PDA

View Full Version : R1b in Central - Eastern Europe



Gravetto-Danubian
01-23-2016, 05:31 AM
It is interesting to look at the frequency of R1b in Europe, and the sub-clade proportions

7410

7411

The absolute nadir is in the forest zone of eastern Europe (Belarus, east Baltic, Russia) suggesting that R1b-M269 groups did not move north much beyond the forest steppe. Much of the small percentage (8%) in Estonia is due to U106 - likely brought in by Germanic contacts. The low frequency is maintained through to Ukraine (5%) - which is situated where western Yamnaya was - suggesting significant turnover since the Copper Age (or western Yamnaya was mostly R1a (?))

Poland is a different story, however, with 21% frequency. A solid chink is U106, but much (10%) is L23 xM412/L51 stuff. This is probably Z2013 groups ?

Going to the Czech lands, of the laudable total 22% M269, 6% is possibly Z2013, 6% U106, and 6% fell in what Myres et al called P312* (x L21, x U152). So this could be stuff like DF27, S182, or other not yet described SNPs.

A decent frequency (~ 20%) is maintained in Slovenia and Hungary. 25% of that would be Z2013, 25% U152, 25% P312*, 25% U106, roughly.

The frequency drops precipitously in the Dinaric region - Herzegovina, Bosnia, especially.

A high frequency in Kosovar Albanians (21%) and Albanians 17% is seen, much of which is likely Z2103. However, some 20-30% is x L23, quite likely that PF7562 described in the ftDNA "ht 35" Project.

Bulgaria also has double-digit frequencies (10%), of which half of probably Z2013, as with Greece (10%, and up to 8% being Z2013).

All in all, much of R1b in the Balkans is probably Z2013 (5-10%), and of the P312 groups, most is U152 (2-6%) - like in next door Italy, followed by a trickle of U106 (2-4%).

I also included Turkey, which had a healthy 15% of R1b, 13% of which is likely Z2103.

For comparisons, I also included some R1b-rich countries - Austria, Northern Italy, Ireland, Germany & Netherlands.

______________________


Its difficult to make any conclusions with such a paucity of aDNA from southeastern and east-central Europe. However, the overall impression one gets is that R1b was probably much higher there during the Iron Age. Its reduction - most felt in the Dinaric hinterland - must have to do with the Slavic migrations following the demographic collapse in Rome's Balkan regions after centuries of making a large bulk of the (East) Empire's soldiery (and Emperors), and having been not so densely populated to begin with. I'd say the same is true for eastern Europe further up, with drops in R1b frequency due to arriving I2a1 and R1a-M458 from Carpatho-Ukraine. I.e. I'd bet R1b had a higher frequency in the Iron Age in places like Czecho-Slovakia and Poland during the Iron Age than now (but in turn it had possibly expanded westward - somewhat diminishing the prevalence of R1a groups which arrived with CWC).

It is difficult to post theories as to the route of M269 migration through Europe based on modern east European M269 frequencies, all the more acute given the dynamic character of the eastern continent - with several migration waves, cultural turnovers, etc, between the EBA and late Antiquity.

Based on this, it is difficult to say which route M269 took to reach western Europe - i.e. south or north of the Carpathians. But given the dearth (? complete absence) of R1b in CWC, south is looking more probable.


Data for frequencies was collated from Myres et al (2011). Additional data for Albanians - Sarno et al (2015), Bulgarians (Karachanuk 2013).

northkerry
01-23-2016, 06:35 AM
Smal's R1b Y-tree is showing a lot of new branches under L23 so there seems to be a major R1b expansion after L23's birth.

alan
01-23-2016, 10:22 AM
It is interesting to look at the frequency of R1b in Europe, and the sub-clade proportions

7410

7411

The absolute nadir is in the forest zone of eastern Europe (Belarus, east Baltic, Russia) suggesting that R1b-M269 groups did not move north much beyond the forest steppe. Much of the small percentage (8%) in Estonia is due to U106 - likely brought in by Germanic contacts. The low frequency is maintained through to Ukraine (5%) - which is situated where western Yamnaya was - suggesting significant turnover since the Copper Age (or western Yamnaya was mostly R1a (?))

