G25 Balkan Calculator

[Rizza]

I didnt say they had a lot of ev13. Maybe they didnt. But none can say they didnt have any when my clade shows clear proof it was Illyrian for 8000 years. Not Thracian, not Roman.

We have several EV-13 ILLYRIAN, one from Iron Age Croatia and several from Roman period and one from medieval

This is another profile of EV-13 found in Naissus

Distance to: Roman_NaissusR6764-K13-sim
0.02213164 MKD_Anc
0.02641717 ITA_Rome_Late_Antiquity
0.02676969 ALB_Mdv
0.02707730 HRV_Pop_CA
0.02861375 ALB_Cinamak_Anc
0.02971800 HRV_BA
0.02987156 ITA_Rome_MA
0.02987771 MKD_BA
0.03030494 ITA_PoggioPelliccia_EMA
0.03039972 SRB_Mokrin_EBA_Maros_oAegean
0.03054305 ALB_PostMdv
0.03096004 DEU_MA_Alemannic_o2
0.03207600 HUN_IA_Syrmian_SremGroup
0.03212464 ITA_Daunian
0.03249035 ITA_Chiusi_EMA
0.03264642 MNE_LBA
0.03441901 ITA_Tarquinia_EMA
0.03443172 HUN_EIA
0.03488768 BGR_TellKran_EBA
0.03497976 MDA_Trypillia_Late
0.03500374 BGR_EBA
0.03528632 GRC_Logkas_MBA
0.03580020 HRV_EBA
0.03605799 ITA_Etruscan
0.03607031 HUN_earlymiddle_Avar

Target: Roman_NaissusR6764-K13-sim
Distance: 1.7677% / 0.01767674
65.4 TUR_Barcin_N
24.2 Yamnaya_RUS_Samara
5.6 SRB_Iron_Gates_HG
4.4 GEO_CHG
0.4 MAR_Taforalt


Like around 40% of Albanian Y-DNA has been found in Western Balkans if we only take into account J2b2 and R1b which her population seems to totally lack. This is not taking into account maternal side. Illyrians also inhabited the central Balkans.


Iron Age and Bronze Age Croatia is filled with J2B2-L283 while seems non existant among her population. I think this troll realized this so she lashes it out on Albanians and takes things out of context. I would ignore this troll from now, incredibly incompetent.

[Rizza]

I think you need to look into a dictionary what defended means if you think this person has defended Albanians. I have seen foreigners who are far more pro Albanian and have far more knowledge about Albanian history . This person is basically just butthurt that Albanians are one of the last survivors of an indigenous Ancient Balkan/Illyrian population. Just some nationalist who is trying to steer us away from our Illyrian ancestors. I have seen enough of this persons posts. Complains about the calc in the OP yet uses near eastern samples to model other people and claims people are 30% near east without understanding overlapping components as explaiend by this Albanian guy who created the calc: https://anthrogenica.com/showthread....l=1#post876427

The guy who claims Albanians and Balkan people are like 30%-40% near east and uses samples from Croatia wrote this:

Originally Posted by bce
I never said those influences were very high in Albanians. I was just responding to Bruzmi who used certain models and PCAs to prove that modern Albanians are 100% identical to Cinamak.

Code:

Croatia_Imperial,0.1066143,0.1411583,-0.0294153,-0.054587,0.0001023,-0.0188717,-0.006267,-0.0050767,-0.0111803,0.0144573,0.005088,0.005545,-0.0006937,-0.0052297,-0.0113553,0.0031823,0.0180363,-0.0019003,0.0023463,-0.0042937,-0.0059477,-0.0001237,-0.0042313,0.0013253,-0.0030737
ALB_Çinamak_Anc,0.123904,0.153200,0.028823,-0.012459,0.029148,-0.005379,0.001175,0.000396,-0.000292,0.024107,0.002598,0.006573,-0.018307,-0.007648,-0.009190,-0.002462,0.010989,0.001412,0.004741,-0.008665,-0.005633,0.001767,-0.003363,0.006593,-0.005474
TUR_Marmara_Ilıpınar_Byz3(Slav),0.130897,0.133034,0.065242,0.037791,0.043085,0.008646,0.008225,0.019615,-0.005931,-0.016401,0.000650,-0.007343,0.008771,0.019955,-0.008143,0.003447,0.009257,-0.004434,-0.004651,-0.002626,-0.007736,-0.005193,0.011216,-0.007712,-0.007664

