Italian 23andme result from southern Apulia -- Lecce, Salento area.

[Sikeliot]

Usually Apulians score high Balkan, but this is the first one from southern Apulia. It is similar to other southern Italians. She also has a grandparent from Lucania, so 100% mainland southern.

http://i57.tinypic.com/wbcby1.jpg
http://i60.tinypic.com/vxkwuc.jpg

[Green]

Speculative, Standard or Conservative?

[Alessio]

http://s20.postimg.org/gxqq3ozd5/Puglia_taranto.jpg (http://postimg.org/image/gxqq3ozd5/)

Taranto

[Neon Knight]

Interesting how there is almost no Ashkenazi whereas GEDmatch has trouble distinguishing between them and Italians.

How do Albanians tend to come out on 23andMe?

[Journeyman26]

Interesting how there is almost no Ashkenazi whereas GEDmatch has trouble distinguishing between them and Italians.

How do Albanians tend to come out on 23andMe?

23andMe is pretty careful with Ashkenazi ancestry, as "Jewish" ancestry is so dependent on where the Jews had been living. In the countries of ancestry section you have to select a box to differentiate between declared ashkenazi heritage. I am not sure what GEDmatch uses as a sample population for Ashkenazi. Usually shows up somewhere #7-12 on my list for some reason

[Longbowman]

ARGH because IGNORE GEDMATCH ORACLES they DON'T SHOW ANCESTRY so much as a line of best fit GEDMATCH measures admixture not ethnicity.

[Era]

Interesting how there is almost no Ashkenazi whereas GEDmatch has trouble distinguishing between them and Italians.

How do Albanians tend to come out on 23andMe?

0 Ashkenazi 0 Middle eastern.

I plot with ashkenazis in the world map ( the rest of Albanians don't) however that's because of lack of eastern and low northern euro. On the Southern Euro map I am right in the middle of Tuscan (there is no ashkenazis there) and on the northern Euro map on the corner of Austrian. It's just a numerical plot in the end.

[Sikeliot]

http://s20.postimg.org/gxqq3ozd5/Puglia_taranto.jpg (http://postimg.org/image/gxqq3ozd5/)

Taranto

High Balkan for an Italian -- either signals Greek ancestry, or ancient Illyrian.

[Petalpusher]

Interesting how there is almost no Ashkenazi whereas GEDmatch has trouble distinguishing between them and Italians.


Because where you plot or the population you are close to, doesn't equal ancestry or ethnicity on Gedmatch. You could plot with Danes and have 25% Ashkenazi or plot with Greeks, Sicilians and have none, you are only relatively close in genome similarity but any combination really can be close to another population. These tiny chunks of Ashkenazi are absorbed genes in the South, not recent ancestry.

[Neon Knight]

Because where you plot or the population you are close to, doesn't equal ancestry or ethnicity on Gedmatch. You could plot with Danes and have 25% Ashkenazi or plot with Greeks, Sicilians and have none, you are only relatively close in genome similarity but any combination really can be close to another population. These tiny chunks of Ashkenazi are absorbed genes in the South, not recent ancestry.

Yeah, I understand that GEDmatch produces an average genetic location, but the fact that a pure Italian can come out as Ashkenazi shows the limitations of the GEDmatch method - it clumps too much together.

[Petalpusher]

Yeah, I understand that GEDmatch produces an average genetic location, but the fact that a pure Italian can come out as Ashkenazi shows the limitations of the GEDmatch method - it clumps too much together.

It's the excess of MENA in the South that makes the difference, Italian have relatively a lot of East-med so some is already integrated in the Italian cluster. When you add up even more, it shows up as some MENA on 23 and drifts some closer to other pop which are basically intermediates euro/eastmed mix. South Italy (and some Greeks) overlap a lot with Ashkenazi at this point.

Benchmarkoid is a particular case


This West Eurasia pca is the most comprehensive imo :

http://s15.postimg.org/y7hux51u3/West_Central_Eurasia.png

[Neon Knight]

This West Eurasia pca is the most comprehensive imo :

http://s15.postimg.org/y7hux51u3/West_Central_Eurasia.png

On a side-note, remove Sheps & Ashk's from that map and the European group becomes significantly clearer. Then ignore Cyprus and it becomes clearer still.

[Petalpusher]

On a side-note, remove Sheps & Ashk's from that map and the European group becomes significantly clearer. Then nuke Cyprus and we are really talking.

Sir, my darkest sense of humor suggest you would release some lest at the wagon's tail.

[Era]

High Balkan for an Italian -- either signals Greek ancestry, or ancient Illyrian.

I dont think it's that ancient. Some arbereshe probably.

[Longbowman]

Yeah, I understand that GEDmatch produces an average genetic location, but the fact that a pure Italian can come out as Ashkenazi shows the limitations of the GEDmatch method - it clumps too much together.

No it doesn't, also there's no 'GEDmatch method,' GEDmatch is a site that hosts various calculators. GEDmatch itself does not create the calculators. The calculators show a variety of components. The fact remains that the Ashkenazi and southern Italians are similar. Not everyone is going to be that different. Sicilians and Ashkenazi overlap on all professional studies so why they wouldn't on Eurogenes is beyond me.

ignore the oracles, always. Always.

