Some experiment with Greeks from Himera

[Francesco]

Some experiments with Greeks from Himera

I was just eperimenting and messing around with G25, so take this thread just as some sort of disengaged divertissement, without any serious or scientific aim.
For now, I'll ust post the following model. I'll expand on the matter in the future, but it seems many of the greek soldiers from the 480 (cluster number 1) could have been of mixed sicanian-greek origin. So, they might not be a good proxy for archaic age colonist from Grecia. Of course, the model could be totally coincidential, so take it with a grain of salt.

 

[Francesco]

I tried to run the same model with Mycenean samples as a target population: interestingly, despite the samples plot roughly in the same spot as the Himeran soldiers, the model shows a slightly worse fit. So, maybe, the cluster of greek soldiers from Himera might actually have harboured some degree of sicanian admixture...I wouldn't bet on it, but it may be a possibility to take in to account nontheless.

 

[peloponnesian]

It's possible they had a small amount of local admix but I don't understand why you used ITA_Sicily_Himera_Civilian as the Greek proxy instead of Empuries/Greece_IA/Mycenaean samples. The Himera_civilians average includes just 2 samples (one of which is an outlier compared to the typical Greek BA/IA profile). Almost all Y DNA branches from the 480BC cluster 1 have close matches with ancient samples from Greece itself (the J-M319 clusters with Minoan samples, the T with Mycenaean from Kastrouli, G-Z42565 with an LBA from Chania, etc).

The 409 BC soldiers had more obvious Sicilian admixture, as the paper itself suggested.

 

[Francesco]

Yes, I could have used Empuries as a proxy for IA Greeks (of which we still lack samples), maybe I will.
I chose those two Himeran civilians just because they are from the same geographical context and because they seem to form some sort of cline with the other two clusters.

 

[Francesco]

Ok, as suggested by peloponnesian, I tried to run the model using Empuries as a proxy for IA_Greece. Interestingly, the fit stays somwhat valid for the Himera cluster, while it reamains not so good for Myceneans...This could indeed confirm that at least some of the sample from Himera have some degree of Sicanian ancestry.

This could aslo be inferred by the PCA itself, where the Himera cluster could be divided into two subgroup: one wich is very close to Empuries and to the Himeran civlians (samples n. 10945 and n. 7221) and the other which is slightly shifted towards the Sicanians.

 

[Francesco]

...This could indeed confirm that at least some of the sample from Himera have some degree of Sicanian ancestry.

I just found out that the same hypothesis is formulated by the authors of the paper as well.

In the supplements, at pag. 25, we read the following statement: "We find among the soldiers of the 480 BCE battle three (I10945/W0494, I10952/W2738 and I7221/W2825) which are best modeled as deriving all of their ancestry from Greece_LBA without added genetic input. For 4 soldiers (I10948/, I7217/W2590 and I7218/W2737 and I7219) the most parsimonious model is an admixture of around 52-86% Greece_LBA and 21-48% Sicily_LBA or Sicily_IA."

 

[peloponnesian]

Yes you are right, I didn't remember the paper's analysis for the 480 BC cluster.