South Italy & Balkans PCA and dendrogram created my user "mlukas"

[Sikeliot]

averaging together the results from all 100+ of my samples.

You can indeed see that Greek islands, Greek mainland, and Sicily all form their own clusters, with the former being Anatolian shifted, the latter Levantine shifted, and Greek mainlanders near Albania.

Dendrogram shows that there is one overarching "Greek-South Italian" cluster, which can then be divided among Albanians and Greeks from the mainland on one side, and Greek islanders/Sicilians/South Italians on the other, and then Cypriots separately. Greeks from Kythera island fall, further, into the South Italian cluster when Greek islands/Sicily divide.

The exception are Cyclades islanders, who instead fall into the mainland Greek and Albanian cluster along with, interestingly, Istanbul Greeks.

Moreover, Messina, Caltanissetta, and Ashkenazim cluster together, as do Catania and Calabria, and then the rest of West Sicily together.

https://s24.postimg.org/l19itfm4j/ward.jpg


Scatter plot shows Dodecanese Greeks up near Armenians/Assyrians, Sicilians drifting right to the Levant and Cyprus, and the rest of Greeks plotting more or less together.

https://s15.postimg.org/xgos714bt/mds.jpg


PCA: shows the same

https://s17.postimg.org/682jfmff1/pca.jpg


Another scaling:

https://s9.postimg.org/9au19v9y5/PCo_A.jpg

[Sikeliot]

I forgot to mention it uses Eurogenes K36.

[Dick]

I forgot to mention it uses Eurogenes K36.
Ok

[Ylla]

Am I included here? Unlikely, but would be cool to see my position.

[Coolguy1]

Could you include mine too

[Kriptc06]

excuse me, what program is that?

[Sikeliot]

Could you include mine too

You're included in one of the Peloponnese groupings, which are close to one another.

Anything surprising to you?

The same seems true here as in that Sazzini et al study -- far eastern Sicily is more genetically outlying than is Calabria, despite that we often expect the opposite.

[Lucas]

Pm me your kit numbers or k36 scores. And i add positions

[Coolguy1]

You're included in one of the Peloponnese groupings, which are close to one another.

Anything surprising to you?

The same seems true here as in that Sazzini et al study -- far eastern Sicily is more genetically outlying than is Calabria, despite that we often expect the opposite.

There is nothing surprising to me here, is k36 a good test?

[Sikeliot]

There is nothing surprising to me here, is k36 a good test?

Yes. It is very good for a PCA because it has so many components that it is reliable and likely not mixing anything up.

[Sikeliot]

Also contrary to what I expected, Calabria does appear Greek-shifted here.. while we think of Calabria as exotic, it is actually plotting Greek-shifted, and the Sicilians are the ones shifting toward Palestine.

[Sikeliot]

Actually I realized, the divide goes like this:

Greek Mainland/Albania
vs
Cyprus and Sicily/Greek Islands

Then the latter splits into Cyprus on one hand, and Sicily and Greek islands in turn split again. So Cypriots are actually closer to Sicilians and Aegean islanders, not to Lebanese.

[Danaan]

I forgot to mention it uses Eurogenes K36.

Can you tell him to add Druze and Lebanese samples?

[Sikeliot]

Can you tell him to add Druze and Lebanese samples?

I think "Levant" there is Lebanese and Druze.

[Sikeliot]

Ok I found out, Levant on there is a mixture of Lebanese and Syrians.

[Argentano]

Pm me your kit numbers or k36 scores. And i add positions

add me please

Population
Amerindian -
Arabian 4.67
Armenian 1.92
Basque -
Central_African -
Central_Euro 1.19
East_African -
East_Asian -
East_Balkan 6.68
East_Central_Asian -
East_Central_Euro -
East_Med 14.43
Eastern_Euro -
Fennoscandian -
French 0.62
Iberian 15.00
Indo-Chinese -
Italian 22.67
Malayan -
Near_Eastern 13.77
North_African -
North_Atlantic 3.17
North_Caucasian -
North_Sea 3.87
Northeast_African -
Oceanian -
Omotic -
Pygmy -
Siberian -
South_Asian -
South_Central_Asian -
South_Chinese -
Volga-Ural -
West_African -
West_Caucasian 6.03
West_Med 5.98

[Sikeliot]

add me please

What are your thoughts on this? The key thing that is surprising to me, but consistent with Sazzini et al, is for how much we view Calabria as the exotic one, it is actually NE Sicily with the most West Asian genes (and here, all of the Sicilians except Trapani, for some reason, seem more Levant shifted than the Calabrese, let alone the Greeks).

It kind of makes us rethink what we thought we knew about population movements.

[Lucas]

add me please

8

Every adding change the plot a little.

You are Iberian shifted here.
It's PCA
https://s7.postimg.org/6r34fs9l7/argentani.jpg

[Argentano]

Every adding change the plot a little.

You are Iberian shifted here.
It's PCA
https://s7.postimg.org/6r34fs9l7/argentani.jpg

why am i so low in this PCA?

[Sikeliot]

Ok in that one, the Greek islanders and Sicilians are closer to one another which makes more sense to me.

[Argentano]

What are your thoughts on this? The key thing that is surprising to me, but consistent with Sazzini et al, is for how much we view Calabria as the exotic one, it is actually NE Sicily with the most West Asian genes (and here, all of the Sicilians except Trapani, for some reason, seem more Levant shifted than the Calabrese, let alone the Greeks).

It kind of makes us rethink what we thought we knew about population movements.

Well all seem very close so not much difference right?

[Sikeliot]

Well all seem very close so not much difference right?

Greek islands/SOuth Italy/Cyprus form one overarching cluster more or less.

[Sikeliot]

I wanted to bump this in case anyone wanted to see where they plot, maybe we can have the plot recreated with you on it.

What is too bad is we did not get a sample from Enna. Enna has a lot of Lombard towns, but they lived very segregated from the mainstream Sicilian population, and apart from some Byzantine Greek settlements, Enna was relatively isolated from foreigners. I suspect them to have been close to those of Catania, Messina, Caltanissetta, and the eastward part of Palermo province, all of which likely have 90%+ "original" Sicilian ancestry with little foreign input, and are the purest examples of the original Sicilian population.

The only foreign input into Messina, apart from a small amount of Greek input, would have been from Calabria which would be close to genetically identical to begin with.

I notice people on one or two other forums are trying to say that these results are invalid because people mixed and moved around the island (or they think I'd lie about the origins of the samples) but there are so many, that the sample size is even larger than in any professional study. I think it is quite accurate.