The Arrival of Steppe & Iranian Related Ancestry in Islands of West Mediterranean

i don't find any info on the Ibiza Phoenician
where is it?

"We tried to model the Phoenician individual from 32 469 using the same 4 distal sources but no470 models produced valid results. When we added Morocco_LN as a fifth possible source, however, we471 obtained a good two-way fit for a model with 18.8 ± 7.9% Anatolia_Neolithic and 81.2 ± 7.9%472 Morocco_LN ancestry (p=0.141)"

Page 19 of the supplement.


Also, page 25.
"Ibiza_Phoenician: We investigated if the published Phoenician individual from Ibiza 64432 was645 consistent with inheriting some ancestry from previous Balearic Islands populations so we used the646 same proximal sources as for Menorca_LBA but then added: Menorca_LBA, Mycenaean, Sardinia_IA, 647 Sicily_MBA4109, Morocco_LN, and Jordan_EBA. Only models with two sources of admixture648 produced valid results, and all of them required Morocco_LN as one of those sources649 (Supplementary Table 11). Even though we used model competition to try to reduce the number650 of working models by adding the unused sources to the “Right” (including the Bronze Age Balearic651 individuals), none of the initially working models failed."
 
Wow, Nelly, did anyone notice this yet????

"As with the Balearic individuals we could not model modern Sicilians using a threshold of558 p>0.05, or even with a more permissive p>0.01 threshold. There is clearly also a North African559 influence, however, as we identify a working model for the 4-way model with 24.8 ± 4.3%560 Anatolia_Neolithic, 12.1 ± 3.1% Iran_Ganj_Dareh_Neolithic, 19.8 ± 1.4% Yamnaya_Samara, and 43.3561 ± 6.1% Morocco_LN (p=0.334)."

I knew there would be North African, but THIS MUCH?

I don't think the y Dna would support that, would it? I was expecting perhaps 15%.

Could qpAdm be a little wonky or is this legit?

Why not use a North African sample closer to the present? Or use a Levantine sample for the Phoenicians, or maybe the "Phoenician" woman from Ibiza?

Plus, this is all done without any Greek migration era samples. Would everything have to be redone based on those samples???

that is what figure 4 b says

do you have data on paternal DNA in Sicily?
in the Balears, Morocco LN would haven arrived with the Phoenicians, and in Sardegna it is absent
 
That is an insane number, I doubt it's correct.

I don't know. It's in the graphs in the body of the paper, and then that one sentence in the Supplement. They sure didn't highlight it.

I mean, I do know the Normans and even Frederick II didn't show the same zeal as Ferdinand and Isabella in hunting them out root and branch and setting the more recalcitrant ones ablaze. The first auto da fe in Sicily occurred after the Spanish took over, but a lot of them were expelled.
 
Anyone know what Moroccan Late Neolithic looked like?

Is it likely North Africans, say, Tunisians, would still be like that in 800 C.E.? That was the staging era for the invasion of Sicily.

I wonder about the implications for southern mainland Italy. The Saracens were barely there.

@Bicicleur,
I think the latest Grugni et al paper would be the best maybe, but I can't get access. Maybe someone else can get the data.

This is from Boattini. The "E" contains E-V13, plus would we attribute all the rest of the E to North Africa? The J2a I assumed was mostly from Anatolia/Crete directly or by way of Greece, although I suppose some could have been from North Africa as well.

Most of Sicily is in Number 7.

mlmEoxz.png
[/IMG]
 
Isn't that what always happens in models with temporally disparate sources - a bias towards the younger samples if they have related ancestries? LN Morocco has Spanish Neocolithic admixture so of course it would be preferred over ANF/Zagros.

Sicilians should derive most of their ancestry from Chalcolithic Iberian + Aegean or Balkan BA, the latter having steppe admixture.
 
Last edited:
Isn't that what always happens in models with temporally disparate sources - a bias towards the younger samples if they have related ancestries? LN Morocco has Spanish Chalcolithic admixture so of course it would be preferred over ANF/Zagros.

Sicilians should derive most of their ancestry from Chalcolithic Iberian + Aegean or Balkan BA, the latter having steppe admixture.

Tomorrow I'll try to find a good model somewhere and see how much of that Moroccan LN sample is Chalcolithic Iberian. It's a bit confusing I must say to use it and not delineate the components for the readers. Or perhaps it's in the paper and I just missed it. There must be a sample elsewhere in another paper as well.

