Modeling Polish ancestry

[Tomenable]

And Georg Hassel's data (for year 1819):

Provinz (Grand Duchy) Posen: https://books.google.pl/books?id=31DMAJgQV28C&pg=PA43



West Prussia 1819: https://books.google.pl/books?id=31DMAJgQV28C&pg=PA42



West Prussia 1824 (Karl Andree*): https://books.google.pl/books?id=xgUEAAAAYAAJ&pg=P212



Pommern: https://books.google.pl/books?id=31DMAJgQV28C&pg=PA31



Oppeln Silesia: https://books.google.pl/books?id=31DMAJgQV28C&pg=PA34



Breslau Silesia: https://books.google.pl/books?id=31DMAJgQV28C&pg=PA33



^^^
Other authors give similar numbers of Poles in Breslau Silesia (Kamusella, Ladenberger and Michałkiewicz for year 1840 give 80,000 Poles in Breslau Regency and 566,000 in Oppeln Regency; S. Plater gives 600,000 Poles in all of Silesia in 1824).

Liegnitz Silesia: https://books.google.pl/books?id=31DMAJgQV28C&pg=PA34



Brandenburg: https://books.google.pl/books?id=31DMAJgQV28C&pg=PA30

But according to T. Kamusella, in year 1815 there were still 200,000 Sorbian-speakers in Brandenburg (and 50,000 in Saxony):



In the link below August von Haxthausen in his 1839 book has data on East Prussia, by county:

https://books.google.pl/books?id=Gsj1FBg0gvUC&pg=RA1-PA78

County Allenstein (Olsztyn) for example:



Karl Andree gives for East Prussia 280,000 Poles and 200,000 Lithuanians (total population 1,080,000):

https://books.google.pl/books?id=xgUEAAAAYAAJ&pg=P218



*Karl Andree, "Polen: in geographischer, geschichtlicher und culturhistorischer Hinsicht", 1831.

Georg Hassel for some reason added East Prussian Poles to East Prussian Lithuanians. However, he also reported 20,000 Curonians and Latvians as well as 2,400 Jews in East Prussia (Andree's data for East Prussia is for year 1824, Hassel's data for year 1819).

[Tomenable]

They might come second, but from my experience its rather Silesia. However, most of the people I could evaluate from the Posener region tested on Ancestry and MyHeritage, which are by far not as reliable if its about small percentages of Ashkenazi and the majority wasn't 4/4 from the area anyway...
I think its not just about the percentage of Jews in a region, but also about the interactions with other people. And in Silesia and some areas of Poland they were probably less formalised-restricted than in others.

But those Posen Jews nearly disappeared by the 1910s.*

So they either all emigrated or partially also assimilated.

Question is if they assimilated with local Poles or Germans.

=====

*In 1921 Jews were only 0.5% of inhabitants of Poznan Voivodeship:



^^^
It can be added that in 1824 Congress Poland had already 10,8% Jews.

[Tomenable]

Polish admixture among Polish Jews is higher than the opposite:

Target: Ashkenazi_Poland
Distance: 1.0799% / 0.01079948
92.4 Ashkenazi_Germany
7.6 Polish

Target: Ashkenazi_Poland
Distance: 1.0862% / 0.01086155
93.0 Ashkenazi_Germany
7.0 Polish_Medieval(n=62)

[Tomenable]

Among my matches with 4/4 grandparents from Germany and who made their ancestry composition public, percentage of AJ admixture is:

1 - 0.0%
2 - 0.0%
3 - 0.0%
4 - 0.0%
5 - 0.0%

I will also add to this one match who has 4/4 grandparents from Wrocław, Poland (in fact pre-war Breslau, this match is ethnically German):

6 - 0.0%

Maybe Rothaer has more matches with 4/4 grandparents from Germany so he could update this statistics with information on his matches?

=====

Edit:

BTW, this is ancestry composition of Reiner, my 4/4 grandparents from Wroclaw match:



So he is more Slavic autosomally than Rothaer. He also has Slavic Y-DNA (R1a-CTS3402).

[Riverman]

Polish admixture among Polish Jews is higher than the opposite:

Target: Ashkenazi_Poland
Distance: 1.0799% / 0.01079948
92.4 Ashkenazi_Germany
7.6 Polish

Target: Ashkenazi_Poland
Distance: 1.0862% / 0.01086155
93.0 Ashkenazi_Germany
7.0 Polish_Medieval(n=62)

Most likely yes, but keep in mind that its more general Slavic, not necessarily Polish only. There seems to have been strong communities in Czechia for example, and of course East of Poland, where Jews picked up Slavic ancestry.

