Greek Discussion Thread

[J Man]

Yes, it's clearly a beautiful place.







Thanks, guys. I have indeed looked into the patronym (Θωμάκος) and wondered if a Maniot connection was possible or even likely. My Greek islander ancestry is obvious, but I wouldn't be surprised if there is more to the story.

From what I can tell, J-L26 is the same subclade found in the only Mycenaean male (from the Peloponnesus no less) that we have. If this is common in Mani, that would be interesting to know. I'm not a Y-hap anorak and haven't had my lineage analyzed in much depth.





Yeah, I'm not getting any specific matches from Karpathos or much from the Aegean in general. Most of my closer matches (3rd and 4th cousins) are from the Peloponnese. A few list ancestry from Athens, Crete, North Aegean, and even Smyrna. Of the Peloponnesians, there are a few Spartans and a couple of Maniots, including apprently a full-blooded Maniot with all four grandparents from Selegoudi. This makes some kind of Mani connection pretty likely, I suppose. That would be personally gratifying because I think Maniots are the coolest Balkan Greeks.

23andme's regional analysis of my "Greek & Balkan" ancestry:



That's great, but my father's autosomal ancestry is still very insular-looking overall and in fact on the Near East-shifted side. 23andme predicts the part of my ancestry owed to him like so:

 

The conundrum is that my/his West Asian is high even for an Aegean Greek. I'm only half-Greek yet 23andme assigns me more West Asian than even some full-blooded Aegeans. This has always made me wonder if there's Cypriot in the woodpile. Romaniote or recent Near Eastern ancestry doesn't seem likely to me, but maybe a Maronite ancestor? I don't know-- that seems like a stretch. It could also be that this more Near East-heavy profile is common in the islands and we just haven't tested enough people yet. There are no Rhodians, Karpathians, or Kastellorizans in the G25. That's where I'd expect to see autosomal profiles like my father's show up.

EDIT: Found this on Reddit. This belongs to someone with ancestry from Karpathos and Rhodes:

 

I remember a number of Y-DNA samples from Mani that were posted in another thread on this forum and quite a few of them are J2a. Do you have cousin matches on 23andme with Maniots?

[23abc]

Ok, good to know, thought there is more to it, apparently there is not.
This really limits its validity.

Yep, and if 'num_gp' is standing for 'number of grandparents' (which I strongly believe it does), it mean all of these could be just because they are half Dodecanese half something else and listed their grandparents like that. There is some common idea floating around that they only use people that list '4 grandparents' from a particular region for this relative sharing, but the 2 people from Kayseri I do match on my relatives list don't list 4 grandparents. 1 lists 2 and the other lists 3, which matches the num_gp for Kayseri being 5 here. So yeah, this method has false positives more than AncestryDNA's region system but I don't mind it as AncestryDNA's one is useless (it can only tell me ancestors came from either Albania, Greece or Turkey... how useless!)

[Riverman]

Yep, and if 'num_gp' is standing for 'number of grandparents' (which I strongly believe it does)

It does, like for me:

"num_gp": 6, "subregion_id": "de:hesse:1", "num_relatives": 3, "state": "Hesse"}

So 6 grandparents from 3 matches.

, it mean all of these could be just because they are half Dodecanese half something else and listed their grandparents like that. There is some common idea floating around that they only use people that list '4 grandparents' from a particular region for this relative sharing, but the 2 people from Kayseri I do match on my relatives list don't list 4 grandparents

They seem to use people with 2 gp, whether they also use some with one is not possible to recalculate with the code for me, unless someone has "1 gp from 1 relative" for a region - if they use that too...

The strange thing which made me believe there is more to it is, that I have people with 4 gp from one proven ancestral region of mine, but they don't give me the region. In another case I'm not even sure I have enough matches from that region they put me. But apparently, its very simple maths plus a filter algorithm, like you suggested.

[23abc]

It does, like for me:


So 6 grandparents from 3 matches.



They seem to use people with 2 gp, whether they also use some with one is not possible to recalculate with the code for me, unless someone has "1 gp from 1 relative" for a region - if they use that too...

The strange thing which made me believe there is more to it is, that I have people with 4 gp from one proven ancestral region of mine, but they don't give me the region. In another case I'm not even sure I have enough matches from that region they put me. But apparently, its very simple maths plus a filter algorithm, like you suggested.

The 'simple maths' part is simply the shading of the regions using a ratio of ibd matches from one place to total ibd matches. Obviously there are more complex filters in play during the actual computation of ibd matches. I have matches that have 4 grandparents from some place and then I don't score that region. Similarly I've scored regions in the past (which are now gone) where I couldn't find anyone with that region on my actual relatives list (Emilia Romagna).