The Albanian guys response:

Originally posted by Bruzmi
You are replying as if I said that Albanians have 0% post-IA admixture which is a statement that nobody can claim for almost all groups anywhere in the world. In a reply to you yesterday I repeated the same view:

Originally Posted by Bruzmi
I don't understand why you're replying as if someone claimed that there is 0% post-IA admixture. At the same time, whichever custom pca we use, we can see that all of these admixtures are low and certainly aren't what earlier claims made them to be.
A key rule in good conversations is to reply to what the other side is saying, not to arguments which don't exist in the conversation.

Now, I'll examine your model. The first issue that I have that as you created a three-way model with non-contemporaneous groups, it's certain that it'll cause overlapping...which it does, but it also doesn't provide a very good fit and I'll tell you why it can't do so.

Target: ALB_Çinamak_Anc
Distance: 1.9872% / 0.01987198
60.4 TUR_Barcin_N
35.8 Yamnaya_RUS_Samara
3.8 SRB_Iron_Gates_HG

Target: Albanian
Distance: 2.4404% / 0.02440353
58.4 TUR_Barcin_N
33.4 Yamnaya_RUS_Samara
3.2 SRB_Iron_Gates_HG
2.6 IRN_Ganj_Dareh_N
1.0 Israel_Natufian
0.8 GEO_CHG
0.6 Han

Target: Croatia_Imperial
Distance: 2.2931% / 0.02293063
52.8 TUR_Barcin_N
15.0 IRN_Ganj_Dareh_N
13.6 Yamnaya_RUS_Samara
10.4 GEO_CHG
8.2 Israel_Natufian

Target: Slavic:TUR_Marmara_Ilipinar_Byz3 (the key point is low ANF, not the lack of a proper base source for the Baltic drift)
Distance: 5.7416% / 0.05741599
43.0 Yamnaya_RUS_Samara
39.2 TUR_Barcin_N
17.8 SRB_Iron_Gates_HG

Your model doesn't have a very good fit because all the post-Çinamak sources have lower ANF than the Albanian average on G25. You can see the overlapping in the 30% "Croatia_Imperial" component which doesn't correspond to the components which Albanians should have if they had ~30% ancestry from this source.

The IA ancestors of Albanians didn't have contacts with East Med-like and Slavic-like groups at the same time, so models of this kind should be just two-way ones: pre-medieval + Migration Period. The second issue is that we shouldn't pick arbitrarily which samples fit the contacts of Albanians with East Med-like groups. We have samples which show that contact (e.g. Alb_Mdv), hence we don't have to calculate it ourselves. Alb_Mdv samples look nothing like the Croatia_Imperial average which you created hence we know for a fact that East Med-like admixture in Roman Albania didn't look at all like what you're describing:

Target: ALB_Mdv
Distance: 2.2292% / 0.02229232
63.6 TUR_Barcin_N
31.0 Yamnaya_RUS_Samara
2.0 GEO_CHG
2.0 IRN_Ganj_Dareh_N
1.4 SRB_Iron_Gates_HG

On the other hand, we can't know the extent of this admixture which for many individuals might have been negligible, so models should work with a range which assumes an admixture rate from 0% to the highest per sampling from that region. In this case, if you want to use samples from Croatia but not the Trogir dataset, one of the samples closer to Alb_Mdv in terms of local + East Med-like admixture is:

Target: Croatia_Novo_selo_Bunje:R3547___AD_571___Coverage_ 57.30%
Distance: 2.9260% / 0.02925961
55.6 TUR_Barcin_N
27.0 Yamnaya_RUS_Samara
8.8 GEO_CHG
6.2 Israel_Natufian
1.0 SRB_Iron_Gates_HG
0.8 IRN_Shahr_I_Sokhta_BA2_I8728
0.4 MAR_Taforalt
0.2 Han

But you have to add it separately as part of a range which assumes 0% (MNE_LBA/HRV_EIA) to R2041-like admixture. This means the creation of a model which includes:
1. ALB sources from 0% to Alb_Mdv-like admixture, hence Alb_Çinamak (0%) + Alb_Mdv
2. Dalmatia/Montenegro sources 0% to R3547-like admixture
3. a Migration Period/Slavic source

Quote Originally Posted by bce View Post
this is the mathematically feasible amount of those influences in Albanians. It means all proportions fit, ANF included.