Poland is a different story, however, with 21% frequency. A solid chink is U106, but much (10%) is L23 xM412/L51 stuff. This is probably Z2013 groups ?

Going to the Czech lands, of the laudable total 22% M269, 6% is possibly Z2013, 6% U106, and 6% fell in what Myres et al called P312* (x L21, x U152). So this could be stuff like DF27, S182, or other not yet described SNPs.

A decent frequency (~ 20%) is maintained in Slovenia and Hungary. 25% of that would be Z2013, 25% U152, 25% P312*, 25% U106, roughly.

The frequency drops precipitously in the Dinaric region - Herzegovina, Bosnia, especially.

A high frequency in Kosovar Albanians (21%) and Albanians 17% is seen, much of which is likely Z2103. However, some 20-30% is x L23, quite likely that PF7562 described in the ftDNA "ht 35" Project.

Bulgaria also has double-digit frequencies (10%), of which half of probably Z2013, as with Greece (10%, and up to 8% being Z2013).

All in all, much of R1b in the Balkans is probably Z2013 (5-10%), and of the P312 groups, most is U152 (2-6%) - like in next door Italy, followed by a trickle of U106 (2-4%).

I also included Turkey, which had a healthy 15% of R1b, 13% of which is likely Z2103.

For comparisons, I also included some R1b-rich countries - Austria, Northern Italy, Ireland, Germany & Netherlands.

______________________


Its difficult to make any conclusions with such a paucity of aDNA from southeastern and east-central Europe. However, the overall impression one gets is that R1b was probably much higher there during the Iron Age. Its reduction - most felt in the Dinaric hinterland - must have to do with the Slavic migrations following the demographic collapse in Rome's Balkan regions after centuries of making a large bulk of the (East) Empire's soldiery (and Emperors), and having been not so densely populated to begin with. I'd say the same is true for eastern Europe further up, with drops in R1b frequency due to arriving I2a1 and R1a-M458 from Carpatho-Ukraine. I.e. I'd bet R1b had a higher frequency in the Iron Age in places like Czecho-Slovakia and Poland during the Iron Age than now (but in turn it had possibly expanded westward - somewhat diminishing the prevalence of R1a groups which arrived with CWC).

It is difficult to post theories as to the route of M269 migration through Europe based on modern east European M269 frequencies, all the more acute given the dynamic character of the eastern continent - with several migration waves, cultural turnovers, etc, between the EBA and late Antiquity.

Based on this, it is difficult to say which route M269 took to reach western Europe - i.e. south or north of the Carpathians. But given the dearth (? complete absence) of R1b in CWC, south is looking more probable.


Data for frequencies was collated from Myres et al (2011). Additional data for Albanians - Sarno et al (2015), Bulgarians (Karachanuk 2013).

I think you are right that R1b has decreased in eastern and central Europe with Slavic expansion being a major factor. The Tyrol study showed for example L51xL11 just suddenly dies off completely in the areas where Slavic placenames survive but had a peak frequency for Europe in areas without such names. This demonstrates that the Slavic expansion was capable of erasing previous yDNA patterns. The Slavs seem to have expanded by folk movement rather than elite dominance. This contrasts with Germanic who sometimes practices elite dominance such as in Spain and France.

alan
01-23-2016, 10:36 AM
Basically I think eastern and to a lesser extent east-central Europe has hopelessly changed over the last 2000 years or so to use modern population studies to infer past situations and migration routes. So its best to concentrate on ancient DNA. I would be wary of ruling out a northerly route west for R1b because so far all the ancient L11 derived stuff is from northern or north-central Europe. There is also the way that even far western R1b derived people like Rathlin are distinctly northern European. If we didnt have archaeological considerations and just looked at the ancient DNA then a northern route west would probably be being concluded. Even today all northern Europeans cluster together. Certainly a truely southern route along the Med. or southern Alps can be ruled out by ancient DNA IMO. Its pretty emphatic on that point.