I take back my suggestion that perhaps they should have used the Ibiza Phoenician just to see what it would show. It's apparently not a very good sample. Now that I know that, the strange place it lands on the PCA makes more sense.
 
Last edited:
Tomorrow I'll try to find a good model somewhere and see how much of that Moroccan LN sample is Chalcolithic Iberian. It's a bit confusing I must say to use it and not delineate the components for the readers. Or perhaps it's in the paper and I just missed it. There must be a sample elsewhere in another paper as well.

I take back my suggestion that perhaps they should have used the Ibiza Phoenician just to see what it would show. It's apparently not a very good sample. Now that I know that, the strange place in lands on the PCA makes more sense.

We discussed the sample in question in this thread about the Rosa Fregel paper.
https://www.eupedia.com/forum/threa...and-Europe?highlight=Rosa+Fregel+North+Africa

I don't know why I so completely forgot it.

Anyway, at least half of it is EEF (or Anatolian Neolithic like) according to the author (by way of Iberia), so the strictly speaking "North African" or IAM we're talking about would be about 21.5% ?, and IAM is highly Levant Neolithic like.

So there, perhaps, is one explanation for the Levant like ancestry in Sicilians which so exercises the interest of some of our colleagues, an ancestry which I believe they have proposed came as a stream specifically from the Levant into Sicily and perhaps Southern Italy in the period of the Empire, and perhaps also in Late Antiquity, a migration for which I confess I could find no basis in the history of either place.

According to this paper on the Mediterranean Islands, the "northern West Asian" or Caucasus ancestry, whatever you wish to call it, has plausibly been in Sicily since the Bronze Age.

This is starting to make more sense.

However, I still really want to see an analysis of the samples from the period AFTER the arrival of the Greek migrations of the first millenium BC, and using a contemporary North African "Saracen" sample to model the percentage of change. They may or may not have been very much like the IAM samples.

Well, we're getting somewhere finally, but always need more...ancient samples. :)
 
We discussed the sample in question in this thread about the Rosa Fregel paper.
https://www.eupedia.com/forum/threa...and-Europe?highlight=Rosa+Fregel+North+Africa

I don't know why I so completely forgot it.

Anyway, at least half of it is EEF (or Anatolian Neolithic like) according to the author (by way of Iberia), so the strictly speaking "North African" or IAM we're talking about would be about 21.5% ?, and IAM is highly Levant Neolithic like.

So there, perhaps, is one explanation for the Levant like ancestry in Sicilians which so exercises the interest of some of our colleagues, an ancestry which I believe they have proposed came as a stream specifically from the Levant into Sicily and perhaps Southern Italy in the period of the Empire, and perhaps also in Late Antiquity, a migration for which I confess I could find no basis in the history of either place.

According to this paper on the Mediterranean Islands, the "northern West Asian" or Caucasus ancestry, whatever you wish to call it, has plausibly been in Sicily since the Bronze Age.

This is starting to make more sense.

However, I still really want to see an analysis of the samples from the period AFTER the arrival of the Greek migrations of the first millenium BC, and using a contemporary North African "Saracen" sample to model the percentage of change. They may or may not have been very much like the IAM samples.

Well, we're getting somewhere finally, but always need more...ancient samples. :)

Given that we have early Medieval samples from Iberia that are half "North African" I wonder how hard it would be to create a "virtual" Early Medieval Saracen and play around with that in admixture. I know Dienekes used to do things like that. Gosh, I miss his input.

I'm also interested to see where the Haak/Krause group differ in their interpretations.
 
The three late antiquity samples are plotting next to Mycenaeans but closer to the Caucasus groups, and I think I saw this same sort of plotting in the Iron Age Greek from the Iberian paper. It may be possible the classical Greeks had more (but slightly more) Caucasus in them in comparison to Mycenaeans...or maybe some Mycenaeans had more of this Caucasus like ancestry than others and their gene pool hasn't gotten anything extra even during the classical period
 
I10554 Nuragic sample may had some steppe admixture 8.5 +- 3.3. but in general it seems to be absent at that time..did all the Beakers from Sardinia left for Sicily ? [emoji16]

Utilizzando Tapatalk
 
As davef noted 3 of the LA samples are nestled between Myceneans and the Central Anatolians who were likely responsible for the spread of bronze technology in West Eurasia. Perhaps in the Central Mediterranean the bearers of bronze technology encountered fewer natives.