[Tomenable]

Most likely yes, but keep in mind that its more general Slavic, not necessarily Polish only. There seems to have been strong communities in Czechia for example, and of course East of Poland, where Jews picked up Slavic ancestry.

Well for East of Poland (historic Eastern Poland) we have other AJ regional references:

https://drive.google.com/file/d/1wZr-UOv...ZWb8y/view

Code:

Ashkenazi_Belarussia,0.1019155,0.1339716,-0.0073393,-0.0395799,0.0107712,-0.0144595,-0.0010123,-0.0010294,0.0057267,0.0146208,0.0024485,-0.0003688,0.00231,-0.001461,-0.002944,0.0027944,0.0018154,-0.0003409,0.0014889,-0.0012602,-0.0038586,-0.0007705,0.0027399,0.0015665,-0.0002949
Ashkenazi_Lithuania,0.0998539,0.1327575,-0.0092224,-0.040375,0.0094843,-0.0141475,-0.0005555,-0.001175,0.0060613,0.0137009,0.0015499,-0.0018529,0.0025678,0.0006257,-0.0044417,0.0034595,0.0016001,0.0001612,0.0004455,-0.0031721,-0.0032555,-0.0033497,0.0029245,0.0024428,-0.001056
Ashkenazi_Russia,0.1011888,0.131511,-0.0064488,-0.0382755,0.0103711,-0.014335,-0.0019505,-0.0001847,0.0065447,0.0140506,0.0011205,0.0002997,0.0021408,0.0003303,-0.0050352,-0.0002651,-0.0012125,-0.0005701,-0.0021367,-0.0031266,-0.0035687,-0.0029553,0.0006286,0.0045788,-0.0012214
Ashkenazi_Ukraine,0.0994244,0.1351162,-0.005506,-0.0401166,0.0120483,-0.0142652,-0.0010928,0.0001154,0.007772,0.0129388,0.0008038,0.0001125,0.0024156,0.0005229,-0.0028297,-0.0018297,-0.0043223,-1.9e-05,0.0012758,-0.0046772,-0.0031258,-0.0025904,0.0007702,0.0021028,0.0010059

I assumed that Ashkenazi_Poland refers to samples from the current borders of Poland.

[Riverman]

Among my matches with 4/4 grandparents from Germany and who made their ancestry composition public, percentage of AJ admixture is:

1 - 0.0%
2 - 0.0%
3 - 0.0%
4 - 0.0%
5 - 0.0%

I will also add to this one match who has 4/4 grandparents from Wrocław, Poland (in fact pre-war Breslau, this match is ethnically German):

6 - 0.0%

Maybe Rothaer has more matches with 4/4 grandparents from Germany so he could update this statistics with information on his matches?

=====

Edit:

BTW, this is ancestry composition of Reiner, my 4/4 grandparents from Wroclaw match:



So he is more Slavic autosomally than Rothaer. He also has Slavic Y-DNA (R1a-CTS3402).

I did a quick check for my kits at 23andMe, and there are not that many 4/4 from Germany which show their ethnicity estimate. Some of the matches from different kits:

Germans:

1/4 0,1 AJ and 0,2 WANA
1/4 0,2 WANA
1/4 0,4 WANA
1/4 nothing exotic

Austrians:

1/7 0,3 WANA
1/7 0,2 WANA and 0,1 SSA
1/7 0,1 CASA
1/7 1,1 WANA
3/7 nothing exotic

Czechs:

1/3 0,2 AJ
1/3 0,4 WANA and 0,1 SSA
1/3 nothing exotic

Poles:

1/1 Pole gets 6,2 AJ and 0,6 WANA and 0,2 SSA

There you get a noise/trace level WANA/AJ almost throughout at a 50 percent rate. This makes me question any such trace level ancestries, unless they being verified by matching. Doesn't mean they are wrong, but I just wouldn't take them for granted since 23andMe's assignment is not bulletproof.

Do the Poles get additional WANA at trace or higher levels, on top of the AJ? This WANA at a trace level seems to be far more common in Germans, Austrians and Czechs than AJ.

