How is that possible? A glitch or coincidence?

[Thracian]

I also tried with this using pen=0 and also added Albanians but the scores stayed same.

You can also share your odd model results here.

https://i.ibb.co/GkmhW60/wtf2.png

[Kaspias]

Turkish_Balkans average is very close to the Bulgarian except their Turkmen admixture. Scoring Greek here means you have not only Bulgarian admixture but also Northern Greek admixture.

Some glitches on this model:

1.Bulgarian samples are not enough to make correct modeling since they are all Northern Shifted than average, there is no Thracian Bulgarian sample.
2. Greek samples are not from all over Greece but only from Thrace.
3. There is a lot of difference between some Turkish_Balkans samples. Ludogorie is even more northern than some Bulgarians while Thrace clearly shows huge Turkic admixtures. This difference shifts the average to the upwards. Thracian Turks need some other admixture(Greek or Macedonian) to balance their admixtures.

[Impaler]

Turkish_Balkans average is very close to the Bulgarian except their Turkmen admixture. Scoring Greek here means you have not only Bulgarian admixture but also Northern Greek admixture.

Some glitches on this model:

1.Bulgarian samples are not enough to make correct modeling since they are all Northern Shifted than average, there is no Thracian Bulgarian sample.
2. Greek samples are not from all over Greece but only from Thrace.
3. There is a lot of difference between some Turkish_Balkans samples. Ludogorie is even more northern than some Bulgarians while Thrace clearly shows huge Turkic admixtures. This difference shifts the average to the upwards. Thracian Turks need some other admixture(Greek or Macedonian) to balance their admixtures.

How is Turkish_Balkans from Silistra and Varna? I am the closest with Varna sample. On some calculators I am 90% the same.

[Impaler]

sample": "Custom:Impaler_scaled",
"fit": 1.9659,
"Turkish_Balkans": 69.17,
"Bulgarian": 21.67,
"Greek": 9.17,
"closestDistances": [
"Turkish_Balkans:Turkish_Varna1: 2.884407",
"Turkish_Balkans:Turkish_Ludogorie2: 3.256418",
"Turkish_Balkans:Turkish_Macedonia1: 3.918638",
"Turkish_Balkans:Turkish_Ludogorie1: 3.983168",
"Turkish_Balkans:Turkish_Silistra1: 4.002772",
"Turkish_Balkans:Turkish_Sumnu1: 4.198224",
"Bulgarian:BulgarianE2: 4.319624",
"Turkish_Balkans:Turkish_Bulgaria1: 4.563744",
"Bulgarian:BulgarianF2: 4.605767",
"Bulgarian:Bulgaria1: 4.641969",
"Greek:GREEKGRALPOP12: 4.765742",
"Turkish_Balkans:Turkish_Thrace1: 4.777900",
"Greek:GREEKGRALPOP10: 4.861372",
"Bulgarian:BulgarianD6: 4.903417",
"Bulgarian:BulgarianH2: 4.939614",
"Greek:NA17372: 5.106202",
"Greek:GREEKGRALPOP11: 5.124275",
"Greek:GREEKGRALPOP8: 5.266610",
"Greek:GREEKGRALPOP13: 5.280908",
"Greek:GREEKGRALPOP5: 5.380745",
"Greek:GREEKGRALPOP17: 5.429290",
"Greek:GREEKGRALPOP18: 5.446868",
"Greek:NA17377: 5.478847",
"Greek:GREEKGRALPOP3: 5.520621",
"Greek:GREEKGRALPOP14: 5.549773",
"Greek:GREEKGRALPOP4: 5.761957",
"Greek:GREEKGRALPOP15: 5.797399",
"Greek:GREEKGRALPOP9: 6.001356",
"Greek:NA17373: 6.240033",
"Greek:GREEKGRALPOP16: 6.302207",
"Greek:NA17375: 6.553955"

[Thracian]

Turkish_Balkans average is very close to the Bulgarian except their Turkmen admixture. Scoring Greek here means you have not only Bulgarian admixture but also Northern Greek admixture.