[Michalis Moriopoulos]

I remember a number of Y-DNA samples from Mani that were posted in another thread on this forum and quite a few of them are J2a. Do you have cousin matches on 23andme with Maniots?

Yep!

You surely know better, but just telling you my personal opinion, it just looks to me like you have a complex Greek ancestry. Probably something along the lines of Peloponnese (paternal line) with islander and Anatolian ancestry? People could migrate in strange patterns. And Athens would be an ideal place for such mixtures to become reality. I'm actually surprised by your strong result for the Peloponnes on 23andme. This gives the surname distribution and the conclusions derived from it much more credibility. I don't think you can have only islander ancestry.

That very well may be the case. I thought the inclusion of Rhodes into the G25 might shed some light, but my father's simulated coordinates still show closest distances to the Ashkenazi, Romaniote, and only then the Kos averages. The Rhodians don't show up high on the list even when I run distances to individuals. Again, though, it is just a sim. Who knows how a real genome would turn out.

Your result looks similar to one of my kids, except they score much lower Greek & Balkan, and more Italian. But this does look close to what a half Dodecanese scores if they're phased, except the Greek & Balkan would be a little lower.

At least Southern Aegean is your second strongest region, but it's not strongly coloured which is weird. It's possible your islander ancestry is from a place which isn't common in the diaspora and so you get no relatives or something.

The absence of cousin clustering is the biggest mystery to me. It could be an undertested small island, or an Anatolian community that moved en masse to Athens, and perhaps sent very few people to America/Australia and other places where testing is common. It could also mean a bigger community (or island) with less endogamy. Dodecaneses shifted Cretan?

Yes, that is strange, isn't it? I would expect to see higher relative sharing in the Southern Aegean. Perhaps the ancestry is from a more obscure area. Or maybe there is some significant Cappadocian ancestry in the mix? I thought that might explain the seeming Near Eastern shift of my father's sim in PCA:

 

Overall though he does seem to be quite Dodecanese-like autosomally, so I'm just not sure. It's quite the mystery. Maybe I shouldn't make too much of a simulation. Perhaps his real genome wouldn't be so outlying and cluster more elegantly with the Dodecanese samples.

By the way, in the current 23andMe versions, the average in Dodecanese is about 50% WANA and 50% Southern European, nearly all of which is Italian. Some people can score up to 70% WANA though, and others more Southern European. Greek & Balkan higher than 15% is usually a sign of admixture from somewhere else.

Thanks for the info. I now feel more confident that my high WANA score (and hence my father's even higher WANA) is not necessarily due to exotic ancestry. My mother doesn't have any significant "West Asian," "Italian," or "Greek & Balkan" ancestry which means I inherited it all from him. So if I inherited ~22% WANA from him, then he was likely ~44% WANA himself. Based on what you've just shown me, that's apparently well in the range of Aegean islanders. Also based on my score, his "Italian" might hover somewhere around 20% and his "Greek & Balkan" ~16%. And the "Broadly Southern Euro" I inherited from him might amount be assigned on his end to ~14% extra "Italian" and/or "Balkan" ancestry, were his genome tested directly as opposed to inferred from mine. Obviously these numbers are rough approximations based on little more than doubling my regional scores, but I think they're close enough to what we'd see if I had his actual genome.

I guess that's like Berlin for me, its about relatives which moved there, rather than me having ancestry from that city. But even then, you have islander ancestry and your Peloponnese is way lower than that of Michalis. And for him its now five strikes:
- surname distribution
- relative matches
- 23andme regional assignment
- higher Balkan (?) ancestral component
- actual known residence of the ancestor

On its own, every point wouldn't mean too much, but taken together, it might be more conclusive.

I think the cousin matches tell the tale. 3rd and 4th cousins from Mani/Peloponnese (closest Greek matches for Michalis, right?) likely didn't immigrate there. It is far more likely Michalis' ancestor moved from there to Athens a couple of generations ago and those cousins show the common great-great-grandparents from Mani.

For sure, I agree again with your first point that his ancestry is probably more complex than simply being from Karpathos. Not ruling that out, but I think he probably does have some ancestry from the mainland if that information about his father is correct.

Indeed, I'm quite amendable to the idea that there might be some minor Balkan ancestry at play here. I looked at the 23andme code for my "Greek & Balkan" ancestry. Here are the specifics:

Peloponnese
num_gp: 77
num_relatives: 43

Southern Aegean
num_gp: 21,
num_relatives: 9

Attica
num_gp: 20,
num_relatives: 12

Northern Aegean
num_gp: 15
num_relatives: 7

Crete
num_gp: 13
num_relatives: 9

Ionian Islands
num_gp: 7
num_relatives: 5

Epirus
num_gp: 4
num_relatives: 2

Michali.. surname, yDNA and autosomal Screams either South Peloponnese(prolly Maniot) or Greek islands(inclunding Crete).Also keep in mind that Maniots and Laconians in general have migrated to islands during the periods. There are many Greek islanders(included Cretans) With Maniot/Laconian roots. Also It might be a mixed situation. But anyway, without misunderstanding and i know that it dosnt concern me at all,but if i was you, i would focus more to your mother's genetics. And Btw, she has Nice genetics.... since she is Anglo-Saxon. Focus there Bro!!!

