nMonte using Global PCA 25 spreadsheet

[Aren]

New run with a big spreadsheet including many modern and ancient samples. Some examples

"distance%=3.9416"

Icelandic

Yamnaya_Kalmykia,33.4
Barcin_N,29.4
Blatterhole_HG,20.6
Yamnaya_Samara,16.6

"distance%=3.7211"

Swedish

Yamnaya_Kalmykia,34.4
Barcin_N,29
Blatterhole_HG,21.6
Yamnaya_Samara,15

"distance%=3.8633"

Scottish

Barcin_N,34.2
Yamnaya_Kalmykia,29.8
Yamnaya_Samara,18.2
Blatterhole_HG,17.8

"distance%=2.2357"

French_East

Barcin_N,43.2
Yamnaya_Kalmykia,23.4
Yamnaya_Samara,18
Blatterhole_HG,15.4

"distance%=1.6084"

Italian_Bergamo

Barcin_N,58
Yamnaya_Samara,32.6
Blatterhole_HG,7.8
Yamnaya_Kalmykia,1.6

"distance%=1.1221"

Italian_South

Barcin_N,46.4
Yamnaya_Samara,23
Iran_IA,21.8
Levant_BA,3.8
Mozabite,3.6
Blatterhole_HG,1.4

"distance%=1.3507"

Sicilian_East

Barcin_N,47.8
Yamnaya_Samara,22
Iran_IA,16.4
Levant_BA,6.8
Mozabite,5.4
Blatterhole_HG,1.6

"distance%=3.1192"

Spanish_Aragon

Barcin_N,52
Yamnaya_Samara,21.4
Blatterhole_HG,16.2
Yamnaya_Kalmykia,7.4
Mozabite,3

"distance%=2.2353"

Spanish_Extremadura

Barcin_N,47.8
Yamnaya_Samara,25
Blatterhole_HG,14
Mozabite,9.8
Yamnaya_Kalmykia,3.4

"distance%=2.5611"

Portuguese

Barcin_N,49
Yamnaya_Samara,28
Blatterhole_HG,14
Mozabite,9

[Sikeliot]

Try Sicilian West, and various Greek groups?

[Aren]

Try Sicilian West, and various Greek groups?

"distance%=1.3615"

Sicilian_West

Barcin_N,42.8
Iran_IA,19.4
Yamnaya_Samara,19
Mozabite,8.2
Blatterhole_HG,6
Levant_BA,4.6

"distance%=1.891"

Greek

Barcin_N,50.6
Yamnaya_Samara,20
Iran_IA,19.6
Yamnaya_Kalmykia,6.2
Blatterhole_HG,3
Levant_BA,0.6

[Sikeliot]

"distance%=1.3615"

Sicilian_West

Barcin_N,42.8
Iran_IA,19.4
Yamnaya_Samara,19
Mozabite,8.2
Blatterhole_HG,6
Levant_BA,4.6

"distance%=1.891"

Greek

Barcin_N,50.6
Yamnaya_Samara,20
Iran_IA,19.6
Yamnaya_Kalmykia,6.2
Blatterhole_HG,3
Levant_BA,0.6


West Sicily clearly has more Near East and Berber (reaching South Spain/Portugal levels) than does East Sicily. West Sicily here = Trapani while East Sicily = Syracuse.

What is "Blatterhole"?

[Aren]

West Sicily clearly has more Near East and Berber (reaching South Spain/Portugal levels) than does East Sicily. West Sicily here = Trapani while East Sicily = Syracuse.

What is "Blatterhole"?

WHG from Germany.

[Aren]

Try Sicilian West, and various Greek groups?
I think this gives a better result

"distance%=1.3608"

Sicilian_West

Barcin_N,43.2
Yamnaya_Samara,19.8
Iran_IA,18.6
Mozabite,8.2
Levant_BA,5.6
WHG,4.6

"distance%=1.3444"

Sicilian_East

Barcin_N,47.8
Yamnaya_Samara,21.8
Iran_IA,16.8
Levant_BA,6.8
Mozabite,5.4
WHG,1.4

[Sikeliot]

I think this gives a better result

"distance%=1.3608"

Sicilian_West

Barcin_N,43.2
Yamnaya_Samara,19.8
Iran_IA,18.6
Mozabite,8.2
Levant_BA,5.6
WHG,4.6

"distance%=1.3444"

Sicilian_East

Barcin_N,47.8
Yamnaya_Samara,21.8
Iran_IA,16.8
Levant_BA,6.8
Mozabite,5.4
WHG,1.4


West Sicily is slightly more MENA than East Sicily still. East Sicily, on the other hand, seems more Neolithic/"original" SE Euro.

