New attempt!!!G25 Balkan calculator with 5 (+1) ancient components

[gixajo]

Do and say what you want.:wink

SCALED

https://vahaduo.github.io/vahaduo/


Balkan:BGR_IA,0.126344,0.157407,0.007165,-0.051034,0.029852,-0.02259,0.00423,-0.001385,0.007772,0.032074,-0.001461,0.005695,-0.013974,-0.003028,-0.015065,-0.001458,0.01369,0.006968,0.006285,-0.014757,-0.006239,0.004699,-0.007888,0.003494,-0.008861
German:DEU_Halberstadt_LBA,0.12862,0.132019,0.0584 54,0.054587,0.032006,0.023985,0.00752,0.005077,0.0 04704,-0.005103,0.002111,0.002248,-0.014271,-0.015001,0.015744,0.017369,0.007562,-0.003801,0.000628,0.004752,0.00262,0.002968,0.0035 74,0.00723,0.002754
Slav:UKR_Srubnaya_MLBA,0.119514,0.12491,0.054682,0 .077843,0.015387,0.027052,-0.00141,-0.009,-0.014521,-0.022415,-0.017538,0.004796,-0.001784,-0.007019,0.020358,0.013657,-0.002868,0.003294,-0.005405,-0.00963,-0.005366,-0.002349,0.022678,0.013134,-0.011496
Turkic:TUR_Kaman-Kalehoyuk_MLBA,0.1050018,0.1515678,-0.042332,-0.082365,-0.0040775,-0.0274705,-0.0024088,-0.0077882,-0.011402,0.028429,0.0097435,0.007006,-0.0120788,0.0030965,-0.0138435,-0.004475,0.0116693,-0.0021538,0.0087988,-0.00741,-0.0031817,0.0061828,-0.0048065,0.0030725,-0.001407
Levant:Levant_Abel_IA,0.087644,0.149283,-0.05506,-0.100776,-0.014156,-0.032072,-0.002115,-0.006231,0.012476,0.007654,0.011205,-0.003297,0.01888,0.003716,0.002443,0.010475,0.0033 9,0.008235,-0.001006,-0.003252,0.005366,0.009027,-0.002835,0.000482,-0.004071

And a Baltic component if someone want to add it.(It will steal % from others ,specially German and Slav)


Baltic:Baltic_EST_MA,0.129379,0.118817,0.0886237,0 .0915167,0.046778,0.0363487,0.013474,0.0148457,0.0 040223,-0.0346857,-0.0089853,-0.0133377,0.0173437,0.02624,-0.003031,0.0135683,0.0149073,-0.0040963,0.0007963,-8.33e-05,0.0007073,0.001319,0.00493,0.0008433,-0.0006783

[Ion Basescul]

MLBA Slavs is something new for sure.

[gixajo]

In my case, it´s the same if I add or I don´t add baltic sample:

Target: gixajo_scaled
Distance: 4.6031% / 0.04603136
50.8 German
49.2 Balkan

Target: gixajo_scaled
Distance: 4.6031% / 0.04603136 | ADC: 0.25x
50.8 German
49.2 Balkan

And if I delete German sample and I add Baltic sample:

Target: gixajo_scaled
Distance: 5.4513% / 0.05451279
63.8 Balkan
19.8 Baltic
16.4 Slav

Target: gixajo_scaled
Distance: 5.4558% / 0.05455840 | ADC: 0.25x
64.4 Balkan
17.8 Baltic
17.8 Slav

So do what you like, but tell in your post which samples you use in sources, please.:thumb001:

[gixajo]

MLBA Slavs is something new for sure.

And if the sample is from Africa Paleolithic and fits well with nowadays Ukranians, Russians and Polishs, I will ad too.:thumb001:

I am very transgressor and and i am very crazy. :crazy:

[Ion Basescul]

And if the sample is from Africa Paleolithic and fits well with nowadays Ukranians, Russians and Polishs, I will ad too.:thumb001:

I am very transgressor and and i am very crazy. :crazy:

Oh I see, so you are trying to find ancient equivalents for moderns?
In that case BGR_IA is too Southern for Balkan, but I guess it also depends on where you draw the boundary for the Balkans. If it's in Albania and Greece then that's fairly ok, but if it extends further North then something like HRV_IA fits better.

[Chris596]

I will come back here with my real G25 coordinates ;) (thanks anyway for trying to create one, but I get Romani with that)

[Grace O'Malley]

Target: Grace_scaled
Distance: 3.3792% / 0.03379216
95.8 German
2.6 Slav
1.6 Balkan

No difference with the Baltic added.

