non-European admixture ?

The authors of the paper were a bit hasty in their conclusion that these Chalcolithic samples are similar or even ancestral to modern Basques. Here is what they wrote in the paper's conclusion :


In summary, our ancient genomic sequence data from the El Portalón individuals and our analyses suggest the following model of events: The incoming early farmers, who could have spoken a non–Indo-European language, assimilated resident hunter–gatherers, with this admixed group becoming the ancestors of modern-day Iberian groups. Basques remained relatively isolated (compared with other Iberian groups) with marked continuity since the Neolithic/Chalcolithic period, but not since the Mesolithic.

Using the same K=17 admixtures, Genetiker gets completely different results for modern Basques. The Basques have approximately 65% of Southern-European, 30% of Northern-European and 3% of European-Hunter-Gatherer and 2% of assorted trace admixtures, but mostly Northern-Middle-Eastern, Veddoid and strangely also Amerindian (well after all the Basques have 0.5% of Y-haplogroup Q1a).

There is hardly any continuity between the Basques and Mesolithic Western Europe considering that La Braña had 100% of European-Hunter-Gatherer admixture at K=17, and modern Basques have only 3% of the same admixture. This isn't clear at all from the Günther at al. paper, which conveniently blends La Braña-like admxiture with Yamna-like admixture, when other recent papers have clearly distinguished between Western and Eastern European Hunter-Gatherers' admixtures. Autosomally, modern Basques appear to be about 2/3 descended from Neolithic farmers and 1/3 from Northeast Europe (Yamna).

What really sets the Basques apart from other Iberians is their lack of Middle Eastern admixture, which ironically was present in several Copper Age samples (ATP3=35%, ATP16=12.5%, ATP17=11%, ATP7=10%, ATP2=5.5%, ATP9=5%). Central and southern Spaniards today have between 15% and 25% of Middle Eastern admixture, which obviously came from the Arabs, Jews and Phoenicians.

Then there is the other contradictory fact that Chalcolithic samples from El Portalón belonged to Y-DNA haplogroups that are virtually absent from the modern Basque population, namely I2a2a-M223 and H2. Even the R1b sample may not be ancestral as the Basque R1b-M153 has a coalescence age of less than 1500 years.

bicicleur
11-09-15, 13:52
R1b-V88 was in Els Trocks >5000 BC
Els Trocks was a shepherd shelter
could he have made all the way around through North-Africa and brought subsharan admixture to the area?

Maciamo
11-09-15, 14:01
R1b-V88 was in Els Trocks >5000 BC
Els Trocks was a shepherd shelter
could he have made all the way around through North-Africa and brought subsharan admixture to the area?

I am not surprised to find African admixture in northern Iberia during the Neolithic or Chalcolithic. Besides R1b-V88, A1a and E1b1b were also present then. My objection was that this population could be called similar to the modern Basques. It's probably not that surprising considering that El Portalon Cave is not even in the greater historical Basque country.

Tomenable
11-09-15, 14:21
ATP2 which has a stunning 28% of sub-Saharan African (14% of Bushman-and-Pygmy + 12.47% Western-Negroid + 1.42% Eastern-Negroid)

That ancestry could be from early North-West Africans, descended from men such as those 160,000 years old found at Jebel Irhoud:

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

Reconstruction of one of 7 men found at Jebel Irhoud (in terms of Y-DNA they could probably belong to A1b and A1a haplogroups):

https://www.pinterest.com/pin/516928863456817198/

http://www.kenniskennis.com/images/site/388.jpg

https://s-media-cache-ak0.pinimg.com/736x/e9/6c/e5/e96ce50ddba1a128c821a2992cd830e1.jpg

http://s27.postimg.org/bmkil791v/388.jpg

Check also:

http://www.ancestraljourneys.org/earlymediterraneans.shtml

http://articles.philly.com/2011-06-16/news/29665786_1_modern-humans-neanderthals-skeleton

bicicleur
11-09-15, 15:53
That ancestry could be from early North-West Africans, descended from men such as those 160,000 years old found at Jebel Irhoud:

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



I doubt that.
Aterian may descend from Jebel Irhoud, which lasted till 40 ka.
Then it was replaced by IberoMaurusian,
later by Capsian
and finally by Berbers (TMRCA just 2100 years !) and Islamic Arabs.
I don't think any descendants of Jebel Irhoud survived.

Tomenable
11-09-15, 20:21
I doubt that.
Aterian may descend from Jebel Irhoud, which lasted till 40 ka.
Then it was replaced by IberoMaurusian,
later by Capsian
and finally by Berbers
Weren't those Aterians just absorbed / assimilated? Otherwise how to explain this:


Another surprising thing in Genetiker's runs of K=17 admixtures (https://genetiker.wordpress.com/2015/09/08/analyses-of-copper-and-bronze-age-spanish-genomes/) is that Copper and Bronze Age samples from El Portalon cave all have between 2 and 5% of Bushman-and-Pygmy admixture, except the unusual ATP20

Was there some additional Bushman-and-Pygmy like migration to the north later on?

