Printable View
Examples of genetically heterogenous large ethnic groups of Europe:
- French
- Italians
- Germans
- Greeks
Examples of genetically homogenous large ethnic groups of Europe:
- English
- Spanish
- Romanians
- Belarusians
Examples of intermediate (neither too diverse nor homogenous) ones:
- Russians
- Poles
- Ukrainians*
- Dutch
*Ukrainians would be homogenous if not for South-Western Ukrainians.
Who are the most heterogenous/homogenous in each category in your view, and why?Quote:
Originally Posted by Peterski
I would say that when it comes to large ethnic groups of Europe (let's say at least 10 million people each), the English are the most homogenous, because despite there being almost 50 million ethnically English people in the UK, when you make a European PCA plot (for example based on K36 results), you will see that English people occupy very little space there relative to their number (about as much space as, for example, Danish people, despite being much more numerous than Danish people).Quote:
Originally Posted by J. Ketch
The Spanish occupy a larger space in this PCA (especially if you include the Basques and Catalans).
And Romanians occupy a similar or perhaps slightly larger space, but they are much less numerous.
Belarusians seem to be even more homogenous than the English, but they are much less numerous.
The Irish are also even more homogenous than the English, but again, they are less numerous.
When it comes to the most heterogenous large ethnic group, this is harder to conclude.
Because France, Italy and Germany occupy similar amounts of space in this European PCA (France and Italy a bit more).
It depends if you add populations like Memelland for Germans, or Corsica for the French.
But when it comes to Germans, there seems to be no such a thing as a "Central German" in genetic sense - they are quite sharply divided into NW Germans, South Germans and East Germans, with very few areas plotting truly "in-between". In case of France and Italy it is different, as in both countries you can discern respectively Central French and Central Italian clusters.
So France and Italy occupy more PCA space than Germany, but at least they have "central" clusters, while Germany doesn't.
As for the Spanish, it is believed that population movements during the Reconquista contributed to their “genetic homogenization.” At that time, there were large-scale migrations, especially from north to south, which led to significant mixing of populations between regions and, as a result, the homogenization of the genetic profile (the greatest differences in Spain are between the east and the west, and much smaller, for example, between the northeast and the southeast; precisely because the mixing of people during the Reconquista took place mainly along the meridian axis).
=====
EDIT:
Here is my K36-based PCA to illustrate the points which I'm making in this thread:
(please note that for example Memelland is not included in Germany in this PCA)
(this PCA is based on regions - each dot is a regional average, not an individual)
https://i.postimg.cc/Td9ZFPSW-/Europe-PCA-New.png
https://i.postimg.cc/Td9ZFPSW-/Europe-PCA-New.png
If you contrast the neighbouring Dutch to the English (or British in general) it's remarkable, despite the Netherlands being about 1/3 the size of England and 1/3 the population they cover maybe 3x the genetic area, overlapping with Scandinavians in the North and French/Austrians in the South.Quote:
Originally Posted by Peterski
The main bulk of the Spanish are quite homogenous but not so much if you include the outliers of the Basques and Canarians.
To me it makes the status of these as proper ethnic nations (based around a single genetic/ancestral community) questionable, or at least the status of outlier groups within them. Cultural empires vs proper nations, there should be a distinction made.Quote:
Originally Posted by Peterski
Why a „proper ethnic nation“ should just be when it’s „based around a single genetic/ancestral community“ seems a little bit unclear. What if it’s based on two, three or four main components? An ancestral community for centuries or more will regularly have existed anyhow.Quote:
Originally Posted by J. Ketch
Would the English not fulfil this criteria if the incoming Germanics in ancestry would just make up 40% of the English gene pool?
In the G25 North Europe PCA the Dutch/Flemish do overlap abt. 90% with the English. There is no overlap with Scandinavians etc.Quote:
Originally Posted by J. Ketch
The idea is that those ancient components are common to the whole population, because they all come from the same early medieval mixture that homogenised. If a nation doesn't have unique common origins (at least a majority of ancestry in common) surely it's more of a cultural entity than a racial one.Quote:
Originally Posted by rothaer
Is that not because you arbitrarily excluded Frisians? If you consider them ethnically distinct that supports the point about some nations being less proper.Quote:
Originally Posted by rothaer
Source? I don’t remember seeing some part of the Dutch cluster overlapping with Austria.Quote:
Originally Posted by J. Ketch
Serbs are homogenous genetically.
I added a K36-based European PCA (check post #4 in this thread).
In my K36 PCA English plot between Dutch and Irish, which I think makes more sense than overlapping with Dutch.Quote:
Originally Posted by rothaer
From back when I used to make G25 PCAs many years ago (small overlap but it's a small sample size):Quote:
Originally Posted by Opie
https://i.postimg.cc/281HTQgF/ryohazuki.png
English overlap with both (all NW European nations overlap with each other) but are closer to the Netherlands on average.Quote:
Originally Posted by Peterski
Many Frisians overlap strongly with Scots and Irish too, more so than with English (Dutch Bronze Age ancestry coming through).
Just ignore it. G25 Plotting is a joke.Quote:
Originally Posted by J. Ketch
Also Groningens iirc but I don't remember seeing a strong overlap.Quote:
Originally Posted by J. Ketch
Indeed.Quote:
Originally Posted by Peterski
You mean G25 samples are a joke? The plotting is just visualising them. Regardless some Dutch overlapping with Austrians is not a stretch. Dutch Limburg is Flemish-like, and Flemish are South German-like.Quote:
Originally Posted by Opie
re: the spread of ethnic Germans/Austrians, I saw some North Italian-like results from supposed German-speakers in Italy on a thread here a while ago.
edit: this is the post
https://www.theapricity.com/forum/sh...61#post7485861
Are we to consider these people ethnic Germans/Austrians?