[Aren]

West Sicily is slightly more MENA than East Sicily still. East Sicily, on the other hand, seems more Neolithic/"original" SE Euro.

Do East Sicilians score higher West/East Med than West Sicilians on K13/K15?

[Sikeliot]

Do East Sicilians score higher West/East Med than West Sicilians on K13/K15?

Depends on the region, but I can say I think what you modeled above is reflective of the trend I see, yes.

[Coolguy1]

[1] "distance%=2.6763 / distance=0.026763"

Italian_South

Mycenaean:I9041 39.50
Yamnaya_Bulgaria 24.50
Mycenaean:I9033 19.10
Mycenaean:I9006 14.15
Russian_Kursk 2.75

[Coolguy1]

[1] "distance%=2.4616 / distance=0.024616"

Sorcelow_scaled

Italian_South 79.20
Russian_Tver 18.45
Russian_Kursk 2.35


[1] "distance%=2.3653 / distance=0.023653"

Italian_South

Mycenaean:I9041 32.0
Mycenaean:I9033 27.9
Yamnaya_Samara:I0443 16.6
Mycenaean:I9010 14.6
Mycenaean:I9006 8.9

[Thambi]

[1] "distance%=3.7764 / distance=0.037764"

Thambi

"ASI_Ghost" 49.05
"Iran_N" 29.25
"Yamnaya_Ukraine_outlier" 21.7

Thambi

"ASI_Ghost" 54.8
"Armenia_EBA" 42.55
"Yamnaya_Bulgaria" 2.65

[1] "distance%=4.4572 / distance=0.044572"

Thambi

"Paniya" 65.05
"Iran_N" 26.75
"Andronovo" 8.2
"Karasuk" 0

roughly equidistant to pashtuns and paniyas. It makes sense but I thought it was interesting.

[1] "distance%=4.9507 / distance=0.049507"

Thambi

"Paniya" 53.35
"Pashtun" 46.65

[ADonkeyBrain]

[1] "distance%=2.4616 / distance=0.024616"

Sorcelow_scaled

Italian_South 79.20
Russian_Tver 18.45
Russian_Kursk 2.35


[1] "distance%=2.3653 / distance=0.023653"

Italian_South

Mycenaean:I9041 32.0
Mycenaean:I9033 27.9
Yamnaya_Samara:I0443 16.6
Mycenaean:I9010 14.6
Mycenaean:I9006 8.9

I'm guessing you are mostly trying to see how these populations broadly relate to each other because the models are rather overfitted (the Myceaneans have correlations with each other in nMonte with Global25 in the high .90s so using an average or the one that works best might be a better idea) and kinda unrealistic at that. They end up giving you kinda large distances too, compared to other, more reasonable combos.

Here's something that you get if you throw everything together and reasonably prune after that:

[1] "distance%=0.6303"

Greek

Minoan_Lasithi,38.2
Varna_outlier,19.6
Armenia_EBA,16.2
Baltic_BA,15.6
Levant_BA,7
Scythian_AldyBel,3.4

[1] "distance%=1.0688"

Italian_South

Minoan_Lasithi,36.8
Nordic_IA,26.4
England_Roman_outlier,19.6
Armenia_EBA,14.2
Srubnaya_outlier,3

You can get similar stuff for the rest of the Balkans if you use something Neolithic, especially with extra CHG, together with a steppe source or alternatively one of the ancient post-steppe influx Balkan genomes we already have (e.g. Mycenaean or Balkan_IA), something Baltic-like (Baltic_BA fits all of the Balkans very well and is likely a decently close proxy for early/proto-Slavs) and a combination of Near Eastern sources, one representing the more Caucasus side of things and the other the more Levantine one.

For Italy, Neolithic with extra CHG and/or WHG, 'Italo-Celto-Germanic' stuff (Nordic_IA, England_IA, various Bell Beakers with more steppe work reasonably well depending on the sample) and similar Near Eastern stuff (especially England_Roman_outlier for Italy, maybe because there's some North African in Italy) works well too.

