Continental ancestry of colonial latinos in the 40-65% euro range from Gedmatch

[Adamastor]

I am from Spain and in my k15 I score 0% SSA and 1,32% Northeast African.

I edited: i score 1,24% in k13, and 1,32 in k15

That is surprising actually. Weren't you fully Basque? Basques are supposed to score 0% SSA + Northeast African.

SSA and Northeast African in Europeans are signs of North African ancestry, so it will be higher in Iberians with more North African and lower/inexistent in individuals with no North African (like the Basques).

If you look at averages in Vahaduo, you would see that Basque average is 0% for both while Spanish_Galicia and Portuguese averages are like 3-4% for both components combined. It happens because in GEDmatch North Africans score 20% SSA, so if you're 1/8 North African you would score ~3% SSA + Northeast African. I generally use these components to calculate North African admixture in Iberians and Italians.

I'm not trying to say the Northeast African + SSA in Latin Americans is Iberian lol, what I'm saying is that Northeast African around the Iberian average (between 1.5% and 3%) likely comes from Iberians rather than from Blacks (of course it applies to people scoring it in lower amounts, not people who score 20%, for example). Canarians and Sephardic Jews score even higher Northeast African, around 5-6%.

[Argentano]

Yeah. I like Argentano, but he seems to have a poor grasp of what GEDmatch is made for. It's not designed to calculate percentages of recent European, Asian and African ancestries, but to dig up components of thousands of years ago. Many Europeans will score things like East_Asian, Siberian and even SSA (in the case of Iberians/South Italians).

If a Latin American is, let's say, 90% Turkish and 10% Amerindian, that Latin American would come out as 80% Caucasoid and 20% Mongoloid in GEDmatch. The same for someone that is 90% Portuguese, Galician or Canarian and 10% SSA, that guy will come out as 85% or 87% European and 13% - 15% SSA. Almost all West Iberians and Canarians score over 2% SSA in GEDmatch.

Yes i understand this are ancient components . But so what? eurogenes k15 is still very good to compare ancestry between latin american countries. Because latin americans dont have turkish or russian blood with high siberian or whatever. They mostly have iberian blood. And iberians DONT score southeast asian, they dont score siberian, they dont score south asian and they dont score high SSA .

Iberians are around 1% SSA in eurogenes K15 . If a dominican or a cuban or a brazilian is scoring 40% SSA in eurogenes K15, that is clearly not iberian blood.

Eurogenes k15 Reference populations (Very similar to the kits i have found in gedmatch but i prefer to post this as "official")

https://docs.google.com/spreadsheets...it?usp=sharing

[Argentano]

I appreciate Argentano's work but unfortunately Eurogenes K15 and other GEDmatch calculators aren't powerful enough to deal with noisy data. It produces noisy results by overfitting the data, which is why Europeans and even the extremely drifted Amerindians can score small amounts of East Asian, South Asian, Siberian and other stuff that aren't really there. And this is why people like me can score over 2% South Asian (with no Balkan shift, so Romani ancestry is impossible). Five years is like a century in the fastly developing field of genetics.

but THIS is clearly not noise. This is a clear pattern

Here one very clear example to show how northeast african is totally related to SSA ancestry.

Northeast African component in Brazilians from gedmatch depending on their SSA range. The more SSA brazilians score, the higher the northeast African component too.

[Argentano]

with all due respect to token and adamastor, 2 posters i appreciate. I dont agree with you. You talk like iberian kits in gedmatch were "full of noise" in eurogenes k15 when they are clearly not. They dont score east asian, they dont score siberian, they dont score high northeast african and they dont score high SSA.

MOST of Those components in latin americans are clearly non white ancestry.

[Adamastor]

Yes i understand this are ancient components . But so what? eurogenes k15 is still very good to compare ancestry between latin american countries. Because latin americans dont have turkish or russian blood with high siberian or whatever. They mostly have iberian blood. And iberians DONT score southeast asian, they dont score siberian, they dont score south asian and they dont score high SSA .

Dude, sorry for being so harsh, but you clearly don't understand what you are saying. I'm not saying 40% SSA in a Cuban or Brazilian comes from Iberians, c'mon, you're smarter than that. Also ''Iberians'' vary a little bit depending on the region. They were supposed to not score Siberian, Southeast Asian etc. But many of them actually score those things in these calculators and it happens because they are designed for measuring things that go beyond recent ancestry from 500 years ago. Besides the fact GEDmatch opens an immense room for noise and lacks references to many ghost populations who got modeled and discovered in the last 6 years. Why do you think even the least SSA admixed North Africans are always 20% SSA there?

Actually the spreadsheet you posted corroborates what I said, Portuguese score ~3% SSA in there, Castillans around 2,7% and Galicians 2,5%, what is in line with the fact all of these guys are around 8-12% North African. So the SSA + Northeast African in West Iberians can range from 2% to 5% actually, given that the average is 2.5-3%. Canarians score even more, so modelling Cubans and Venezuelans the way you do is to miss a lot of stuff.

Canarian averages:



Davidski himself, the guy who created these Eurogenes calculators, said to me that GEDmatch isn't good to model Latin Americans and other recently admixed populations because it opens a lot of room for noise.