[rothaer]

Polish admixture among Polish Jews is higher than the opposite:

Target: Ashkenazi_Poland
Distance: 1.0799% / 0.01079948
92.4 Ashkenazi_Germany
7.6 Polish

Target: Ashkenazi_Poland
Distance: 1.0862% / 0.01086155
93.0 Ashkenazi_Germany
7.0 Polish_Medieval(n=62)

Haha, I get what you mean. (Askenazi_Germany are known to be notably "purer".) Considering that we talk of two people of a very different size mixing it can hardly became a different way. Not for a relative but for an absolute DNA exchange we had to multiplay 7.5% with the Askenazi_Poland (pre holocaust) figures and multiply the number of Poles with 0.8%. I actually think that the resuklt is pretty even.

[Tomenable]

Do the Poles get additional WANA at trace or higher levels, on top of the AJ?

No the Poles don't get WANA. But the Germans do, I also noticed it among my six 4/4 German matches.

This must be related to something else, not to AJ admixture.

[Tomenable]

Considering that we talk of two people of a very different size mixing it can hardly became a different way.

It could become a different way. We can imagine such scenarios. For example if only Jews converted to Catholicism but not the other way around, then there would be Jewish admixture among Poles but no Polish admixture among Jews. Or if only Poles converted to Judaism then there would be Polish admixture among Jews, but no Jewish admixture among Poles. But as we can see in reality admixture was in both directions.

[rothaer]

Among my matches with 4/4 grandparents from Germany and who made their ancestry composition public, percentage of AJ admixture is:

1 - 0.0%
2 - 0.0%
3 - 0.0%
4 - 0.0%
5 - 0.0%

I will also add to this one match who has 4/4 grandparents from Wrocław, Poland (in fact pre-war Breslau, this match is ethnically German):

6 - 0.0%

Maybe Rothaer has more matches with 4/4 grandparents from Germany so he could update this statistics with information on his matches?

=====

Edit:

BTW, this is ancestry composition of Reiner, my 4/4 grandparents from Wroclaw match:



So he is more Slavic autosomally than Rothaer. He also has Slavic Y-DNA (R1a-CTS3402).

This is also essentially in line with my hitherto impression.

People withn 4/4 grandparents born in Germany and seemingly being ethnic Germans:

1. 0.2%
2. 0.0%
3. 0.0%
4. 0.0%

Unfortunately there are not more such matches that do display their Ancestry Composition. There are no additional matches with at least 3/4 such grandparents.

I want to add that #1 above is an Eastern German hailing from Central Poland with ancestry that likely has been there since 1800. He has 62.3% EE (I've 50.7%) which is inconspicious for an Eastern German from Pomerania or (Lower) Silesia but Polish proper contributions can not totally be ruled out.

Interesting is one more individual that is an ethnic German from rural western fringe Masuria, located in the former Duchy of Prussia. This area was Slavic speaking a pretty long time and the population might have been Germanised as late as 1850/1900, maybe even later. All four grandparents are from the small village Peterswalde (Pietrzwald in Polish). Maybe you find some language figures applicable to that village.
He has 88.3% EE and 0.3% AJ.

This fact is now really interesting as this area became Slavic by the Teutonic Order there settling Polish Masovians neatly before 1500, after there were no German settlers available anymore and then the country switched to Protestantism in 1525 (The Duchy of Prussia was the first Protestant state worldwide). Such a thing regularly was a marriage barrier with Catholics. This suggests that the Masovians pre 1525 already had such contributions. An admixture with Germans will have happened since then but it must have been very modest (as suggested by his 88.3% EE) and also AJ contributions in German rural populations are essentially inexistent. A single case with a not fully known pedigree doesn't prove anything but I consider it a puzzle piece in this topic.

[Tomenable]

Congress Poland had 10,8% Jews in 1824 according to S. Plater (Karl Andree quotes Plater's figures) and Galicia together with Cracow Republic had 7,54% Jews in 1824 according to the same source (Cracow Republic had 100,000 Poles and 10,000 Jews according to Plater):

[Tomenable]

Interesting is one more individual that is an ethnic German from rural western fringe Masuria, located in the former Duchy of Prussia. This area was Slavic speaking a pretty long time and the population might have been Germanised as late as 1850/1900, maybe even later. All four grandparents are from the small village Peterswalde (Pietrzwald in Polish). Maybe you find some language figures applicable to that village.

He has 88.3% EE and 0.3% AJ.