[Neon Knight]

No it doesn't, also there's no 'GEDmatch method,' GEDmatch is a site that hosts various calculators. GEDmatch itself does not create the calculators. The calculators show a variety of components. The fact remains that the Ashkenazi and southern Italians are similar. Not everyone is going to be that different. Sicilians and Ashkenazi overlap on all professional studies so why they wouldn't on Eurogenes is beyond me.

ignore the oracles, always. Always.

The GEDmatch calculators (the main ones anyway) are similar in that they all analyse a person's ancestry in terms of a matrix of rather arbitrary genetic pools like East Med and North Sea then position an individual somewhere between them. It roughly works but a lot of people are reading too much into it IMO. Ashkenazis are certainly very close to south Italians on those genetic maps but if they are 30-40% Palestinian then they must be about 60-70% central Italian on average. I don't know if other studies back that up.

[Longbowman]

The GEDmatch calculators (the main ones anyway) are similar in that they all analyse a person's ancestry in terms of a matrix of rather arbitrary genetic pools like East Med and North Sea then position an individual somewhere between them. It roughly works but a lot of people are reading too much into it IMO. Ashkenazis are certainly very close to south Italians on those genetic maps but if they are 30-40% Palestinian then they must be about 60-70% central Italian on average. I don't know if other studies back that up.

More or less, they're like 40-45% Levantine and 50-55% European, of which mostly North Italian, and some spare change from Asia and Africa, but the oracles, which is what I'm talking about, are next to useless.

[Neon Knight]

More or less, they're like 40-45% Levantine and 50-55% European, of which mostly North Italian, and some spare change from Asia and Africa, but the oracles, which is what I'm talking about, are next to useless.

That's basically the equivalent of being half Italian, half Russian - not especially 'mixed' really. The European genetic cluster is such a long shape that it doesn't really qualify as a cluster. We could call it a geno-cultural construct.

[Longbowman]

That's basically the equivalent of being half Italian, half Russian - not especially 'mixed' really. The European genetic cluster is such a long shape that it doesn't really qualify as a cluster. We could call it a geno-cultural construct.

No it isn't? I meant ~50% recent European, if we go by archaic admixture, they're about 20% WHG, 70% Neolithic, most of the rest ANE with some small Asian and SSA components. However, they certainly fit in to the 'European genetic cluster,' at the far end.

[Alessio]

Whereas 23andme go's by precise haplotype matches to confirm ancestry (this method needs lots of samples), Gedmatch relies on allele frequencies for analyzing ancestry which usually goes back thousands of years and thus trying to make one cluster with populations with a more similar ancient admixture or common ancestry from prehistoric ancestral groups which no longer exist.

GEDmatch calculators are very useful for spotting recent admixture when comparing your results with others who share the same ancestry.

[Neon Knight]

No it isn't? I meant ~50% recent European, if we go by archaic admixture, they're about 20% WHG, 70% Neolithic, most of the rest ANE with some small Asian and SSA components. However, they certainly fit in to the 'European genetic cluster,' at the far end.

I think you might have misunderstood. I didn't mean the Ashkenazi cluster was not really a cluster (it obviously is) but that the European cluster itself is not really a cluster. In other words, there is no such thing as genetically European. So the question of whether Ashkenazis are genetically European or not disappears.

[Longbowman]

I think you might have misunderstood. I didn't mean the Ashkenazi cluster was not really a cluster (it obviously is) but that the European cluster itself is not really a cluster. In other words, there is no such thing as genetically European unless we combine that idea with culture. So the question of whether Ashkenazis are genetically European or not disappears.

Yes, I concur. There is a 'European component' - WHG - that's more or less unique to Europe but it is only a co-modal component.

[Alessio]

No it doesn't, also there's no 'GEDmatch method,' GEDmatch is a site that hosts various calculators. GEDmatch itself does not create the calculators. The calculators show a variety of components. The fact remains that the Ashkenazi and southern Italians are similar. Not everyone is going to be that different. Sicilians and Ashkenazi overlap on all professional studies so why they wouldn't on Eurogenes is beyond me.

ignore the oracles, always. Always.

Why would one have to ignore the oracles ? From the comparisons between people of similar ancestry I did, the oracles can actually be very useful and usually they split/predict someones ancestries fairly correct (if the 2 populations are distant enough)

[Longbowman]

Why would one have to ignore the oracles ? From the comparisons between people of similar ancestry I did, the oracles can actually be very useful and usually they split up someones two ancestries fairly correct (if the 2 populations are distant enough)

The oracles are a line of best fit. People take them as signifying actual ancestry. They do more harm than good.

[Alessio]

The oracles are a line of best fit. People take them as signifying actual ancestry. They do more harm than good.

If you'd treat them as such they'd actually do harm to someones knowledge of how the oracles work.

[Longbowman]

If you'd treat them as such they'd actually do harm to someones knowledge of how the oracles work.

Pardon?

[Alessio]

Pardon?

you as in a general person 'you'