I wouldn't be surprised so much by some Western-Negroid, but Bushman-and-Pygmy?

bicicleur
11-09-15, 21:14
Weren't those Aterians just absorbed / assimilated? Otherwise how to explain this:



Was there some additional Bushman-and-Pygmy like migration to the north later on?

I wouldn't be surprised so much by some Western-Negroid, but Bushman-and-Pygmy?

http://dienekes.blogspot.be/2013/07/origin-of-iberomaurusian.html

We present 54 AMS radiocarbon dates on bone and charcoals from a sequence of late MSA and LSA occupation levels of the cave. Using Bayesian modelling we show that an MSA non-Levallois flake industry was present until ∼24.5 ka Cal BP (calibrated years before present), followed by a gap in occupation and the subsequent appearance of an LSA Iberomaurusian industry from at least 21,160 Cal BP.

Furthermore : is there a link between 160 ka Irhoud and LGM subsaharan or Bushman-and-Pygmies?

Fire Haired14
11-09-15, 23:38
@Maciamo,

I think you should say "component" not "ancestry". A score in a North European component for example is not the amount of North European ancestry a person has. Components are relative. It is named after the region it peaks in(usually at 50% at most). It can be confusing for people who are not familiar with ADMIXTURE tests.

Also, ATP3 is of very low coverage. Any exotic(African, East Asian, South Asian) score is almost certainly noise. The authors would have picked up on African ancestry. Formal statistics will confirm or disprove African ancestry. No Pre-Historic European genomes show any signs of African ancestry(including all Neolithic ones, even the Spainard with R1b-V88). Today African ancestry is only found in Iberia(x Basque) and Southern Italy and in both regions it is very insignificant(like a few percent).

Because of ATP3s low coverage we should be hesistent of estimating any-type of ancestry(Steppe vs Neolithic European) with a K17 ADMIXTURE test. Davidski is currently doing analysis of these new Copper age Spanish genomes, and in a PCA ATP3 clusters by Basque and Northern Spanish. It suggests Steppe ancestry but very very very little. We can only confirm sometype of ANE ancestry in the high coverage El Portalon genomes.

Fire Haired14
11-09-15, 23:43
That ancestry could be from early North-West Africans, descended from men such as those 160,000 years old found at Jebel Irhoud:


IMO, 160,000 years is way way way too far back. Our oldest human genome is 45,000 years old and is equally related to all Eurasians. The earliest split between Eurasians we know of, is Basal vs Crown maybe 50,000-60,000. 160,000 years ago is like at the begging of humanity. Any genome from that time period would probably be equally related to all humans living today.

Greying Wanderer
12-09-15, 00:32
Just an idea but if there was an OoA expansion followed by a back-migration from SE Asia then maybe mountain refuges along the Atlantic coast had/have larger surviving amounts of that OoA dna and these results from Iberia picking up lots of different kinds of African dna is actually picking up a signal from this OoA population that the software has no reference for (edit: so it jumps to the closest).

Maciamo
12-09-15, 14:39
@Maciamo,

I think you should say "component" not "ancestry". A score in a North European component for example is not the amount of North European ancestry a person has. Components are relative. It is named after the region it peaks in(usually at 50% at most). It can be confusing for people who are not familiar with ADMIXTURE tests.

Also, ATP3 is of very low coverage. Any exotic(African, East Asian, South Asian) score is almost certainly noise. The authors would have picked up on African ancestry. Formal statistics will confirm or disprove African ancestry. No Pre-Historic European genomes show any signs of African ancestry(including all Neolithic ones, even the Spainard with R1b-V88). Today African ancestry is only found in Iberia(x Basque) and Southern Italy and in both regions it is very insignificant(like a few percent).

Because of ATP3s low coverage we should be hesistent of estimating any-type of ancestry(Steppe vs Neolithic European) with a K17 ADMIXTURE test. Davidski is currently doing analysis of these new Copper age Spanish genomes, and in a PCA ATP3 clusters by Basque and Northern Spanish. It suggests Steppe ancestry but very very very little. We can only confirm sometype of ANE ancestry in the high coverage El Portalon genomes.

You may be right about the noise. I didn't check the samples myself and don't know how good their coverage is.

Maciamo
13-09-15, 08:54
I have updated my analysis of the ATP3.

Tomenable
24-09-15, 02:30
I have updated my analysis of the ATP3.

The discovery of ATP3 proves, that R1b-M269 was present both in Iberia and in the Pontic-Caspian steppe at the same time.