The most interesting find to me are the LBA Sicilians from Trapani whom I would consider to be speakers of Indo-European. They have very little if any Caucasus admixture, and no steppe admixture. They cannot be Neolithic holdovers either since they carry Bronze Age TMRCA G2a-Z1903, one of them with a subclade specific to present day Scandinavia.

I'm pretty sure these are Balkaners. Ultimately from Chalcolithic Bulgaria, perhaps by way of Baden-Boleraz.
 
Brachycephal invasion of Anatolia and expansion of Iranian related ancestry to Europe seem sequential?
 
Brachycephal invasion of Anatolia and expansion of Iranian related ancestry to Europe seem sequential?

Seems like it, lots of Iranian and minor Levantine admixture arriving in Anatolia in the CA/BA.
 
Isn't it also odd that their supposedly best model for the Ibiza Phoenician is Anatolia_Neolithic + Morocco_LN? Maybe they just lack more proximate sources, and therefore a third componente is hidden in those other two? I find it unlikely that a Levantine Phoenician would lack Levant_Neolithic and Iranian_Chalcolithic ancestry.

The observation that there may be an unknown Iberia_Chalcolithic and Nuragic Sardinia connection is interesting, because some linguists have speculated that the Nuragic language might've been related to Iberian and/or Basque (not mainstream hypothesis, of course, the evidence is too scant, mostly based on toponyms). That might hark back to the EEF colonization, but by the IA the linguistic differentiation should be very profound, so what if the connection was actually much more recent? Just a curious and not totally off base speculation.

***

If the steppe ancestry arrived in the Balearic islands roughly at the same time it arrived in Iberia from Central Europe and looks most similar to the steppe-admixed Iberian BB, but their material culture is not BB, then it could again strengthen the idea that BB was a cultural phenomenon, a bit like modern pop Western culture (okay, not the best example, but you get it), which results in some similar aesthetics and material products produced by completely different populations and cultures. Some of the steppe+EEF people in Central Europe (or maybe also France?) would've picked up that "brand-new cultural trend" and made it their own, whereas others would be more conservative and isolated and maintained their original ways.
 
Isn't it also odd that their supposedly best model for the Ibiza Phoenician is Anatolia_Neolithic + Morocco_LN? Maybe they just lack more proximate sources, and therefore a third componente is hidden in those other two? I find it unlikely that a Levantine Phoenician would lack Levant_Neolithic and Iranian_Chalcolithic ancestry.

The observation that there may be an unknown Iberia_Chalcolithic and Nuragic Sardinia connection is interesting, because some linguists have speculated that the Nuragic language might've been related to Iberian and/or Basque (not mainstream hypothesis, of course, the evidence is too scant, mostly based on toponyms). That might hark back to the EEF colonization, but by the IA the linguistic differentiation should be very profound, so what if the connection was actually much more recent? Just a curious and not totally off base speculation.

***

If the steppe ancestry arrived in the Balearic islands roughly at the same time it arrived in Iberia from Central Europe and looks most similar to the steppe-admixed Iberian BB, but their material culture is not BB, then it could again strengthen the idea that BB was a cultural phenomenon, a bit like modern pop Western culture (okay, not the best example, but you get it), which results in some similar aesthetics and material products produced by completely different populations and cultures. Some of the steppe+EEF people in Central Europe (or maybe also France?) would've picked up that "brand-new cultural trend" and made it their own, whereas others would be more conservative and isolated and maintained their original ways.

Plus, their Ibiza Phoenician plots right on top of Mycenaeans. Did they have a lot of Morocco LN too?

I'm really not understanding this. Going by the PCA plot the sample is Greek, isn't it?

fEefemp.png
[/IMG]
 
Look at the PCA (especially of the other paper), the Morocco_LN shift does not make too much sense for the shift to present-day Sardinia. It's also really low quality individuals (<250k SNPs covered), so don't take fits with it too seriously - in general the qpAdm in this paper is very wonky. If you look at results, every SE > about 5% should set off an alarm that the data is weak.
 
Plus, their Ibiza Phoenician plots right on top of Mycenaeans. Did they have a lot of Morocco LN too?

I'm really not understanding this. Going by the PCA plot the sample is Greek, isn't it?

fEefemp.png
[/IMG]

So are the Sardinian Late Antiquity samples. Mycenaean, I mean.
 

This thread has been viewed 103810 times.

Back
Top