[Hapanuwa]

Target: Hapanuwa_scaled
Distance: 7.2784% / 0.07278388 | ADC: 0.5x
60.0 Turkic
29.0 Slav
11.0 Levant

Target: Hapanuwa_scaled
Distance: 7.1679% / 0.07167918 | ADC: 0.25x
45.4 Turkic
31.0 Slav
22.2 Levant
1.4 Baltic

Target: Hapanuwa_scaled
Distance: 7.1538% / 0.07153801
40.6 Turkic
28.2 Slav
26.2 Levant
5.0 Baltic

[gixajo]

Oh I see, so you are trying to find ancient equivalents for moderns?
In that case BGR_IA is too Southern for Balkan, but I guess it also depends on where you draw the boundary for the Balkans. If it's in Albania and Greece then that's fairly ok, but if it extends further North then something like HRV_IA fits better.

Ok, so I will check your proposal and maybe I add in the future in another calculator.

My idea is to find several good samples from every group and make averages with them ,Usual samples are clear, but I need more not so usually used samples to do it.

And the sample I used has been already used by others users in Balkan calculators and I have not seen criticism about that sample, but...as we used to be here "nunca llueve a gusto de todos", "it never rains to everyone's liking".

[gixajo]

Target: Grace_scaled
Distance: 3.3792% / 0.03379216
95.8 German
2.6 Slav
1.6 Balkan

No difference with the Baltic added.

And removing German and adding Baltic?

[gixajo]

I will come back here with my real G25 coordinates ;) (thanks anyway for trying to create one, but I get Romani with that)

Is Davidsky already doing G25 coordinates? Did you send an email to him?

[Coastal Elite]

Target: DacoCeltic_scaled
Distance: 2.0270% / 0.02026963
44.4 German
34.4 Balkan
21.2 Slav

Target: DacoCeltic_scaled
Distance: 2.0530% / 0.02052994 | ADC: 0.25x
51.2 German
32.6 Balkan
16.2 Slav

Distance to: DacoCeltic_scaled
0.05086350 German: DEU_Halberstadt_LBA
0.08120968 Slav:UKR_Srubnaya_MLBA
0.10079358 Balkan:BGR_IA
0.15428957 Turkic:TUR_Kaman-Kalehoyuk_MLBA
0.18209304 Levant:Levant_Abel_IA

With Baltic

Target: DacoCeltic_scaled
Distance: 1.9977% / 0.01997744
39.8 German
35.6 Balkan
19.8 Slav
4.8 Baltic

Target: DacoCeltic_scaled
Distance: 2.0530% / 0.02052994 | ADC: 0.25x
51.2 German
32.6 Balkan
16.2 Slav

[gixajo]

Poll added.:rolleyes:

[JohnnyP]

You can add also HRV_IA for Balkans.
My results with HRV_IA plus ,because im closer to HRV_IA than BGR_IA.

Target: me
Distance: 2.0495% / 0.02049548 | ADC: 0.25x
42.4 Balkan:HRV_IA
32.2 Baltic:Baltic_EST_MA
24.4 Balkan:BGR_IA
1.0 Slav:UKR_Srubnaya_MLBA

[Grace O'Malley]

And removing German and adding Baltic?

Target: Grace_scaled
Distance: 5.1197% / 0.05119742
47.2 Slav
28.8 Baltic
24.0 Balkan

Higher distance of course.

[Luke35]

Target: MrG_scaled
Distance: 2.3664% / 0.02366386
53.0 German
24.8 Balkan
22.2 Slav

Target: MrG_scaled
Distance: 2.3922% / 0.02392158 | ADC: 0.25x
60.2 German
22.6 Balkan
17.2 Slav


With Baltic added:

Target: MrG_scaled
Distance: 1.9434% / 0.01943399
36.2 German
26.6 Balkan
19.4 Baltic
15.4 Slav
2.4 Turkic

Target: MrG_scaled
Distance: 2.0270% / 0.02027024 | ADC: 0.25x
48.4 German
25.8 Balkan
13.0 Baltic
12.8 Slav

These are all very reasonable results for me, Gixajo. Nice work, thank you.