I wonder if ATP3 was positive for L51 clade, now prevalent in Western Europe? All Yamnaya samples were negative for L51:

http://s24.postimg.org/be36zsq5h/West_East_Split.png

http://s24.postimg.org/be36zsq5h/West_East_Split.png

It seems that L23 split very early on (before 3300 BC) and one branch (L51) went to Iberia, another branch to the steppe.

The idea of the "massive migration from the steppe" of L51 may be wrong. It could be spreading from Iberian Bell Beaker.

Of course ATP3 itself predates Bell Beaker, but - as we know - the oldest Bell Beaker sites are from Iberia.

So if R1b-L51 was present in Iberia already before Beaker, it could be spreading outside of Iberia later on, with Beaker.

Athiudisc
28-09-15, 02:32
It seems that L23 split very early on (before 3300 BC) and one branch (L51) went to Iberia, another branch to the steppe

This would seem more likely if ATP3 didn't evince steppe ancestry. As is, it appears as if APT3's paternal ancestor came from the steppe, bringing M269 of some sort with him.

Sigfrido
28-09-15, 06:46
Today African ancestry is only found in Iberia(x Basque) and Southern Italy and in both regions it is very insignificant

If you talking about Dodecad, African admixture is also found in France and in the Balkans so you are wrong. Look at the eupedia maps about autosomal dna. Both Boatigue and Lazaridis found no African admixture on mainland Italy.

MOESAN
01-10-15, 13:22
The discovery of ATP3 proves, that R1b-M269 was present both in Iberia and in the Pontic-Caspian steppe at the same time.

I wonder if ATP3 was positive for L51 clade, now prevalent in Western Europe? All Yamnaya samples were negative for L51:

http://s24.postimg.org/be36zsq5h/West_East_Split.png


It seems that L23 split very early on (before 3300 BC) and one branch (L51) went to Iberia, another branch to the steppe.

The idea of the "massive migration from the steppe" of L51 may be wrong. It could be spreading from Iberian Bell Beaker.

Of course ATP3 itself predates Bell Beaker, but - as we know - the oldest Bell Beaker sites are from Iberia.

So if R1b-L51 was present in Iberia already before Beaker, it could be spreading outside of Iberia later on, with Beaker.

Not a contradiction but an other way to look at things:
1- Y-R1b-L23>>L51 could have taken a southern road and could be either 'basquic' or I-Ean but I don't believe in a southern road for all L-51 look at North-Eastern Europe - that said it doesn't exclude some little groups of only males, porspectors, in South-West
2- if I-E, what excludes a cut off North the Caspian between groups heading Northwest and West and groups heading East and North-East, in the Steppes hypothesis???
&: the L23 of South caucasus seem to me more a genetic dead-end than a cradle of demic explosion for the R1b ligneages... And R1b is surrounded by other (dominant) Y-DNA ligneages there. If Caucasus surroundings were overcrowded for civilisation advance, what prevented to other ligneages than Y-R1b or Y-R1a to "invade" North Pontic steppes? Only Y-G2a seems having had a (light) weight Just trying to understand upon the facts I have at hand.
&&&: concerning BBs we spoke about that long time ago: I think they put other people in move rather than they had a great demic proper posterity

ToBeOrNotToBe
11-10-18, 13:14
What happened about this?

markod
11-10-18, 20:38
What happened about this?

I think we can be reasonably sure that ATP3 had M269. See Genetiker's post here:

https://anthrogenica.com/showthread.php?9662-The-Neolithic-Transition-in-the-Baltic-Was-Not-Driven-by-Admixture-with-Early-Europea/page48

ToBeOrNotToBe
11-10-18, 20:42
I think we can be reasonably sure that ATP3 had M269. See Genetiker's post here:

https://anthrogenica.com/showthread.php?9662-The-Neolithic-Transition-in-the-Baltic-Was-Not-Driven-by-Admixture-with-Early-Europea/page48


Yamnaya trembles...

Its going to need a confirmed L51 to topple it, but still, Yamnaya aint gonna give us that any time soon.

bicicleur
11-10-18, 23:59
Yamnaya trembles...
It�s going to need a confirmed L51 to topple it, but still, Yamnaya ain�t gonna give us that any time soon.
atp3 is dated 5.4 ka
check Olalde 2017
it's a dead end, even if it is M269

ToBeOrNotToBe
12-10-18, 01:21
atp3 is dated 5.4 ka
check Olalde 2017
it's a dead end, even if it is M269

What's wrong with that, it's pretty much Copper Age, and I think that was spread to Iberia by L51 immigrants. Why is it a dead end?

Olympus Mons
12-10-18, 15:08
What's wrong with that, it's pretty much Copper Age, and I think that was spread to Iberia by L51 immigrants. Why is it a dead end?

Yes. Will love to see what Bicicleur says...

Prior to 3500 bc Iberia was scarcely populated. It was the arriving pops from 3700bc on that are the incognito.