Based on the above:

Code:

Albania_IA_Rom:ALB_Cinamak_Anc,0.1239044,0.1531999,0.0288229,-0.0124586,0.0291481,-0.0053786,0.001175,0.0003956,-0.000292,0.0241073,0.0025981,0.0065729,-0.0183066,-0.0076479,-0.0091901,-0.0024623,0.0109894,0.0014116,0.0047407,-0.0086647,-0.005633,0.0017666,-0.0033629,0.006593,-0.0054743
Albania_IA_Rom:ALB_Mdv,0.120273,0.1502983,0.0158387,-0.0279933,0.0196957,-0.007251,-0.00141,-0.0031537,0.0025907,0.0223547,0.000487,0.0052957,-0.014519,9.2e-05,-0.0152457,-0.0100767,0.0092573,0.000718,0.0085057,-0.0132563,-0.0108557,0.005276,-0.0013557,-0.0038957,-0.003393
HRV/MNE_LBA/IA_Rom:HRV_EIA,0.1276082,0.1472518,0.0320973,-0.0011843,0.0354596,-0.0009297,0.0016711,0.002359,0.0107942,0.0275177,-0.0005954,0.008992,-0.0170796,-0.0077679,-0.0019151,-0.0021068,0.003868,0.0031109,0.0071369,-0.0045022,-0.0027036,0.0016899,0.0006983,5.34e-05,-0.0025813
HRV/MNE_LBA/IA_Rom:Croatia_Novo_selo_Bunje:R3547___AD_571___Coverage_57.30%,0.112685,0.144205,0.002263,-0.031654,0.021542,-0.010877,0.00517,-0.002538,0.001636,0.01877,0,0.006145,-0.00996,0.004817,-0.014386,-0.012463,-0.004303,-0.005954,-0.003394,-0.000875,-0.011729,0.000866,0.002218,-0.000482,0.000239
HRV/MNE_LBA/IA_Rom:MNE_LBA,0.1270678,0.1491904,0.0278726,-0.0069885,0.0273337,-0.00573,-0.0005768,0.0010698,0.0039603,0.0292077,0.0040745,0.0097005,-0.01342,-0.006781,-0.0104628,3.6e-05,0.006377,0.0004607,0.0056794,-0.0089245,-0.0069195,0.0050474,-0.0004818,0.0054114,-0.0012845
Migration_Period:TUR_Marmara_Ilipinar_Byz3,0.130897,0.133034,0.065242,0.037791,0.043085,0.008646,0.008225,0.019615,-0.005931,-0.016401,0.00065,-0.007343,0.008771,0.019955,-0.008143,0.003447,0.009257,-0.004434,-0.004651,-0.002626,-0.007736,-0.005193,0.011216,-0.007712,-0.007664

Another close sample is:

Target: Croatia/Montenegro_LBA/IA_Rom:Croatia_Sisak_Pogorelec:R2041___AD_303___Co verage_50.40%
Distance: 1.6494% / 0.01649370
54.4 TUR_Barcin_N
29.4 Yamnaya_RUS_Samara
4.8 SRB_Iron_Gates_HG
4.6 GEO_CHG
3.4 IRN_Ganj_Dareh_N
3.4 Israel_Natufian

but we only have g25 sims for it. Just for comparison purposes, I created a separate model which replaces R3547 with R2041

Code:

Albania_IA_Rom:ALB_Cinamak_Anc,0.1239044,0.1531999,0.0288229,-0.0124586,0.0291481,-0.0053786,0.001175,0.0003956,-0.000292,0.0241073,0.0025981,0.0065729,-0.0183066,-0.0076479,-0.0091901,-0.0024623,0.0109894,0.0014116,0.0047407,-0.0086647,-0.005633,0.0017666,-0.0033629,0.006593,-0.0054743
Albania_IA_Rom:ALB_Mdv,0.120273,0.1502983,0.0158387,-0.0279933,0.0196957,-0.007251,-0.00141,-0.0031537,0.0025907,0.0223547,0.000487,0.0052957,-0.014519,9.2e-05,-0.0152457,-0.0100767,0.0092573,0.000718,0.0085057,-0.0132563,-0.0108557,0.005276,-0.0013557,-0.0038957,-0.003393
HRV/MNE_LBA/IA_Rom:HRV_EIA,0.1276082,0.1472518,0.0320973,-0.0011843,0.0354596,-0.0009297,0.0016711,0.002359,0.0107942,0.0275177,-0.0005954,0.008992,-0.0170796,-0.0077679,-0.0019151,-0.0021068,0.003868,0.0031109,0.0071369,-0.0045022,-0.0027036,0.0016899,0.0006983,5.34e-05,-0.0025813
HRV/MNE_LBA/IA_Rom:Croatia_Sisak_Pogorelec:R2041___AD_303___Coverage_50.40%,0.1201,0.1444,0.0077,-0.0179,0.0246,-0.0079,-0.0002,0.0011,0.0026,0.013,0.0015,0.0047,-0.0102,0.0001,-0.0079,0.0017,0.0101,0,0.0013,-0.0004,-0.0028,0.0013,-0.0018,-0.001,-0.0017
HRV/MNE_IA_Rom:MNE_LBA,0.1270678,0.1491904,0.0278726,-0.0069885,0.0273337,-0.00573,-0.0005768,0.0010698,0.0039603,0.0292077,0.0040745,0.0097005,-0.01342,-0.006781,-0.0104628,3.6e-05,0.006377,0.0004607,0.0056794,-0.0089245,-0.0069195,0.0050474,-0.0004818,0.0054114,-0.0012845
Migration_Period:TUR_Marmara_Ilipinar_Byz3,0.130897,0.133034,0.065242,0.037791,0.043085,0.008646,0.008225,0.019615,-0.005931,-0.016401,0.00065,-0.007343,0.008771,0.019955,-0.008143,0.003447,0.009257,-0.004434,-0.004651,-0.002626,-0.007736,-0.005193,0.011216,-0.007712,-0.007664

Both are better fits than your model and they correspond to base components in a much better way. Albanian_G25 is the existing Albanian average on G25 and Alb_new_G25 is an average I created with just the official Albanian G25 coords which I've gathered.

Code:

Albanian_new_G25,0.1250916,0.1457284,0.0166129,-0.0182818,0.0257279,-0.0083667,0.003243,0.0013614,-0.0001021,0.0190802,-0.0007146,0.0044958,-0.0071359,0.0063443,-0.0201952,-0.0018031,0.0129469,0.0003802,0.0066997,-0.0073536,-0.0101322,0.0004452,0.0017624,0.0034221,-0.0034847

Judging from the G25 sims that we have already, when/if we get G25 coords from samples like the J-L283 from Viminacium or the E-V13 from Naissus as a replacement for Dalmatia/Montenegro, the Migration Period admixture would drop even further. Such figures are consistent with the conclusions from the Southern Arc papers.

PS I don't want to start another discussion about this subject. You posted some models and I showed how models with better fits can show different results. In the past, I've posted models with even better fits in the same direction but I want to move to other topics. The lesson to learn is that Migration Period admixture is much lower than what some people assume without testing all possibilities.

EDIT:

PS.2. It also doesn't lower Migration Period admixture in terms of a "model effect", as you can see for Serbs, for whom it reproduces the Lalueza-Fox (2022) conclusion (50-60% Slavic/40-50% Balkan) without increasing the pre-Slavic component:

Target: Serbian
Distance: 1.3535% / 0.01353505
56.0 Migration_Period
38.2 HRV/MNE_LBA/IA_Rom
5.8 Albania_IA_Rom

Near Eastern + Slavic component will get dropped usually anyway , although it is there and has shifted people a bit more east maybe I believe but not that huge, nothing compared to some Italians, Greeks, Romanians, Macedonians, Bulgarians etc.

[Ylla]

that’s my take on it aswell. The calculators are already limited and biased as they depend on data in the source. They can only tell you equivalence. There could be multiple sources of east med input and nobody seems to consider the huge Greek settlement in Italy and across the Balkans. Roman settlement is too limited of an explanation for my liking.
I probably do have excess Roman admix most likely eastern Christians, but majority of Albanians have too low Near Eastern admixture to account for that. Also just because the E-V13 Croatian sample has left no descendants doesn’t mean there weren’t clades that have. Isn’t it common sense to think the fact that one sample has been identified is more than enough evidence that Illyrians we’re carriers of EV13 although as a minority clade.