This fact is now really interesting as this area became Slavic by the Teutonic Order there settling Polish Masovians neatly before 1500, after there were no German settlers available anymore and then the country switched to Protestantism in 1525 (The Duchy of Prussia was the first Protestant state worldwide). This regularly was a marriage barrier with Catholics. This suggests that the Masovians pre 1525 already had such contributions. An admixture with Germans will have happened since then but it must have been very modest (as suggested by his 88.3% EE) and also AJ contributions in German rural popultions are essentially inexistent. A single case with a not fully known pedigree doesn't prove anything but I consider it a puzzle piece in this topic.

I don't have data for this specific village (Pietrzwald), but the county where it is located (Kreis Osterode) was:

Year - German (%) / Polish (%) / Lithuanian

1825 - 8920 (28,3%) / 22552 (71,7%) / 0
1828 - 9268 (28,2%) / 23577 (71,8%) / 0
1837 - 10735 (30,6%) / 24333 (69,4%) / 0

So you are probably right that Germanisation of most of the territory of this county really started only after 1850.

I agree that his AJ is probably older than 1525, since Jews had no right to settle in the Duchy of Prussia, IIRC.

This would indicate that Poles from Masovia probably have a low level AJ admixture dating back to the Middle Ages.

[rothaer]

I'm curious if anyone is able to come up with better models? Especially I wonder if you can get a better distance for Polish_Medieval.

Here are the references you used and additionally a bunch of references otherwise used for proto Slavic ((Balto-)Slavic)

Code:

(Balto-)Slavic:HUN_Avar_Szolad_Av2,0.134311,0.126941,0.081458,0.065569,0.035391,0.033746,0.00987,0.005769,0.004704,-0.02278,-0.002436,-0.005395,0.01219,0.020643,-0.015201,-0.003845,0.005867,0.004561,0.008673,5e-04,0.001497,-0.00272,0.013804,-0.007109,0.002634
(Balto-)Slavic:RUS_Sunghir_MA_Sunghir6,0.129758,0.116786,0.070899,0.060078,0.041238,0.018965,0.009165,0.011307,0.001432,-0.017859,0.000487,-0.008093,0.014271,0.023809,-0.014658,-0.005834,0.00013,-0.002154,0.000628,-0.001251,-0.007986,-0.005812,0.012695,-0.010724,-0.005868
(Balto-)Slavic:DEU_Krakauer_Berg_MA_KRA001,0.129758,0.125926,0.075424,0.072029,0.046778,0.026774,0.011986,0.010384,-0.001636,-0.021322,-0.003085,-0.006294,0.008028,0.021194,-0.013843,0.013789,0.019036,-0.005701,0.00352,0.005378,0.004991,-0.008408,0.000986,-0.006145,0.001796
(Balto-)Slavic:DEU_Krakauer_Berg_MA_KRA009,0.126344,0.127957,0.067127,0.070737,0.042162,0.02008,0.011281,0.014999,-0.001023,-0.032802,-0.001949,-0.01124,0.017542,0.01968,-0.012893,0.007558,0.011604,0.00114,0.005782,0.011506,-0.010357,0.003091,0.007765,-0.008796,-0.001676
(Balto-)Slavic:DEU_Krakauer_Berg_MA_KRA011,0.125205,0.129988,0.078818,0.072675,0.047701,0.027889,0.013161,0.006692,0.005522,-0.020593,-0.00065,-0.009292,0.018583,0.027249,-0.011672,-0.010342,-0.006389,0.00266,0.008045,0.002876,-0.011729,-0.008037,0.007272,-0.003494,0.003952
(Balto-)Slavic:Moravia_POH13,0.127482,0.132019,0.083721,0.078489,0.050471,0.026495,0.00517,0.018922,0.004295,-0.028429,-0.004384,-0.006744,0.017988,0.043076,-0.008143,-0.011668,0.005346,0.000507,0.004022,0.012256,-0.012228,-0.008903,0.01368,-0.013496,-0.002395
(Balto-)Slavic:Moravia_POH28,0.141141,0.136081,0.087492,0.081719,0.047086,0.027052,0.019976,0.023999,0.010635,-0.01385,-0.001299,-0.014537,0.027502,0.03647,-0.015065,-0.007425,-0.014473,-0.005954,-0.005531,0,-0.007237,-0.004204,0.000616,-0.00964,-0.002275
(Balto-)Slavic:VK2020_UKR_Lutsk_MA:VK541,0.127482,0.125926,0.062602,0.053618,0.035699,0.02008,0.00658,0.016845,-0.0045,-0.015308,0.000974,-0.008093,0.016947,0.02202,-0.016286,0.001989,0.007562,-0.001647,-0.002011,0.001626,-0.006738,-0.013107,0.012941,-0.00723,0.00012 

displayed in the North Europe PCA:



You used for proto Slavic two Baltic-like Himera samples vom 480 BCE and a sample from Viminacium (Serbia) from abt. 146 AD. I consider that assignment arbitrary and not legit. First, these samples are from notably before Slavs at all showed up in history and second, when Slavs showed up in history, they were genetically not like that. And third, you have no clue where these samples from Sicily and Serbia hail from.