I do think a mixed situation is likely based on what I've learned in this thread. I hope I can iron out the specifics one day.

And thanks for the compliment, bro, but let's get real. People of NW Euro descent like my mama are easily the most studied people on Planet Earth. Do they really need my attention? I'm naturally much more interested in understanding that side of my ancestry that makes me physically and genetically distinctive from most of the WASPs I interact with on a daily basis. And come on, who doesn't love being Greek?

[J Man]

Yep!



That very well may be the case. I thought the inclusion of Rhodes into the G25 might shed some light, but my father's simulated coordinates still show closest distances to the Ashkenazi, Romaniote, and only then the Kos averages. The Rhodians don't show up high on the list even when I run distances to individuals. Again, though, it is just a sim. Who knows how a real genome would turn out.





Yes, that is strange, isn't it? I would expect to see higher relative sharing in the Southern Aegean. Perhaps the ancestry is from a more obscure area. Or maybe there is some significant Cappadocian ancestry in the mix? I thought that might explain the seeming Near Eastern shift of my father's sim in PCA:

 

Overall though he does seem to be quite Dodecanese-like autosomally, so I'm just not sure. It's quite the mystery. Maybe I shouldn't make too much of a simulation. Perhaps his real genome wouldn't be so outlying and cluster more elegantly with the Dodecanese samples.



Thanks for the info. I now feel more confident that my high WANA score (and hence my father's even higher WANA) is not necessarily due to exotic ancestry. My mother doesn't have any significant "West Asian," "Italian," or "Greek & Balkan" ancestry which means I inherited it all from him. So if I inherited ~22% WANA from him, then he was likely ~44% WANA himself. Based on what you've just shown me, that's apparently well in the range of Aegean islanders. Also based on my score, his "Italian" might hover somewhere around 20% and his "Greek & Balkan" ~16%. And the "Broadly Southern Euro" I inherited from him might amount be assigned on his end to ~14% extra "Italian" and/or "Balkan" ancestry, were his genome tested directly as opposed to inferred from mine. Obviously these numbers are rough approximations based on little more than doubling my regional scores, but I think they're close enough to what we'd see if I had his actual genome.







Indeed, I'm quite amendable to the idea that there might be some minor Balkan ancestry at play here. I looked at the 23andme code for my "Greek & Balkan" ancestry. Here are the specifics:

Peloponnese
num_gp: 77
num_relatives: 43

Southern Aegean
num_gp: 21,
num_relatives: 9

Attica
num_gp: 20,
num_relatives: 12

Northern Aegean
num_gp: 15
num_relatives: 7

Crete
num_gp: 13
num_relatives: 9

Ionian Islands
num_gp: 7
num_relatives: 5

Epirus
num_gp: 4
num_relatives: 2

I do think a mixed situation is likely based on what I've learned in this thread. I hope I can iron out the specifics one day.

And thanks for the compliment, bro, but let's get real. People of NW Euro descent like my mama are easily the most studied people on Planet Earth. Do they really need my attention? I'm naturally much more interested in understanding that side of my ancestry that makes me physically and genetically distinctive from most of the WASPs I interact with on a daily basis. And come on, who doesn't love being Greek?

What are the Y-DNA results of your Maniot matches?

[losAntonis]

Regarding Y-haplogroups from Maniots.
I am from north Greece and we have a family legend, that my paternal line came from a certain family of Mani. (Now, proven by Y-DNA to be wrong.)
Therefore, I did test some people with this surname from Mani.
All of them were J2a, but all with different subclades. So, they have the same surname, without beeing related. One other, with same surname is G2a.

[Johnny ola]

Regarding Y-haplogroups from Maniots.
I am from north Greece and we have a family legend, that my paternal line came from a certain family of Mani. (Now, proven by Y-DNA to be wrong.)
Therefore, I did test some people with this surname from Mani.
All of them were J2a, but all with different subclades. So, they have the same surname, without beeing related. One other, with same surname is G2a.

Do you know his G2a clade?

[losAntonis]

Deleted

[losAntonis]

Do you know his G2a clade?

SmartSelect_20201128-222144_Firefox.jpg

Based on Y37, predicted U1