Gravetto-Danubian
01-23-2016, 10:43 AM
Basically I think eastern and to a lesser extent east-central Europe has hopelessly changed over the last 2000 years or so to use modern population studies to infer past situations and migration routes. So its best to concentrate on ancient DNA. I would be wary of ruling out a northerly route west for R1b because so far all the ancient L11 derived stuff is from northern or north-central Europe. There is also the way that even far western R1b derived people like Rathlin are distinctly northern European. If we didnt have archaeological considerations and just looked at the ancient DNA then a northern route west would probably be being concluded. Even today all northern Europeans cluster together. Certainly a truely southern route along the Med. or southern Alps can be ruled out by ancient DNA IMO. Its pretty emphatic on that point.

Yes I meant we're not sure at this juncture if R1b went west : north over the Carpathians - eg through Poland , or south of them- through Hungary. Not via the Mediterranean ;)

rms2
01-23-2016, 03:19 PM
I would not be so quick to rule R1b out of Corded Ware. RISE1 from the CW site at Oblaczkowo, Poland, is a bad sample and not a firm result, but it does show L1345+, an R1b SNP. That might be an error because of the bad sample, but it might be real, as well, and Corded Ware was a cultural horizon with a number of regional variants, not an absolute monolith. No ancient y-dna from its western precincts has yet been recovered. I still think U106 may have come west in the vanguard of Corded Ware. Obviously I could be wrong about that.

I think P312 was part of the southwestern branch of Yamnaya, the segment that went around the southern end of the Carpathians and up the Danube, eventually becoming the core of Bell Beaker.

lgmayka
01-23-2016, 04:30 PM
Poland is a different story, however, with 21% frequency. A solid chink is U106, but much (10%) is L23 xM412/L51 stuff. This is probably Z2013 groups ?
As your map correctly shows, your statistics for "Poland" are actually for northwestern Poland. Southeastern Poland has proportions more like Slovakia--a fairly even split among U106, U152, DF27, and Z2103.

razyn
01-23-2016, 05:29 PM
In the bar graph, examples at the far right (N. Italy and Ireland): what's the difference between "other P312" (lilac, at the top) and "P312*" (chocolate, near the bottom), which you have said means x-U152, x-L21. Would "stuff like DF27" not fit either description?

Gravetto-Danubian
01-23-2016, 11:59 PM
In the bar graph, examples at the far right (N. Italy and Ireland): what's the difference between "other P312" (lilac, at the top) and "P312*" (chocolate, near the bottom), which you have said means x-U152, x-L21. Would "stuff like DF27" not fit either description?

Good point, my error - "other P312" shouldn't exist. So I imagine most P312* would be stuff like DF 27, DF 19 & S182 - which was not tested in Myres.
Here's a corrected bar graph.

7430

Gravetto-Danubian
01-24-2016, 12:00 AM
As your map correctly shows, your statistics for "Poland" are actually for northwestern Poland. Southeastern Poland has proportions more like Slovakia--a fairly even split among U106, U152, DF27, and Z2103.

Yes. For Poland, I merely averaged the data from Myres, which had "West Poland", "East Poland", "North Poland", etc.

lgmayka
01-24-2016, 01:30 AM
For Poland, I merely averaged the data from Myres, which had "West Poland", "East Poland", "North Poland", etc.
Then you miscalculated. For example, your bar graph shows absolutely no U152 in Poland, whereas Myres found plenty.

This is Myres' page of supplementary data (http://www.nature.com/ejhg/journal/v19/n1/suppinfo/ejhg2010146s1.html?url=/ejhg/journal/v19/n1/full/ejhg2010146a.html). The actual percentages are in Supplementary Table S4 (http://www.nature.com/ejhg/journal/v19/n1/extref/ejhg2010146x5.xls).