Here are two other, random models based on that sort of principle and you can try various combos:

[1] "distance%=1.4468"

Greek

Balkans_IA,58
Baltic_BA,17.8
Armenia_EBA,17.6
Levant_BA,6.6

[1] "distance%=1.362

Italian_South

Mycenaean,46.6
Nordic_IA,23.6
Levant_BA,15.7
Armenia_EBA,14.0

Btw, care to share your coordinates here or in PM? I'm curious to see what you can be modelled as along those lines, I'm guessing you're very ancient Balkan-like.

[Coolguy1]

I'm guessing you are mostly trying to see how these populations broadly relate to each other because the models are rather overfitted (the Myceaneans have correlations with each other in nMonte with Global25 in the high .90s so using an average or the one that works best might be a better idea) and kinda unrealistic at that. They end up giving you kinda large distances too, compared to other, more reasonable combos.

Here's something that you get if you throw everything together and reasonably prune after that:

[1] "distance%=0.6303"

Greek

Minoan_Lasithi,38.2
Varna_outlier,19.6
Armenia_EBA,16.2
Baltic_BA,15.6
Levant_BA,7
Scythian_AldyBel,3.4

[1] "distance%=1.0688"

Italian_South

Minoan_Lasithi,36.8
Nordic_IA,26.4
England_Roman_outlier,19.6
Armenia_EBA,14.2
Srubnaya_outlier,3

You can get similar stuff for the rest of the Balkans if you use something Neolithic, especially with extra CHG, together with a steppe source or alternatively one of the ancient post-steppe influx Balkan genomes we already have (e.g. Mycenaean or Balkan_IA), something Baltic-like (Baltic_BA fits all of the Balkans very well and is likely a decently close proxy for early/proto-Slavs) and a combination of Near Eastern sources, one representing the more Caucasus side of things and the other the more Levantine one.

For Italy, Neolithic with extra CHG and/or WHG, 'Italo-Celto-Germanic' stuff (Nordic_IA, England_IA, various Bell Beakers with more steppe work reasonably well depending on the sample) and similar Near Eastern stuff (especially England_Roman_outlier for Italy, maybe because there's some North African in Italy) works well too.

Here are two other, random models based on that sort of principle and you can try various combos:

[1] "distance%=1.4468"

Greek

Balkans_IA,58
Baltic_BA,17.8
Armenia_EBA,17.6
Levant_BA,6.6

[1] "distance%=1.362

Italian_South

Mycenaean,46.6
Nordic_IA,23.6
Levant_BA,15.7
Armenia_EBA,14.0

Btw, care to share your coordinates here or in PM? I'm curious to see what you can be modelled as along those lines, I'm guessing you're very ancient Balkan-like.

What is Balkan IA?

[Kelmendasi]

What is Balkan IA?
Balkan Iron Age I am assuming. How do you run this?

[Coolguy1]

Balkan Iron Age I am assuming. How do you run this?

You need to pay Davidski for the coordinates

[Aren]

A new sample was added, presumably a Roman soldier found in Munich

"1. CLOSEST SINGLE ITEM DISTANCE%"
Spanish_Baleares Spanish_Cataluna Spanish_Murcia Spanish_Cantabria Spanish_Andalucia Spanish_Castilla_La_Mancha Spanish_Aragon Spanish
2.403033 2.657016 2.766396 2.908725 2.940358 3.080830 3.128336 3.171545

"distance%=2.3108"

Germany_Roman

Spanish_Baleares,52.6
Spanish_Murcia,26.8
Spanish_Cantabria,11
French_East,9.6

[ADonkeyBrain]

What is Balkan IA?

It's a 5th century BC likely Thracian woman from northern Bulgaria. Very close to the Mycenaeans too due to a similar profile - high EEF, some CHG, some steppe.


Balkan Iron Age I am assuming. How do you run this?

If you want to run your own self, as Sorcelow mentioned, you need to pay Eurogenes to get your coordinates. If you just want to try stuff out in general, get nMonte here (https://www.dropbox.com/sh/1iaggxyc2alafow/AACIjLtnkuaNNsJ5oKME_3XHa?dl=0), instructions on how to run things are here (https://anthrogenica.com/showthread.php?6452-How-to-Use-nMonte&) and the datasheets for Global25 are here (http://eurogenes.blogspot.gr/2018/02/unleash-power-global-25-test-drive.html)

[Aren]

Something interesting I noticed using the Balkan_Chl sample. Basically every Euro group especially Southern Euros and Eastern Euros prefer it as an additional input to Bell Beakers/Corded Ware. When removed it does not always pick up other Chalcolithic samples from Europe but instead gives more Anatolia_BA or Armenia_EBA(Kura Axes) or Greece_N.