[Jana]

Target: Feiichy_scaled
Distance: 4.5132% / 0.04513188

66.0 German
25.8 Balkan
8.2 Slav

Target: Feiichy_scaled
Distance: 4.5266% / 0.04526561 | ADC: 0.25x

73.2 German
23.8 Balkan
3.0 Slav

Target: Stears_scaled
Distance: 3.9027% / 0.03902658

47.4 German
29.4 Slav
16.6 Balkan
5.4 Levant
1.2 Turkic

Target: Stears_scaled
Distance: 3.9201% / 0.03920139 | ADC: 0.25x

52.6 German
25.8 Slav
16.6 Balkan
4.0 Levant
1.0 Turkic

[Jana]

with Baltic

Target: Feiichy_scaled
Distance: 2.5487% / 0.02548691

51.8 Baltic
33.8 Balkan
9.2 German
5.2 Turkic

Target: Feiichy_scaled
Distance: 2.6131% / 0.02613086 | ADC: 0.25x

45.0 Baltic
36.6 Balkan
18.4 German

Target: Stears_scaled
Distance: 2.0768% / 0.02076771

49.6 Baltic
29.2 Turkic
14.0 German
7.0 Slav
0.2 Balkan

Target: Stears_scaled
Distance: 2.1883% / 0.02188337 | ADC: 0.25x

42.0 Baltic
25.6 German
23.2 Turkic
5.8 Slav
3.4 Balkan

[gixajo]

You can add also HRV_IA for Balkans.
My results with HRV_IA plus ,because im closer to HRV_IA than BGR_IA.

Target: me
Distance: 2.0495% / 0.02049548 | ADC: 0.25x
42.4 Balkan:HRV_IA
32.2 Baltic:Baltic_EST_MA
24.4 Balkan:BGR_IA
1.0 Slav:UKR_Srubnaya_MLBA

But you score only 1% Slav with both Balkans samples, are you only 1% slav actually? Where are the results with german sample?...

[JohnnyP]

But you score only 1% Slav with both Balkans samples, are you only 1% slav actually? Where are the results with german sample?...

Idk , if i remove Baltic i will score Slavic.
This is without Baltic but fit is 4
Distance: 4.0490% / 0.04049049 | ADC: 0.25x
46.8 Balkan:BGR_IA
23.8 Slav:UKR_Srubnaya_MLBA
20.2 German:DEU_Halberstadt_LBA
9.2 Balkan:HRV_IA

[kleenex]

Distance: 3.4565% / 0.03456509
80.2 Balkan
19.8 Slav

[17571imre]

Target: mh_scaled
Distance: 3.1392% / 0.03139165
48.0 German
34.0 Balkan
18.0 Slav

Baltic added:
Target: mh_scaled
Distance: 1.8932% / 0.01893245
40.8 Balkan
35.2 Baltic
15.6 German
6.6 Slav
1.8 Turkic

[Luke35]

Target: mh_scaled
Distance: 3.1392% / 0.03139165
48.0 German
34.0 Balkan
18.0 Slav

Baltic added:
Target: mh_scaled
Distance: 1.8932% / 0.01893245
40.8 Balkan
35.2 Baltic
15.6 German
6.6 Slav
1.8 Turkic

Any idea why you are scoring this much German, when the Baltic sample is not present? It's especially interesting because I still score in my normal range, but you, Fiiechy and Stears all get higher than normal German.

[17571imre]

Any idea why you are scoring this much German, when the Baltic sample is not present? It's especially interesting because I still score in my normal range, but you, Fiiechy and Stears all get higher than normal German.

yes i saw this too but i have no idea maybe Fiiechy or Kaspias know why this happens?

[17571imre]

Any idea why you are scoring this much German, when the Baltic sample is not present? It's especially interesting because I still score in my normal range, but you, Fiiechy and Stears all get higher than normal German.

double

[Kaspias]

yes i saw this too but i have no idea maybe Fiiechy or Kaspias know why this happens?

Because reference of Slav component which you supposed to get is not Slavic. German component is more Slavic than it. So the algorithm prefers it.

Turkic is also not Turkic but Levantine like native Anatolian.

[celticdragongod]

Distance to: CDG_scaled
0.03649073 German:DEU_Halberstadt_LBA
0.06863219 Slav:UKR_Srubnaya_MLBA
0.09579517 Baltic:Baltic_EST_MA
0.14036009 Balkan:BGR_IA
0.19402032 Turkic:TUR_Kaman-Kalehoyuk_MLBA
0.21790152 Levant:Levant_Abel_IA

Target: CDG_scaled
Distance: 3.6491% / 0.03649073 | ADC: 2x
100.0 German

Target: CDG_scaled
Distance: 3.6491% / 0.03649073 | ADC: 1x
100.0 German

Target: CDG_scaled
Distance: 3.6381% / 0.03638082 | ADC: 0.5x
99.6 German
0.4 Balkan

Target: CDG_scaled
Distance: 3.5531% / 0.03553120 | ADC: 0.25x
90.8 German
4.8 Slav
4.4 Balkan

Target: CDG_scaled
Distance: 3.5411% / 0.03541111
84.8 German
8.6 Slav
6.4 Balkan
0.2 Baltic

[RyoHazuki]

Not sure what advice I can give since I'm not Eastern European, but I think you're on the right track

Distance: 5.3191% / 0.05319131
64.2 German
18.4 Balkan
15.6 Slav
1.8 Baltic

[Impaler]

Target: Impaler_scaled
Distance: 3.5360% / 0.03536028 | ADC: 0.25x
39.4 Turkic
23.6 Slav
15.8 Baltic
15.4 German
5.8 Levant

[gixajo]

Because reference of Slav component which you supposed to get is not Slavic. German component is more Slavic than it. So the algorithm prefers it.