R1a-M420* M343* L11* S116*
Region/Country/Population Regional population code* N R-M207* M479* R2-M124 (xSRY10831,M17) (xM73xM269) V88 M73 M269 all M269(xL23) L23(xM412) M412(xL11) (xU106xS116) U106(xU198) U198 (xM529xU152) M529(xM222) M222 U152
Poland Southwest (Wroclaw) POL 93 0 0 0 0 0 0 0 0.129 0 0.043 0 0 0.032 0 0.022 0.011 0 0.022
Poland East POL 13 0 0 0 0 0 0 0 0.154 0 0.077 0 0 0.077 0 0 0 0 0
Poland North POL 17 0 0 0 0 0 0 0 0.235 0 0.059 0 0.059 0.059 0 0 0 0 0.059
Poland POL 42 0 0 0 0 0 0 0 0.214 0.024 0.095 0 0 0.071 0 0 0.024 0 0
Poland South POL 22 0 0 0 0 0 0 0 0.227 0 0.045 0.045 0 0.091 0 0 0 0 0.045
Poland West POL 15 0 0 0 0 0 0 0 0.333 0 0 0 0 0.133 0 0 0 0 0.2

When I simply total up all the Polish samples, I get:
N = 202
M269 (all) = 37 (18.3%)
M269(xL23) = 1 (0.5%)
L23(xM412) = 11 (5.4%)
M412(xL11) = 1 (0.5%)
L11(xU106,S116) = 1 (0.5%)
U106 = 12 (5.9%)
S116(xM529,U152) = 2 (1.0%)
M529 = 2 (1.0%)
U152 = 7 (3.5%)

The sample counts for individual directions (east, west, north, and south) are low, and do not correspond to historic regions anyway. Perhaps a better division would be northwest, southwest, northeast, southeast, and central.

Gravetto-Danubian
01-24-2016, 01:48 AM
Then you miscalculated. For example, your bar graph shows absolutely no U152 in Poland, whereas Myres found plenty.

This is Myres' page of supplementary data (http://www.nature.com/ejhg/journal/v19/n1/suppinfo/ejhg2010146s1.html?url=/ejhg/journal/v19/n1/full/ejhg2010146a.html). The actual percentages are in Supplementary Table S4 (http://www.nature.com/ejhg/journal/v19/n1/extref/ejhg2010146x5.xls).


R1a-M420* M343* L11* S116*
Region/Country/Population Regional population code* N R-M207* M479* R2-M124 (xSRY10831,M17) (xM73xM269) V88 M73 M269 all M269(xL23) L23(xM412) M412(xL11) (xU106xS116) U106(xU198) U198 (xM529xU152) M529(xM222) M222 U152
Poland Southwest (Wroclaw) POL 93 0 0 0 0 0 0 0 0.129 0 0.043 0 0 0.032 0 0.022 0.011 0 0.022
Poland East POL 13 0 0 0 0 0 0 0 0.154 0 0.077 0 0 0.077 0 0 0 0 0
Poland North POL 17 0 0 0 0 0 0 0 0.235 0 0.059 0 0.059 0.059 0 0 0 0 0.059
Poland POL 42 0 0 0 0 0 0 0 0.214 0.024 0.095 0 0 0.071 0 0 0.024 0 0
Poland South POL 22 0 0 0 0 0 0 0 0.227 0 0.045 0.045 0 0.091 0 0 0 0 0.045
Poland West POL 15 0 0 0 0 0 0 0 0.333 0 0 0 0 0.133 0 0 0 0 0.2

When I simply total up all the Polish samples, I get:
N = 202
M269 (all) = 37 (18.3%)
M269(xL23) = 1 (0.5%)
L23(xM412) = 11 (5.4%)
M412(xL11) = 1 (0.5%)
L11(xU106,S116) = 1 (0.5%)
U106 = 12 (5.9%)
S116(xM529,U152) = 2 (1.0%)
M529 = 2 (1.0%)
U152 = 7 (3.5%)

The sample counts for individual directions (east, west, north, and south) are low, and do not correspond to historic regions anyway. Perhaps a better division would be northwest, southwest, northeast, southeast, and central.

Thanks. Corrected.