Some glitches on this model:

1.Bulgarian samples are not enough to make correct modeling since they are all Northern Shifted than average, there is no Thracian Bulgarian sample.
2. Greek samples are not from all over Greece but only from Thrace.
3. There is a lot of difference between some Turkish_Balkans samples. Ludogorie is even more northern than some Bulgarians while Thrace clearly shows huge Turkic admixtures. This difference shifts the average to the upwards. Thracian Turks need some other admixture(Greek or Macedonian) to balance their admixtures.

Well. To clarify, I also modeled myself with Balkan Turks using different test pop's. We had more or less similar results in those models. Yet, on this one, I am the only one who get them. I also tried with some other ethnicities such as Romanians, Serbians etc... but the scores remained the same. As you know, Balkan Turk average pop based on those samples. So, I was asking they all scored 100% Balkan Turks just because they are part of this average? Or I just have some foreign ancestry and that's why I get the results like this.

I agree with you. There are only 5 Bulgarian samples on Global 25. I believe that just like Turkish samples, they also should separate Bulgarian and Greek (Mainlander) samples into different regions.

Aren't there any Greek samples on Global 25 from Peloponnese, Epirus etc...?

[Kaspias]

How is Turkish_Balkans from Silistra and Varna? I am the closest with Varna sample. On some calculators I am 90% the same.

They are very close to me too, with slightly less Turkmen admixture and a slight southern shift. (Carrying more WA)

[Kaspias]

Well. To clarify, I also modeled myself with Balkan Turks using different test pop's. We had more or less similar results in those models. Yet, on this one, I am the only one who get them. I also tried with some other ethnicities such as Romanians, Serbians etc... but the scores remained the same. As you know, Balkan Turk average pop based on those samples. So, I was asking they all scored 100% Balkan Turks just because they are part of this average? Or I just have some foreign ancestry and that's why I get the results like this.

I agree with you. There are only 5 Bulgarian samples on Global 25. I believe that just like Turkish samples, they also should separate Bulgarian and Greek (Mainlander) samples into different regions.

Aren't there any Greek samples on Global 25 from Peloponnese, Epirus etc...?


Well. To clarify, I also modeled myself with Balkan Turks using different test pop's. We had more or less similar results in those models. Yet, on this one, I am the only one who get them. I also tried with some other ethnicities such as Romanians, Serbians etc... but the scores remained the same. As you know, Balkan Turk average pop based on those samples. So, I was asking they all scored 100% Balkan Turks just because they are part of this average? Or I just have some foreign ancestry and that's why I get the results like this.


They all scored 100% Balkan Turks because they are part of this average. But scoring Greek in your case gives some clues and pretty expected. ("Thracian Turks need some other admixture(Greek or Macedonian) to balance their admixtures because of Northern shifted Turkish_Balkan avg.")


I agree with you. There are only 5 Bulgarian samples on Global 25. I believe that just like Turkish samples, they also should separate Bulgarian and Greek (Mainlander) samples into different regions.


Greek samples are from Lazaridis. He registered them as "Thessaloniki" but in another article that he was mentioned author states samples collected from between Eastern Macedonia and Thrace. Includes possibly Pontic mix too, with this sense. User data will always be more reliable than academic samples that collected because of one specific aim but not for general analyze.

[Thracian]

They all scored 100% Balkan Turks because they are part of this average. But scoring Greek in your case gives some clues and pretty expected. ("Thracian Turks need some other admixture(Greek or Macedonian) to balance their admixtures because of Northern shifted Turkish_Balkan avg.")



Greek samples are from Lazaridis. He registered them as "Thessaloniki" but in another article that he was mentioned author states samples collected from between Eastern Macedonia and Thrace. Includes possibly Pontic mix too, with this sense. User data will always be more reliable than academic samples that collected because of one specific aim but not for general analyze.

Great answer, thank you. Yes, it gives some clues and as you said and some other stuff like matches or oracle's support it.

I understood. I agree with you, the user data is more reliable.