[Rizza]

that’s my take on it aswell. The calculators are already limited and biased as they depend on data in the source. They can only tell you equivalence. There could be multiple sources of east med input and nobody seems to consider the huge Greek settlement in Italy and across the Balkans. Roman settlement is too limited of an explanation for my liking.
I probably do have excess Roman admix most likely eastern Christians, but majority of Albanians have too low Near Eastern admixture to account for that. Also just because the E-V13 Croatian sample has left no descendants doesn’t mean there weren’t clades that have. Isn’t it common sense to think the fact that one sample has been identified is more than enough evidence that Illyrians we’re carriers of EV13 although as a minority clade.

I don't score any Near East

Target: Rizza
Distance: 2.1362% / 0.02136161
55.6 TUR_Barcin_N
34.4 Yamnaya_RUS_Samara
5.2 WHG
4.8 Kura-Araxes_ARM_Kaps



Anyway here is a PCA Map an Albanian guy made

[Rizza]

............

[Rizza]

These Illyrians from Northern Albania belong to J2b2-L283 and R-PF7563>Z29758 one of the most major and diverse Albanian lineages https://www.yfull.com/tree/R-Z29758/ ,

I used that Sisak sample and modelled it with Ancient samples and many others, the components these samples score causes overlap.

[Rizza]

I was told that it's common to not cluster exactly 100% with Iron Age or Bronze Age samples. Apparently Basque are one of the few.

[Rizza]

I def don't think Albanians have much Near Eastern ancestry. I cluster very West right next to these samples. Every single Balkan ethnicity clusters more east than Albanians including non-Balkanites like Southern Italians and even some central Italians who according to a paper have more Near Eastern ancestry from Roman period. And many of these mixed samples from the Balkans are locals that mixed with Near Easterns apparently. If you look at the map I posted some of those Iron Age Sampels from North Macedonia and Cetina Bronze and even some Cinamaks literally touch the Albanian cluster. So we are most definitely not talking about ''different clines'' here once we add individuals rather than averages.

And on that same map even some medieval samples are clustering like some moderns. Also on another map I made post medieval samples where touching Ancient Macedonia cluster.The change is overall small if anything for a place like the Balkans that has been a crossroad of so many different people. We have to wait for more samples. Some people are shifted more East-North East maybe due to Slavic component overall + some Imperial but for a lot of others such ancestry is very negligible. Depends also how East-West or South-North your pre-Slavic ancestors were.


EV-13 I def don't believe is a Thacian marker alone nor spread supposedly with East Med, it's an Y-DNA widespread across Europe and much more numerous than J2b2-L283.

[alexmegas777]

Distance to: alexmegas(Romanian)_scaled
0.04042761 HRV/MNE_LBA/IA_Rom:HRV_EIA
0.04164363 HRV/MNE_IA_Rom:MNE_LBA
0.04663844 HRV/MNE_LBA/IA_Rom:Croatia_Sisak_Pogorelec:R2041___AD_303___Co verage_50.40%
0.04669961 Albania_IA_Rom:ALB_Cinamak_Anc
0.05737017 Albania_IA_Rom:ALB_Mdv
0.06741847 Migration_Period:TUR_Marmara_Ilipinar_Byz3

Target: alexmegas(Romanian)_scaled
Distance: 2.4454% / 0.02445431
33.8 HRV/MNE_LBA/IA_Rom
33.6 Migration_Period
32.6 HRV/MNE_IA_Rom

[vasiliki23]

Distance to: VASILIKI_scaled
0.03163724 Albania_IA_Rom:ALB_Mdv
0.03475496 HRV/MNE_LBA/IA_Rom:MNE_LBA
0.03614463 Albania_IA_Rom:ALB_Cinamak_Anc
0.03847839 HRV/MNE_LBA/IA_Rom:HRV_EIA
0.04090221 HRV/MNE_LBA/IA_Rom:Croatia_Novo_selo_Bunje:R3547___AD_571___Co verage_57.30%
0.09411179 Migration_Period:TUR_Marmara_Ilipinar_Byz3

Target: VASILIKI_scaled
Distance: 2.3305% / 0.02330466
45.4 Albania_IA_Rom
42.0 HRV/MNE_LBA/IA_Rom
12.6 Migration_Period