Seems pretty reasonable with Albanians, when removed the fit get's significantly worse and most of the Balkan_Chl goes to Anatolian_BA. It must be a sample with CHG/Iran_N admix already present to a small degree and probably more basal than Barcin and the older European farmers. Anyone know what Y-DNA the Balkan_Chl belonged to?

"distance%=2.0488"

Albanian

Balkans_ChL,29.4
Anatolia_BA,23.8
CWC_Baltic,21.2
Yamnaya_Bulgaria,20.6
Sidon_BA,4
Ottoman,1.2

Now with a more northern Balkan group, it's about the same %

"distance%=2.6528"

Slovenian

CWC_Baltic,59.6
Balkans_ChL,28.8
Yamnaya_Bulgaria,11.2
Sidon_BA,0.4

Even Polish sample prefers it over GAC and other neolithic samples from central Europe.

"distance%=3.9043"

Polish

CWC_Baltic,78.8
Balkans_ChL,17.8
Comb_Ceramic,3.4

When removed

"distance%=4.1569"

Polish

CWC_Baltic,78
Anatolia_BA,9.8
Globular_Amphora,9.8
Comb_Ceramic,2.4

Also present in Iberia

"distance%=1.2688"

Spanish_Castilla_La_Mancha

Beaker_The_Netherlands,43.6
Iberia_Southwest_CA,22.2
Balkans_ChL,17
Sidon_BA,6.8
Mozabite,5.8
Iberia_Central_CA,4.6

And very high in northern Italy

"distance%=1.1348"

Italian_Bergamo

Balkans_ChL,46
Beaker_The_Netherlands,26.6
Battle_Axe_Sweden,17
Sidon_BA,5.2
Armenia_EBA,3
Iberia_Southwest_CA,2.2

And in another very Italo-Celtic admixed people, like the French

"distance%=1.2558"

French

Beaker_The_Netherlands,46.6
Balkans_ChL,23.2
Battle_Axe_Sweden,16
Iberia_Central_CA,8
Iberia_Southwest_CA,5
Sidon_BA,1.2

Basically only absent or close to absent in NW Europe. Also notice how it doesn't give British_MN or Swedish_MN but instead GAC. Maybe due to lack of TRB samples from Denmark and the Netherlands?

"distance%=2.2121"

Norwegian

Beaker_The_Netherlands,80.6
Globular_Amphora,16
CWC_Baltic,3.4

"distance%=2.0046"

English

Beaker_The_Netherlands,56.6
Globular_Amphora,20.8
Battle_Axe_Sweden,19.6
Balkans_ChL,3


I'm thinking groups rich in Balkans_Chl admix moved up to most of central Europe during the early Bronze Age and it must've been the major non-Steppe component amongst proto Italo-Celts and even the proto Balto-Slavs. The huge thing here though is that it's very low in supposedly very Celtic areas like Britain, almost absent.

[FilhoV]

FilhoV_scaled

Beaker_the_Netherlands- 36.8%
France_MLN -32.4%
Anatolia_EBA 22%
Moroccan 8.8%

Distance. 2.3782%

[FilhoV]

A new sample was added, presumably a Roman soldier found in Munich

"1. CLOSEST SINGLE ITEM DISTANCE%"
Spanish_Baleares Spanish_Cataluna Spanish_Murcia Spanish_Cantabria Spanish_Andalucia Spanish_Castilla_La_Mancha Spanish_Aragon Spanish
2.403033 2.657016 2.766396 2.908725 2.940358 3.080830 3.128336 3.171545

"distance%=2.3108"

Germany_Roman

Spanish_Baleares,52.6
Spanish_Murcia,26.8
Spanish_Cantabria,11
French_East,9.6

Amazing

[Rgvgjhvv]

FilhoV_scaled

Beaker_the_Netherlands- 36.8%
France_MLN -32.4%
Anatolia_EBA 22%
Moroccan 8.8%

Distance. 2.3782%

How is this done?

[FilhoV]

How is this done?

Just plug in the coordinates on the run you need done I’m horrible and have others do it for me or my spouse doing it