Southern Europeans' being mistaken as MENA in Northern Europe is simply a lie!!!

[Komintasavalta]

A whole paper is dedicated to demonstrate that PCAs (and therefore genetic distances) aren't accurate or meaningfull but this guy simply wants to deny it to support his point.

If Euclidean distances calculated with G25 are squared, they're not that far off from f2 distances, except that G25 underestimates the distance to isolated drifted populations like Saami, Tubalars, and Kalashes. Also G25 seems to overestimate the distance to mixed populations with large effective population size, like Karakalpaks, Uzbeks, and Tatars.




Source: https://anthrogenica.com/showthread....830#post785830

[Komintasavalta]

Here's another plot which includes all modern populations that have identical names in G25 and 1240K+HO. I used SmartPCA to calculate FST (https://anthrogenica.com/showthread....l=1#post786269). I didn't square the G25 distances so the regression line is now curved.

Again G25 underestimates the distance to drifted populations like Kalashes. But I think the reason why Evens are so far above the regression line is that the Even samples in 1240K+HO have more Caucasoid ancestry than the Even samples in G25.



With unscaled coordinates, G25 underestimates the distance of Moroccans to SSAs (because PC1 is not given enough weight) and to Mongoloids (because PC2 is not given enough weight), so it produces this regression line with a long flat section between Europeans and East Asians:

[Damião de Góis]

Genetic distances on G25 aren't perfect but they certainly can be used as guideline. My top 3 populations are Andalalucia, Extremadura and Galicia, and i'm from southern Portugal (Portugal comes in 5th). Distances are small though, and it guessed me correctly in western Iberia.

Hamilcar just didn't like how far a north african found in Copper Age Iberia (which he was using to make a point) was from other Copper Age iberian samples. To put in perspective he was as distant as i am from north indians on G25. And that how modern iberians (which he claims are as foreign as the said north african compared to the Copper Age samples) are in fact a lot closer to Copper Age Iberians:

Distance to: Iberia_Central_CA:I6617
0.11211008 Portuguese
0.23542486 Iberia_Central_CA_Afr:I4246

Target: Iberia_Central_CA:I6617
Distance: 11.2084% / 0.11208388
98.8 Portuguese
1.2 Iberia_Central_CA_Afr


So yeah, now G25 is garbage.

[Token]

sorry damiao but genetic distance are not really meaningfull :

https://www.biorxiv.org/content/10.1...F6peBsHfsNavBo

the only difference is that overall portuguese like yourself have 5% more EEF ancestry than me.

Eran Elhaik's paper on the validity of PCA has zero chance of surviving peer-review (not for nothing nobody co-authored the paper). Elhaik's whole career since 2013 or so has been trying to prove that Ashkenazi Jews are actually Khazar converts (the famous Khazar hypothesis), which is obviously wrong to anyone with a basic knowledge of genetics and history. His methods have been criticized by dozens of other geneticists, this quote in particular is very insightful:

Marcus Feldman, director of Stanford’s Morrison Institute for Population and Resource Studies, has said that Elhaik is "just wrong", with regard to the Khazar hypothesis where, in Feldman's view, he "appears to be applying the statistics in a way that gives him different results from what everybody else has obtained from essentially similar data"

At this point he is attempting to create a smokescreen in which his Khazar hypothesis will have more chance of survival by attacking an important pillar of modern genetic analysis. Any vestige of objectivity in this guy has been lost.

[Hamilcar]

Eran Elhaik's paper on the validity of PCA has zero chance of surviving peer-review (not for nothing nobody co-authored the paper). Elhaik's whole career since 2013 or so has been trying to prove that Ashkenazi Jews are actually Khazar converts (the famous Khazar hypothesis), which is obviously wrong to anyone with a basic knowledge of genetics and history. His methods have been criticized by dozens of other geneticists, this quote in particular is very insightful:



At this point he is attempting to create a smokescreen in which his Khazar hypothesis will have more chance of survival by attacking an important pillar of modern genetic analysis. Any vestige of objectivity in this guy has been lost.

I see but can you pinpoint what is wrong in his demonstration for example ?

[Hamilcar]

Genetic distances on G25 aren't perfect but they certainly can be used as guideline. My top 3 populations are Andalalucia, Extremadura and Galicia, and i'm from southern Portugal (Portugal comes in 5th). Distances are small though, and it guessed me correctly in western Iberia.

Hamilcar just didn't like how far a north african found in Copper Age Iberia (which he was using to make a point) was from other Copper Age iberian samples. To put in perspective he was as distant as i am from north indians on G25. And that how modern iberians (which he claims are as foreign as the said north african compared to the Copper Age samples) are in fact a lot closer to Copper Age Iberians:

Distance to: Iberia_Central_CA:I6617
0.11211008 Portuguese
0.23542486 Iberia_Central_CA_Afr:I4246

Target: Iberia_Central_CA:I6617
Distance: 11.2084% / 0.11208388
98.8 Portuguese
1.2 Iberia_Central_CA_Afr


So yeah, now G25 is garbage.

Your argument doesn't even make sense since I never implied nor said that copper age iberians were NAs or similar to us. I simply said they had a different genetic profile in comparison to modern iberians which is true.

[Chocolate_Hound]

Levant and Maghreb have a lot of overlap with Southern Europe. North Africans are considered white by sub-Saharans, which is understandable from their perspective.

[lustermoo]

Levant and Maghreb have a lot of overlap with Southern Europe. North Africans are considered white by sub-Saharans, which is understandable from their perspective.

Not in Europe they aren’t by ssa

[Cristiano viejo]

Levant and Maghreb have a lot of overlap with Southern Europe. North Africans are considered white by sub-Saharans, which is understandable from their perspective.

So MENAs have overlap with South Europe because blacks think so...

[BeSustainable5903]

To be honest, even an untrained eye like mine can easily distinguish SEs from MENAs, they just do not look the same. Its not just skin colour, its the features that are distinct between the two population groups.

Obviously , MENAs are caucasoids, so there obviously is some superficial resemblance between some samples from both groups. But no one can rationally claim that both groups are regularly mistaken for each other in Europe or elsewhere. A Spaniard is distinct from a Moroccan, even to a layman with zero knowledge about physical anthropology.