Turkic is also not Turkic but Levantine like native Anatolian.

Fantastic, an expert about Balkan ancient components.

Whic are best samples for everyone?

[Ion Basescul]

East Euro admix bleeds into NW Euro again for me.

https://i.imgur.com/D6gTkLL.png

https://i.imgur.com/nhL7pWw.png

+Baltic
https://i.imgur.com/gaXIb3W.png

https://i.imgur.com/cCbBQ5O.png

[Kaspias]

Fantastic, an expert about Balkan ancient components.

Whic are best samples for everyone?

For everyone?

Scythian_MDA is pretty much in between of BGR_IA and HRV_IA, should work for all while determining Balkanic ancestry.
For Slav it is Avar Szolad.
For Turkic DA89 is proxy of Oghuz and Rus_Nomad is proxy of Steppe Turks(Cumans, Bulgars and so on)

You don't need any other component but if you want to add Germanic, Deu_ma should work.

[Gallop]

Target: Gallop_scaled
Distance: 5.0726% / 0.05072575
50.4 German
49.6 Balkan


Distance to: Gallop_scaled
0.08731051 German:DEU_Halberstadt_LBA
0.08803563 Balkan:BGR_IA
0.12321748 Slav:UKR_Srubnaya_MLBA
0.14449229 Turkic:TUR_Kaman-Kalehoyuk_MLBA
0.15089078 Baltic:Baltic_EST_MA
0.16710271 Levant:Levant_Abel_IA

[Impaler]

For everyone?

Scythian_MDA is pretty much in between of BGR_IA and HRV_IA, should work for all while determining Balkanic ancestry.
For Slav it is Avar Szolad.
For Turkic DA89 is proxy of Oghuz and Rus_Nomad is proxy of Steppe Turks(Cumans, Bulgars and so on)

You don't need any other component but if you want to add Germanic, Deu_ma should work.


Target: Impaler_scaled
Distance: 3.3350% / 0.03335028
79.0 Balkan:Scythian_MDA
13.6 Slav:HUN_Avar_Szolad
7.4 Turkic:KAZ_Turk: DA89
0.0 German: DEU_MA
0.0 Turkic:RUS_Nomad_MA

[kleenex]

Distance: 3.2284% / 0.03228365
57.2 Balkan
34.4 Scythian_MDA
8.4 Slav

[Luke35]

Here is me with Kaspias suggested pops vs. gixajo pops:

Kaspias calculator:

Target: MrG_scaled
Distance: 1.6654% / 0.01665364
48.2 German
31.8 Balkan
20.0 Slav

gixajo calculator (no Baltic):

Target: MrG_scaled
Distance: 2.3664% / 0.02366386
53.0 German
24.8 Balkan
22.2 Slav

[gixajo]

For everyone?

Scythian_MDA is pretty much in between of BGR_IA and HRV_IA, should work for all while determining Balkanic ancestry.
For Slav it is Avar Szolad.
For Turkic DA89 is proxy of Oghuz and Rus_Nomad is proxy of Steppe Turks(Cumans, Bulgars and so on)

You don't need any other component but if you want to add Germanic, Deu_ma should work.

And RUS_Sunghir_MA or CZE_Early_Slav for slavs?

Can´t we change one Anatolian sample by one Levantine and one Turkic sample? That was my idea originally in this model.

I don´t want most usual references if possible, at least I was trying not to plagiarize other sources.

And thanks.:thumb001: a little light is always appreciated among so many darkness.

[Kaspias]

And RUS_Sunghir_MA or CZE_Early_Slav for slavs?

Can´t we change one Anatolian sample by one Levantine and one Turkic sample? That was my idea originally in this model.

I don´t want most usual references if possible, at least I was trying not to plagiarize other sources.

And thanks.:thumb001: a little light is always appreciated among so many darkness.

Both Sunghir and Early Slav would work efficiently. The reason why I prefer Szolad is it is the earliest Slavic(in modern form) sample we have. Sunghir dates to MA and might have some other influences, and CZE Early Slav have roughly a quarter native/german like addition. Besides them, I like to use European Horde sample which is Slavic actually.