The distance for Polish medieval gets that bad because your three Baltic-like alleged proto Slavs likely were no. In your model you don't provide the somewhat more southernness that was common to early Slavs and in particular your single Przeworsk sample is not suitable for that. The early Slavs - (Balto-)Slavic in the PCA - are not in the cline between your alleged proto Slavs and and that Przeworsk sample.

I'm not sure why you include the two depcited expressed Wielbark outliers in the Wielbark mean. Although these individuals doubtlessly existed and thus were present, they have obviously a notably different ancestry.

In general there is not much to oppose to picking suitable samples. Out of just two Przeworsk samples - one is Germanic-like and the other is Celtic-like - you pick just the Celtic-like. Okay, but it seems not legit to label that Przeworsk as if it would be representative of the Przeworsk culture. In the matrieal culture the latter was a Celtic-Germanic mixture and the sum of BOTH samples are genetically well representative of that.

I'd simply re-name that reference Celtic-like and select a proxy for Germanic that is from the center of that cluster and the same as for Proto_Slavic.

I then suggest for your Polish_Medieval model the following source:

Code:

Proto_Slavic:DEU_Krakauer_Berg_MA_KRA009,0.126344,0.127957,0.067127,0.070737,0.042162,0.02008,0.011281,0.014999,-0.001023,-0.032802,-0.001949,-0.01124,0.017542,0.01968,-0.012893,0.007558,0.011604,0.00114,0.005782,0.011506,-0.010357,0.003091,0.007765,-0.008796,-0.001676
Celtic-like:Przeworsk_PCA0012,0.135449,0.145221,0.054682,0.029716,0.050163,0.008646,-0.002585,0.008538,0.01309,0.017677,-0.002436,0.009142,-0.010109,-0.007294,0.011265,0.011403,0.004172,0.008361,0.00905,-0.000375,-0.000125,0.005317,-0.003697,-0.003976,0
Germanic:Weklice_R10634,0.133173,0.136081,0.071276,0.067507,0.041546,0.023706,0.003995,0.015461,0.005931,-0.008018,-0.009906,0.004046,-0.008028,-0.006055,0.021851,0.002652,-0.00665,-0.00228,0.004777,0.001,0.004243,-0.003462,0.00419,0.013255,-0.005389

This would yield the following results and you see that the distance for Polish_Medieval is better than in your model:

Target: Polish_Medieval(n=62)
Distance: 2.0932% / 0.02093248
82.2 Proto_Slavic
17.8 Germanic

Target: Polish
Distance: 1.9904% / 0.01990359
68.6 Proto_Slavic
18.0 Germanic
13.4 Celtic-like

Target: rothaer_scaled
Distance: 1.8334% / 0.01833374
41.2 Germanic
32.2 Proto_Slavic
26.6 Celtic-like


Actually, Polish_Medieval is pretty northern and doesn't need any separate Celtic-like.

[Galadhorn]

Tomenable, regarding the Jewish admixture:

(1) such a small percentage (less than 1%) may primarily be noise or a "system" error;

(2) I myself have 0.9% on 23andMe common genes with Ashkenazi Jews (this is not the same as admixture), and since I know my genealogy well, I will explain to you what it is.

My Catholic and Polish/Slavic great-great-grandparents from near Rzeszów - Jakub Pękala and his wife Katarzyna née Samek - apparently had a daughter who married a Jew. Today, the offspring of this union are quite a large group of Ashkenazi Jews from the USA, including a New York Times journalist Emma Goldberg. We have common genes, 23andMe counts it as common autosomes with Jews, but the thing is that these Jews (influential and very culturally Jewish) have the genes of Catholics from Rzeszów, and not I have the genes of Jews from Rzeszów. Emma Goldberg herself is my closest match on many websites (in GEDmatch, 23andMe, MyHeritage), because she sent DNA to the same websites as me.

Regards