There’s something wrong about them. I am far away from my closest matches in g25 PCAs.Quote:
Originally Posted by J. Ketch
A freshly prepared G25 PCA with some of the overlapping groups mentioned, from the very inadequate and limited G25 datasheet.
https://i.postimg.cc/LRX71Xcs/Vahadu...Views-(11).png
Portugal is as homogeneous as it can get in Europe, possibly the most Mediterranean country in Europe. Spain is phenotypically more varied imo.
What about Greece?Quote:
Originally Posted by Sebastianus Rex
Compared to the Portuguese (and probably Greeks too), I find that Spaniards have higher proportions per capita of both very blonde and very dark types. So yeah.Quote:
Spain is phenotypically more varied imo.
Are you even asking this? Portuguese is far more homogeneous than Greeks. Pontic and Cypriots Greeks do not even cluster with Europeans.Quote:
Originally Posted by Tooting Carmen
Greece has a strong dinaric component.Quote:
Originally Posted by Tooting Carmen
gotta give you a downvote for this bro. feel bad doing it but this is bull. Russians are the most genetically heterogenous nation in europe by far. Central Russians are closer to anatolian turks than to northern russians.Quote:
Originally Posted by Peterski
Peterski made a great point as usual (he is one of the better contributors to this forum), so take a deserved downvote from me.Quote:
Originally Posted by AndreiDNA
Anthrogenica and TA single ethnicity members' G25 coords aggregated in the North Europe PCA
https://i.imgur.com/xuLBSQ5.jpg
i can prove that Russians are the most diverse ethnicity in europe.Quote:
Originally Posted by Orion Arm
Attachment 148221
According to F2 statistics, Russian.HO is as close to Turks as to russians from leshukonsk. According to F2 statistics, Russian.HO is 3 times closer to Bulgarian than to Leshukonsk Russians.
Attachment 148222
Fst statistics show the same thing
Keep in mind that Russian.HO includes NORTHERN russians from the HGDP dataset. A group made up of ONLY central/southern russians would be even more distant from leshukonsk russians.
There is no other ethnicity in europe with a genetic range like that. so the thread is beyond silly, claiming Germans are more diverse than russians genetically
Now, deduce the logical explanation for why all Russians from Anthrogenica and TA cluster tightly together, and why not a single Russian individual from either forum clusters even remotely close to the Leshukonsky.Quote:
Originally Posted by AndreiDNA
Why Leshukonsky average would even be relevent?Quote:
Originally Posted by Orion Arm
Less than 10,000 people live in that area.
Andrei, are you using Leshukonsky samples from Jeong & Balanovsky 2019 study?
These samples:
Rakrlsh-002
Rakrlsh-140
Rakrlsh-143
Rakrlsh-144
Rakrlsh-149
I added them to my K36-based PCA and this is how they plot (see 5 black dots):
https://i.postimg.cc/Bqq1m6SJ/Leshukonsky.png
https://i.postimg.cc/Bqq1m6SJ/Leshukonsky.png
Maybe it is a limitation of K36 but as you can see they are not drastically different.
Their average / midpoint plots within the Russian area based on other averages.
In fact their midpoint plots right next to my average for Arkhangelsk Oblast (which includes also Krasnoborsky samples).
Can we see how Russians from Arkhangelsk Oblast plot in a G25-based PCA?
=====
Andrei can you show us a PCA where Central Russians plot closer to Anatolian Turks than to Northern Russians?
It can be a PCA from some academic study.
https://journals.plos.org/plosone/ar...l.pone.0058552Quote:
Originally Posted by Peterski
on this PCA in particular, central russians are way closer to Italians than to Northern Russians
https://journals.plos.org/plosone/ar...e.0058552.g003
The thing is you can setup PCA in any way you want. It's not objective, unlike F statistics.
How????Quote:
Originally Posted by AndreiDNA
PCA is very easy to manipulate. I can make a PCA showing anything I want.
Here's a PCA I made using G25 coordinates and only 3 anchors, Volga_IA, Latvian, and Mbuti. I set it up this way to maximize the distance between Northern Russians and other populations in europe, and minimize the differences between northern russians and any population "east" of the Urals. This PCA shows northern russians clustering closer to Yakuts than to central russians. Quite profound stuff.
Attachment 148250
I am not a big fan of PCA based calculators and especially using something this unstable and manipulatable to compare how diverse different ethnicities are. Of course using K36, Germans will be more "diverse" than Russians. Russian diversity is in the Eastern Europe - Volga Ural spectrum. Only 2 anchors. German diversity hits much more anchors. "French" "North Sea" "North Atlantic" "Iberian" "East Central European" and many more anchors. The same can be said of the anchors in G25. Although in G25, Russians are still the most diverse europeans with a max distance of 0.09 between russian Leshukonsk and Belgorod. I'm pretty sure the distance from Danish to Spanish is like 0.08...
Here's an F2 based PCA system I made with 3 anchors, which are Sardinian, Mansi, and Lithuanian. This is an "intellectually honest" way to envision european ancestry.
Attachment 148251