For Balkan, Scythian_MDA covers Romania. BGR_IA covers Bulgaria, Northern Macedonia and Northern Greece. HRV_IA covers Western Balkans and Mycen covers Southern Greece. So, it is hard to choose one sole reference for it. But MDA is kind of the middle point.

For Turkic, it is always best to use Kaz Turk samples and Rus nomad. Ottoman avg is 1/4 Anatolian, I wouldn't use it.

Anatolian and Levantine samples work in same function here, just check their single pops and you will see same similar populations. TUR_IA and TUR_IA_low_res is not Anatolian, so just abandon them. On the other hand, there is also no need to another Anatolian component if there is already Levantine component. If you want to add Anatolian component, then I can recommend outlier roopkund b sample, who is a Anatolian Greek from 1700s.

These are basic materials of creating an efficient Balkan calculator. You can make it more interesting by checking single population sharings of the samples I stated and find funnier pops from the list.

[gixajo]

Both Sunghir and Early Slav would work efficiently. The reason why I prefer Szolad is it is the earliest Slavic(in modern form) sample we have. Sunghir dates to MA and might have some other influences, and CZE Early Slav have roughly a quarter native/german like addition. Besides them, I like to use European Horde sample which is Slavic actually.

For Balkan, Scythian_MDA covers Romania. BGR_IA covers Bulgaria, Northern Macedonia and Northern Greece. HRV_IA covers Western Balkans and Mycen covers Southern Greece. So, it is hard to choose one sole reference for it. But MDA is kind of the middle point.

For Turkic, it is always best to use Kaz Turk samples and Rus nomad. Ottoman avg is 1/4 Anatolian, I wouldn't use it.

Anatolian and Levantine samples work in same function here, just check their single pops and you will see same similar populations. TUR_IA and TUR_IA_low_res is not Anatolian, so just abandon them. On the other hand, there is also no need to another Anatolian component if there is already Levantine component. If you want to add Anatolian component, then I can recommend outlier roopkund b sample, who is a Anatolian Greek from 1700s.

These are basic materials of creating an efficient Balkan calculator. You can make it more interesting by checking single population sharings of the samples I stated and find funnier pops from the list.

As you said, every component must be "purest" possible in the sense of, with the less admixture with other refrences with which it shares the model,but an Anatolian Greek sample would not have the same problem some samples I used? I mean, that individual has not any input from Balkans? I did not check it in any model to know it,( I am judging it only by his "Greek" included in the sample name.)

One more question/doubt about the same, sorry I know you already have spoken about this.

So you advise use this:

Balkan+Slav+German+(Anatolian)

But if I am interested in separate well every input coming from nowadays considered European references and current considered non-European components, which problems could have in trying:

Balkan+Slav+German+(Turkic+Levantine+(Caucasus))

In Iberian peninsula calculators we do something similar than the latter I wrote, we separate the purest Berber input and a purer SSA input, because if the sample is not so pure, any Northafrican sample has itself Berber+SSA.

I don't know if I explain myself well enough for you to understand what I'm trying to say.

And sorry, I don't want to seem like I'm bugging you.

I asked to another person here about this issue time ago, but I don´t know where I saved references that person recommend me, and I wouldn´t bother her again with this issue.

[Ylla]

Distance: 2.7858% / 0.02785826
75.6 Balkan
20.4 Slav
4.0 Turkic

[paradox]

Target: scaled
Distance: 2.9901% / 0.02990121
31.4 Balkan
26.0 Slav
15.4 Turkic
14.0 German
13.2 Levant


My father's

Target: Father_scaled
Distance: 4.0083% / 0.04008322
50.0 Balkan
35.8 Slav
10.0 German
4.2 Levant


With Baltic added

Target: scaled
Distance: 1.7381% / 0.01738127
42.2 Turkic
34.8 Baltic
13.6 Balkan
6.0 Levant
3.4 Slav


My father's

Target: Father_scaled
Distance: 3.0938% / 0.03093846
35.6 Baltic
33.2 Balkan
20.8 Turkic
9.0 Slav
1.4 Levant

[Kaspias]

As you said, every component must be "purest" possible in the sense of, with the less admixture with other refrences with which it shares the model,but an Anatolian Greek sample would not have the same problem some samples I used? I mean, that individual has not any input from Balkans? I did not check it in any model to know it,( I am judging it only by his "Greek" included in the sample name.)

One more question/doubt about the same, sorry I know you already have spoken about this.

So you advise use this:

Balkan+Slav+German+(Anatolian)

But if I am interested in separate well every input coming from nowadays considered European references and current considered non-European components, which problems could have in trying:

Balkan+Slav+German+(Turkic+Levantine+(Caucasus))

In Iberian peninsula calculators we do something similar than the latter I wrote, we separate the purest Berber input and a purer SSA input, because if the sample is not so pure, any Northafrican sample has itself Berber+SSA.