7433

I agree it's best to split, eg east Poland would be different to west. But this little exercise did not pretend to be 100 % specific for one specific region - as you say, numbers are very low. More an illustrative exercise - more for my own re-acquaintance - as to what subclades distributions we are seeing in Europe's east, and what we might be looking for in the future.

MitchellSince1893
01-24-2016, 03:43 AM
Thanks. Corrected

Maybe you are in the process of updating it but your chart still doesn't show a red segment for U152 for Poland

Gravetto-Danubian
01-24-2016, 04:18 AM
Darn, i just re-attached the old one. Hope this works..

7436

David Mc
01-24-2016, 06:39 AM
Nicely done, G-D! And very interesting.

Silesian
01-24-2016, 01:45 PM
It is interesting to look at the frequency of R1b in Europe, and the sub-clade proportions

7410

7411

Data for frequencies was collated from Myres et al (2011). Additional data for Albanians - Sarno et al (2015), Bulgarians (Karachanuk 2013).
Nice work. You can parse/define the R1b Eastern clades even finer in your graph, by adding R1b-9219+/cts1450, http://www.yfull.com/tree/R-CTS1450/TMRCA 4400 YBP+/-
found among the following,
https://www.familytreedna.com/groups/jaszsag/dna-results -Jasz_people
https://www.familytreedna.com/groups/albanian-bloodlines/dna-results-Albanian
https://www.familytreedna.com/groups/polish/about/background-Poles

& surprisingly outlier group Ossetians
https://www.familytreedna.com/groups/ossetian/about/background

Gravetto-Danubian
01-24-2016, 07:50 PM
Nice work. You can parse/define the R1b Eastern clades even finer in your graph, by adding R1b-9219+/cts1450, http://www.yfull.com/tree/R-CTS1450/TMRCA 4400 YBP+/-
found among the following,
https://www.familytreedna.com/groups/jaszsag/dna-results -Jasz_people
https://www.familytreedna.com/groups/albanian-bloodlines/dna-results-Albanian
https://www.familytreedna.com/groups/polish/about/background-Poles

& surprisingly outlier group Ossetians
https://www.familytreedna.com/groups/ossetian/about/background

Is CTS1450 the major represented subgrpup of Z2103 in EE ?

Silesian
01-24-2016, 08:18 PM
Is CTS1450 the major represented subgrpup of Z2103 in EE ?

I'm suspect about the placement of z2103/05 so many/all around the 6200 mark, Z2103 & Z2106 &Z2109 & Z2110, but we will just have to be patient and see how everything gets resolved. Smal has done an outstanding job.
http://www.kumbarov.com/ht35/aDNA_02_11_30_2015.pdf

Numerically, that would be a good conclusion, although I'm not sure about Russians (Northern)(Arkhangelsk region) IE (Slavic, East) 114 14.0[6] &Komi Uralic (Finnic) 94 16.0[41] 94. For example I cannot see the mutation in the https://www.familytreedna.com/groups/arab-tribes/dna-results project which has 4060 members and about 30+/- r1b samples, just eyeballing looks like they have more or equal L51+ than Z2103, FWIW.
Once you go away from Poland and Czech and the Ossetians Ossetians IE (Iranian, NE) 47 42.6[25] / in the Northern Caucasus it becomes pretty rare.

https://en.wikipedia.org/wiki/Y-DNA_haplogroups_by_ethnic_group

smal
01-24-2016, 09:03 PM
You can find more correct data for Belarusians here (http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0066499)


In Belarusians, another branch of haplogroup R, haplogroup R1b(M269), makes up around 5% (Figure 6).

Tomenable
02-17-2016, 01:52 AM
Yes. For Poland, I merely averaged the data from Myres, which had "West Poland", "East Poland", "North Poland", etc.