[Argentano]

Dude, sorry for being so harsh, but you clearly don't understand what you are saying. I'm not saying 40% SSA in a Cuban or Brazilian comes from Iberians, c'mon, you're smarter than that. Also ''Iberians'' vary a little bit depending on the region. They were supposed to not score Siberian, Southeast Asian etc. But many of them actually score those things in these calculators and it happens because they are designed for measuring things that go beyond recent ancestry from 500 years ago. Besides the fact GEDmatch opens an immense room for noise and lacks references to many ghost populations who got modeled and discovered in the last 6 years. Why do you think even the least SSA admixed North Africans are always 20% SSA there?

Actually the spreadsheet you posted corroborates what I said, Portuguese score ~3% SSA in there, Castillans around 2,7% and Galicians 2,5%, what is in line with the fact all of these guys are around 8-12% North African. So the SSA + Northeast African in West Iberians can range from 2% to 5% actually, given that the average is 2.5-3%. Canarians score even more, so modelling Cubans and Venezuelans the way you do is to miss a lot of stuff.

Canarian averages:

Davidski himself, the guy who created these Eurogenes calculators, said to me that GEDmatch isn't good to model Latin Americans and other recently admixed populations because it opens a lot of room for noise.

I call what you are saying bullshit. Show me proof of 20 or 30 iberians scoring relevant siberian, southeast asian , etc. Common prove me wrong.

You are just inventing things . SHOW ME PROOF. SHOW ME KITS, IBERIANS SCORING HIGH OF THOSE COMPONETS YOU MENTIONED. I WAIT.


How many iberians are there in gedmatch? 5.000 ? 20.000? Show me 40 scoring high siberian....


And obvipsuly we are talking about eurogenes k15. Everything i posted is on eurogenes k15. Dont answer me with data from another calculator...

[Adamastor]

I call what you are saying bullshit. Show me proof of 20 or 30 iberians scoring relevant siberian, southeast asian , etc. Common prove me wrong.

You are just inventing things . SHOW ME PROOF. SHOW ME KITS, IBERIANS SCORING HIGH OF THOSE COMPONETS YOU MENTIONED. I WAIT.

Jut look it up in this forum:

Gaditanian, Gallop, Spaniard 2.0 (I think that's his name), Pedro Ruben, FilhoV etc All of them score at least 2-3% East Eurasian in multiple GEDmatch calculators. I have more kits in my possession scoring beyond noise levels of these things (I can send you the kits). I can post even Sardinians with 2% East_Asian.

I score 1.76% East_Asian myself in K15 and 0% Amerindian (and I have no Amerindian ancestry), so there's no way it is real. The fact that GEDmatch isn't good for modelling proportions in Latin Americans is more than known by everyone in the genetics community. I'm not saying the variations will be huge, they will not, but they can be considerable.

A Cuban guy with 3 Canarian grandparents and a quadroon grandparent will seem more SSA than he really is using GEDmatch (and that's just one example).

[Token]

I call what you are saying bullshit. Show me proof of 20 or 30 iberians scoring relevant siberian, southeast asian , etc. Common prove me wrong.

You are just inventing things . SHOW ME PROOF. SHOW ME KITS, IBERIANS SCORING HIGH OF THOSE COMPONETS YOU MENTIONED. I WAIT.


How many iberians are there in gedmatch? 5.000 ? 20.000? Show me 40 scoring high siberian....


And obvipsuly we are talking about eurogenes k15. Everything i posted is on eurogenes k15. Dont answer me with data from another calculator...

So are you saying there is actual Southeast Asian, Siberian and South Asian admixture in Latin Americans? Come on.

[Argentano]

Jut look it up in this forum:

Gaditanian, Gallop, Spaniard 2.0 (I think that's his name), Pedro Ruben, FilhoV etc All of them score at least 2-3% East Eurasian in multiple GEDmatch calculators. I have more kits in my possession scoring beyond noise levels of these things (I can send you the kits). I can post even Sardinians with 2% East_Asian.

I score 1.76% East_Asian myself in K15 and 0% Amerindian (and I have no Amerindian ancestry), so there's no way it is real. The fact that GEDmatch isn't good for modelling proportions in Latin Americans is more than known by everyone in the genetics community. I'm not saying the variations will be huge, they will not, but they can be considerable.

A Cuban guy with 3 Canarian grandparents and a quadroon grandparent will seem more SSA than he really is using GEDmatch (and that's just one example).

you said there is a lot of noise in gedmatch. Iberians scoring high southeast asian, siberian, SSA in eurogenes k15. Post the kits. I wait.

You talked a lot. Now its your time to show proof of what you are saying.

I call what you are saying bullshit until you back that up. And i aksed for 40 kits. Since you say this noise thing is so constnt i guess you have seen hudreds of kits like that

[Argentano]

So are you saying there is actual Southeast Asian, Siberian and South Asian admixture in Latin Americans? Come on.

the one who score high siberian tend to be the ones with higher native american. Its probably part of the native american ancestry. Similar to the northeast african and SSA


Honestly i can expect this from adamastor because he/she is always claiming shit without proof. But it surprises me from you tbh