I don't know if I explain myself well enough for you to understand what I'm trying to say.

And sorry, I don't want to seem like I'm bugging you.

I asked to another person here about this issue time ago, but I don´t know where I saved references that person recommend me, and I wouldn´t bother her again with this issue.

Well, Kaman-Kalehöyük or other MLBA samples from Anatolia are just a copy of each other and show a perfect genetic continuum of the Levant area with no addition. Yet, the later Greek Anatolian sample also carries the Caucasian shift which today present in the region. I mean; from MLBA to just before Turkic conquest something has changed in the region but we don't have samples from that time of period. By suggesting using Roopkund sample I assume that there have been no changes in DNA of Greek speaker Anatolians from pre-Turkic to during-Turkic. If you worry if these Greek speakers carry Greek/Balkan admixture, this is not the case. They speak Greek because they Hellenized, otherwise they don't carry actual Greek/Balkan DNA.

I actually prefer to use only three components on Balkans(Balkan-Slav-Turk). Because such problems occur after adding other components:

- When adding Germans, for example Western Balkanites and Romanians usually get German scores and they think they really may have some type of Germanic. No, this is not the case. Remember the Balkan samples, Scythian_MDA is Northern than BGR_IA because MDA carry more Steppe admixture, the same goes for HRV_IA. And when you combine German with BGR_IA it gives you additional Germanic to cover that surplus Steppe admixture.
- When adding Levant or Anatolian, for example in my case, I usually modeled like 50% Levant and 50% Baltic which is funny. I lack of west_med admixture but I have rich EHG at the same time, so calculator prefers to give me East Med + Eastern Euro instead of balancing Balkanic with other components. This is something very common for Eastern Balkans as we have additional East Med input.
- Something similar goes for Caucasian. Bulgarians are usually rich on CHG compared to other Balkaners, so they tend to be modeled like high Caucasian + Slav + East Med... weird combinations.

But there is no such chance if we use only three major components present in the region. I got what you mean, but it is kind of impossible to a perfect calculator for everyone while using high K's. Such a calculator might work awesome on 80% of the portion, but will be terrible in 20% at least.

[Thracian]

Target: Thracian_scaled
Distance: 3.3636% / 0.03363592 | ADC: 0.25x
37.2 German
32.8 Turkic
15.4 Slav
14.6 Levant

[InfamousAngel99]

Without Baltic:
0.03251362 German:DEU_Halberstadt_LBA
0.07044988 Slav:UKR_Srubnaya_MLBA
0.12755493 Balkan:BGR_IA
0.18257754 Turkic:TUR_Kaman-Kalehoyuk_MLBA
0.20632082 Levant:Levant_Abel_IA

Distance: 2.6550% / 0.02654975
79.0 German
13.8 Balkan
7.2 Slav

With Baltic:
0.03251362 German:DEU_Halberstadt_LBA
0.07044988 Slav:UKR_Srubnaya_MLBA
0.09476882 Baltic:Baltic_EST_MA
0.12755493 Balkan:BGR_IA
0.18257754 Turkic:TUR_Kaman-Kalehoyuk_MLBA
0.20632082 Levant:Levant_Abel_IA

Distance: 2.5752% / 0.02575220
70.4 German
16.2 Balkan
9.0 Baltic
4.4 Slav

[17571imre]

Both Sunghir and Early Slav would work efficiently. The reason why I prefer Szolad is it is the earliest Slavic(in modern form) sample we have. Sunghir dates to MA and might have some other influences, and CZE Early Slav have roughly a quarter native/german like addition. Besides them, I like to use European Horde sample which is Slavic actually.

For Balkan, Scythian_MDA covers Romania. BGR_IA covers Bulgaria, Northern Macedonia and Northern Greece. HRV_IA covers Western Balkans and Mycen covers Southern Greece. So, it is hard to choose one sole reference for it. But MDA is kind of the middle point.

For Turkic, it is always best to use Kaz Turk samples and Rus nomad. Ottoman avg is 1/4 Anatolian, I wouldn't use it.

Anatolian and Levantine samples work in same function here, just check their single pops and you will see same similar populations. TUR_IA and TUR_IA_low_res is not Anatolian, so just abandon them. On the other hand, there is also no need to another Anatolian component if there is already Levantine component. If you want to add Anatolian component, then I can recommend outlier roopkund b sample, who is a Anatolian Greek from 1700s.