I think you made a mistake because Myres has ~18% not 21% of R1b for Poland:

Poland (n=202):

L51 (= 11,91% of Y-DNA and 64,9% of R1b):

U106(xU198) ----------- 0,0594 (= 5,94%)
U152 ---------------------- 0,0347 (= 3,47%)
S116*(xM529xU152) ---- 0,0101 (= 1,01%)
M529(xM222) ------------ 0,0099 (= 0,99%)
L11*(xU106xS116) ------ 0,005 (= 0,5%)

xL51 (= 6,44% of Y-DNA and 35,1% of R1b):

L23(xM412) -------------- 0,0544 (= 5,44%)
M269(xL23) -------------- 0,005 (= 0,5%)
M412(xL11) -------------- 0,005 (= 0,5%)

M269 all ------------------ 0,1835 (= 18,35%)

====================================

What is interesting, is that there is more U106 in Poland than in Czech Republic:

Czech Rep. (n=87):

U106(xU198) ----------- 0,057 (= 5,7%)
L23(xM412) -------------- 0,057 (= 5,7%)
S116*(xM529xU152) ---- 0,057 (= 5,7%)
U152 ---------------------- 0,034 (= 3,4%)
M529(xM222) ------------ 0,011 (= 1,1%)

R1b-M269 all ------------------ 0,218 (= 21,8%)

Slovakia (n=276):

S116*(xM529xU152) ---- 0,043 (= 4,3%)
U106(xU198) ----------- 0,04 (= 4,0%)
L23(xM412) -------------- 0,036 (= 3,6%)
U152 ---------------------- 0,025 (= 2,5%)
L11*(xU106xS116) ------ 0,007 (= 0,7%)
M412(xL11) -------------- 0,007 (= 0,7%)
M529(xM222) ------------ 0,004 (= 0,4%)

R1b-M269 all ------------------ 0,163 (= 16,3%)
R1b-M343*(xM73xM269) ----- 0,007 (= 0,7%)

Gravetto-Danubian
02-18-2016, 03:26 AM
I think you made a mistake because Myres has ~18% not 21% of R1b for Poland:

Poland (n=202):

L51 (= 11,91% of Y-DNA and 64,9% of R1b):

U106(xU198) ----------- 0,0594 (= 5,94%)
U152 ---------------------- 0,0347 (= 3,47%)
S116*(xM529xU152) ---- 0,0101 (= 1,01%)
M529(xM222) ------------ 0,0099 (= 0,99%)
L11*(xU106xS116) ------ 0,005 (= 0,5%)

xL51 (= 6,44% of Y-DNA and 35,1% of R1b):

L23(xM412) -------------- 0,0544 (= 5,44%)
M269(xL23) -------------- 0,005 (= 0,5%)
M412(xL11) -------------- 0,005 (= 0,5%)

M269 all ------------------ 0,1835 (= 18,35%)

====================================

What is interesting, is that there is more U106 in Poland than in Czech Republic:

Czech Rep. (n=87):

U106(xU198) ----------- 0,057 (= 5,7%)
L23(xM412) -------------- 0,057 (= 5,7%)
S116*(xM529xU152) ---- 0,057 (= 5,7%)
U152 ---------------------- 0,034 (= 3,4%)
M529(xM222) ------------ 0,011 (= 1,1%)

R1b-M269 all ------------------ 0,218 (= 21,8%)

Slovakia (n=276):

S116*(xM529xU152) ---- 0,043 (= 4,3%)
U106(xU198) ----------- 0,04 (= 4,0%)
L23(xM412) -------------- 0,036 (= 3,6%)
U152 ---------------------- 0,025 (= 2,5%)
L11*(xU106xS116) ------ 0,007 (= 0,7%)
M412(xL11) -------------- 0,007 (= 0,7%)
M529(xM222) ------------ 0,004 (= 0,4%)

R1b-M269 all ------------------ 0,163 (= 16,3%)
R1b-M343*(xM73xM269) ----- 0,007 (= 0,7%)

The Myres supp. data had 5 Polish sub-categories - one titled 'Poland', and 4 others west, east, north & south. I think I just averaged out the latter 4, so as not to duplicate
Whatever the case, 18 vs 21% changes the overall picture very little.

wombatofthenorth
04-13-2016, 05:25 AM
Not that it means much since it's so far only traced back to about 1826 but my strictly paternal line is R-L20 in Latvia to an apparent Latvian serf born around 1826.