These are basic materials of creating an efficient Balkan calculator. You can make it more interesting by checking single population sharings of the samples I stated and find funnier pops from the list.

so i decided to use all the samples that you mentiond instead of one sample for Balkan and this is the result:

Target: mh_scaled
Distance: 1.7479% / 0.01747893
37.4 HUN_Avar_Szolad
26.2 BGR_IA
23.6 DEU_MA
8.0 Scythian_MDA
4.4 HRV_IA
0.4 RUS_Nomad_MA

Target: cb_scaled
Distance: 2.3627% / 0.02362669
53.6 HUN_Avar_Szolad
26.2 BGR_IA
9.4 DEU_MA
6.8 RUS_Nomad_MA
4.0 Scythian_MDA

it works really good for us. so i can also see from this that my Balkan side is mostly from the East Balkan.

[Luke35]

so i decided to use all the samples that you mentiond instead of one sample for Balkan and this is the result:

Target: mh_scaled
Distance: 1.7479% / 0.01747893
37.4 HUN_Avar_Szolad
26.2 BGR_IA
23.6 DEU_MA
8.0 Scythian_MDA
4.4 HRV_IA
0.4 RUS_Nomad_MA

Target: cb_scaled
Distance: 2.3627% / 0.02362669
53.6 HUN_Avar_Szolad
26.2 BGR_IA
9.4 DEU_MA
6.8 RUS_Nomad_MA
4.0 Scythian_MDA

it works really good for us. so i can also see from this that my Balkan side is mostly from the East Balkan.

Stalking you as usual :)

Here is me, my Hungarian side, and the G25 Hungarian average, with those same sources:


Target: MrG_scaled
Distance: 1.6302% / 0.01630201
46.6 DEU_MA
25.4 Scythian_MDA
20.8 HUN_Avar_Szolad
7.2 HRV_IA


Target: MrG-dad
Distance: 2.5028% / 0.02502841
42.2 Scythian_MDA
32.8 HUN_Avar_Szolad
19.2 DEU_MA
3.6 HRV_IA
2.2 KAZ_TurkA89


Target: Hungarian
Distance: 1.0635% / 0.01063528
52.0 HUN_Avar_Szolad
25.6 DEU_MA
15.8 BGR_IA
4.6 Scythian_MDA
2.0 RUS_Nomad_MA

[Beowulf]

Target: Beowulf_Scaled
Distance: 0.0131% / 0.01314326
55.5 Iberian
22.8 German
16.8 Balkan
4.9 Slav

Using iberian for me of course

[noricum]

w/o Baltic

Target: G25noricum_scaled
Distance: 4.8345% / 0.04834451
49.4 German
27.8 Slav
22.8 Balkan

Distance to: G25noricum_scaled
0.05773045 German:DEU_Halberstadt_LBA
0.07886508 Slav:UKR_Srubnaya_MLBA
0.12648572 Balkan:BGR_IA
0.18118422 Turkic:TUR_Kaman-Kalehoyuk_MLBA
0.20426063 Levant:Levant_Abel_IA

w Baltic

Target: G25noricum_scaled
Distance: 3.3938% / 0.03393761
47.8 Baltic
35.4 Balkan
13.2 Slav
3.6 German

Distance to: G25noricum_scaled
0.05773045 German:DEU_Halberstadt_LBA
0.07886508 Slav:UKR_Srubnaya_MLBA
0.08711793 Baltic:Baltic_EST_MA
0.12648572 Balkan:BGR_IA
0.18118422 Turkic:TUR_Kaman-Kalehoyuk_MLBA
0.20426063 Levant:Levant_Abel_IA

[ScandinavianCelt]

Target: Beowulf_Scaled
Distance: 0.0131% / 0.01314326
55.5 Iberian
22.8 German
16.8 Balkan
4.9 Slav

Using iberian for me of course

Which Iberian sample did you use? I want to try it with the Iberian sample also.

This is what I get for all the original samples he posted:


Target: Brad.Davidski.AncestryDNA.Official.G25
Distance: 3.4427% / 0.03442666

53.4 German:DEU_Halberstadt_LBA
27.6 Slav:UKR_Srubnaya_MLBA
13.0 Balkan:BGR_IA
6.0 Baltic:Baltic_EST_MA

[cass]

Target: cass
Distance: 4.5819% / 0.04581866
69.0 German:DEU_Halberstadt_LBA
24.4 Slav:UKR_Srubnaya_MLBA
6.6 Balkan:BGR_IA

Target: cass
Distance: 2.1761% / 0.02176095
55.8 Baltic:Baltic_EST_MA
21.2 Balkan:BGR_IA
15.6 German:DEU_Halberstadt_LBA
7.4 Slav:UKR_Srubnaya_MLBA

:bored:

[Fabricius]

Target: Fabricius_K13-G25
Distance: 6.8843% / 0.06884341
53.8 German
20.6 Balkan
18.0 Levant
7.6 Slav

Target: Fabricius_K15-G25
Distance: 7.1048% / 0.07104808
53.0 German
21.0 Balkan
17.8 Levant
8.2 Slav

Target: Fabricius_K36-G25
Distance: 7.0749% / 0.07074887
50.2 German
23.4 Balkan
19.8 Levant
6.6 Slav

Target: Fabricius_MyHeritage-to-g25(simulated)
Distance: 6.7811% / 0.06781144
47.8 German
22.8 Levant
19.0 Balkan
10.4 Slav


Plus Baltic:
Target: Fabricius_K13-G25
Distance: 6.8106% / 0.06810623
42.2 German
21.2 Balkan
20.2 Levant
14.2 Baltic
2.2 Slav

Target: Fabricius_K15-G25
Distance: 6.9982% / 0.06998171
38.4 German
22.2 Balkan
20.2 Levant
17.4 Baltic
1.8 Slav

Target: Fabricius_K36-G25
Distance: 7.0152% / 0.07015157
39.6 German
24.2 Balkan
21.6 Levant
13.0 Baltic
1.6 Slav

Target: Fabricius_MyHeritage-to-g25(simulated)
Distance: 6.7044% / 0.06704363
36.2 German
25.0 Levant
19.6 Balkan
14.4 Baltic
4.8 Slav


Individual distances:
Distance to: Fabricius_K13-G25
0.09493262 German:DEU_Halberstadt_LBA
0.10816406 Balkan:BGR_IA
0.11820989 Slav:UKR_Srubnaya_MLBA
0.14350084 Baltic:Baltic_EST_MA
0.14987084 Turkic:TUR_Kaman-Kalehoyuk_MLBA
0.16780675 Levant:Levant_Abel_IA

Distance to: Fabricius_K15-G25
0.09660472 German:DEU_Halberstadt_LBA
0.10944514 Balkan:BGR_IA
0.11943103 Slav:UKR_Srubnaya_MLBA
0.14349778 Baltic:Baltic_EST_MA
0.15087630 Turkic:TUR_Kaman-Kalehoyuk_MLBA
0.16874386 Levant:Levant_Abel_IA

Distance to: Fabricius_K36-G25
0.10176783 German:DEU_Halberstadt_LBA
0.10410393 Balkan:BGR_IA
0.12497154 Slav:UKR_Srubnaya_MLBA
0.14362671 Turkic:TUR_Kaman-Kalehoyuk_MLBA
0.15093204 Baltic:Baltic_EST_MA
0.16200659 Levant:Levant_Abel_IA

Distance to: Fabricius_MyHeritage-to-g25(simulated)
0.09974840 German:DEU_Halberstadt_LBA
0.10407827 Balkan:BGR_IA
0.12170323 Slav:UKR_Srubnaya_MLBA
0.14208579 Turkic:TUR_Kaman-Kalehoyuk_MLBA
0.14863377 Baltic:Baltic_EST_MA
0.15999353 Levant:Levant_Abel_IA

[vasiliki23]

Target: VASILIKI_scaled
Distance: 3.3874% / 0.03387368
58.4 Balkan
16.4 Slav
15.6 German
9.6 Turkic

With Baltic added
Target: VASILIKI_scaled
Distance: 3.1466% / 0.03146625
53.0 Balkan
18.6 Baltic
17.4 Turkic
7.8 Slav
3.2 German

[RyoHazuki]

Target: G25K13SimRyoHazuki
Distance: 3.0150% / 0.03014965
76.0 German
18.0 Balkan
6.0 Baltic

Target: Ryo_scaled
Distance: 5.3191% / 0.05319131
64.2 German
18.4 Balkan
15.6 Slav
1.8 Baltic

[ScandinavianCelt]

Which Iberian sample did you use? I want to try it with the Iberian sample also.

This is what I get for all the original samples he posted:


Target: Brad.Davidski.AncestryDNA.Official.G25
Distance: 3.4427% / 0.03442666

53.4 German:DEU_Halberstadt_LBA
27.6 Slav:UKR_Srubnaya_MLBA
13.0 Balkan:BGR_IA
6.0 Baltic:Baltic_EST_MA

using official 2-set ave


Target: SC_Ave_IllustrativeDNA
Distance: 2.6316% / 0.02631624
63.2 German
20.0 Slav
10.2 Balkan
6.6 Baltic

Target: SC_23andMe_Scaled
Distance: 2.8836% / 0.02883596
55.4 German
22.0 Slav
11.8 Balkan
10.8 Baltic