Latin American genetic studies [official archive]

[BirdMan]

I'm the only one transparent enough who has provided ALL the individual results for the samples I used.


So as much as I like Birdman, until he doesn't do the same, I'll have to discard those results.

I've been away for a while so I've been busy, but now that I'm back I'll try to add more because 20 isn't that many.

I blacked out the names/gedmatch numbers of the people I found on gedmatch because I'm wary of violating people's privacy. I can PM you individual numbers if you're interested.

More or less similar to 23andMe results, with the bulk of unassigned seeming to go to non-Euro caucasoid categories.

Here is the updated spreadsheet, it took longer than I hoped but I managed to add 14 more Puerto Ricans.



If anyone is interested here are the averages based on those 20 people. It'd be interesting to see this as a dot on the plot.

Group Average
Caucasoid Total 77.3
Euro Caucasoid 63.7
Non0Euro Caucasoid 13.7
SSA 8.7
Western Hemisphere 13.9

Amerindian 12.8
Arabian 1.2
Armenian 0.3
Basque 4.4
Central_African 0.4
Central_Euro 2.8
East_African 1.1
East_Asian 0.1
East_Balkan 1.5
East_Central_Asian 0.3
East_Central_Euro 1.5
East_Med 1.9
Eastern_Euro 1.0
Fennoscandian 1.6
French 4.1
Iberian 18.8
Indo0Chinese 0.1
Italian 12.7
Malayan 0.0
Near_Eastern 2.7
North_African 4.9
North_Atlantic 5.8
North_Caucasian 0.3
North_Sea 4.8
Northeast_African 1.4
Oceanian 0.1
Omotic 0.3
Pygmy 0.4
Siberian 0.4
South_Asian 0.1
South_Central_Asian 0.1
South_Chinese 0.0
Volga0Ural 0.2
West_African 6.5
West_Caucasian 0.8
West_Med 4.6

[Costas]

I've been away for a while so I've been busy, but now that I'm back I'll try to add more because 20 isn't that many.

I blacked out the names/gedmatch numbers of the people I found on gedmatch because I'm wary of violating people's privacy. I can PM you individual numbers if you're interested.

More or less similar to 23andMe results, with the bulk of unassigned seeming to go to non-Euro caucasoid categories.

Yay, thank you for posting them!

Yeah it is indeed very Italian.

As you say maybe it's the amount of samples.

Which ones do you have as Euro Caucasoid and which ones as non-Euro Caucasoid? I wanna calculate the same

[Costas]

i already did and added puerto rico too. the prolem is that k36 has so many componentes that its impossible to compare. For example i wouldnt score more italianj than you would, but i would score more east med and west caucaasian. Thats how hard it is to read the differences.

So do you have the comparison for the 4 countries anyway?

[BirdMan]

Yay, thank you for posting them!

Yeah it is indeed very Italian.

As you say maybe it's the amount of samples.

Which ones do you have as Euro Caucasoid and which ones as non-Euro Caucasoid? I wanna calculate the same

I actually posted them with the original average, but you must have missed it

Not sure, why they score so much Italian. As Argentano says there seems to be a big degree of overlap between the two groups, so I guess I wouldn't read too much into it. I doubt many of them are heavily Corsican/Italian. Maybe it just represents more Eastern or more Southern Iberian types (Mallorcans and Canarians being two of the largest groups to immigrate to PR).


I'll add more samples though, and see if anything changes.

I based the groupings on this post by Litvin, which seemed like a good way to divide things.

You could make a 4-components PCA based on Eurogenes K36:

1) European Caucasoid = sum of these 12 admixtures:

Basque
Central_Euro
East_Balkan
Eastern_Euro
East_Central_Euro
Fennoscandian
French
Iberian
Italian
North_Atlantic
North_Sea
West_Med

2) Non-Euro Caucasoid = sum of these 11 admixtures:

Arabian
Armenian
East_Med
Near_Eastern
North_African
Northeast_African
North_Caucasian
South_Asian
South_Central_Asian
Volga-Ural
West_Caucasian

3) Sub-Saharan African = sum of these 5 admixtures:

Central_African
East_African
Omotic
Pygmy
West_African

4) Western Hemisphere = sum of these 8 admixtures:

Amerindian
East_Asian
East_Central_Asian
Indo-Chinese
Malayan
Siberian
South_Chinese
Oceanian

Reference populations for each admixture are listed here:

https://docs.google.com/file/d/0B9o3...xQdGpZdW8/edit

E.g. "East_Med" is based on Cypriots, Druzes, Lebanese Christians.

[Argentano]

I actually posted them with the original average, but you must have missed it

Not sure, why they score so much Italian. As Argentano says there seems to be a big degree of overlap between the two groups, so I guess I wouldn't read too much into it. I doubt many of them are heavily Corsican/Italian. Maybe it just represents more Eastern or more Southern Iberian types (Mallorcans and Canarians being two of the largest groups to immigrate to PR).


I'll add more samples though, and see if anything changes.

I based the groupings on this post by Litvin, which seemed like a good way to divide things.

So do you have the comparison for the 4 countries anyway?

Here it is. I did it in a hurry let me know if there is a typo .

Anyway i dont like it much, too many componentes make it hard to compare plus iberian isnt iberian, italian isnt italian, etc.

[Costas]

For Chile

Caucasian 71.36
Amerindian 26.47
SSA 1.24

Euro Caucasian 59.48
non-Euro Caucasian 11.89

[Costas]

Here it is. I did it in a hurry let me know if there is a typo .

Anyway i dont like it much, too many componentes make it hard to compare plus iberian isnt iberian, italian isnt italian, etc.

Good one!!! Thanks

[BirdMan]

Here it is. I did it in a hurry let me know if there is a typo .

Anyway i dont like it much, too many componentes make it hard to compare plus iberian isnt iberian, italian isnt italian, etc.

Something really interesting to look at is the Gedcom + DNA matches.

That way if you put in a colonial #, it will only show you relatives of theirs with genealogical information attached. In most cases, it will also show place of birth if you click on the profile number.

I have a lot of free time tonight, so I was planning on adding many more samples with genealogical info for a more interesting comparison.

The only thing is, like you, I don't love k36 espescially for Latinos. My favorite is MDLP k23b, so I think I'll make my updated chart (hopefully with as many samples as yours + regional info) using that calculator. I espescially like it because on their spreadsheet they have many Spanish groups to compare the Latino results with. I'm open to using another one though, if anyone has suggestions.

[Argentano]

Something really interesting to look at is the Gedcom + DNA matches.

That way if you put in a colonial #, it will only show you relatives of theirs with genealogical information attached. In most cases, it will also show place of birth if you click on the profile number.

I have a lot of free time tonight, so I was planning on adding many more samples with genealogical info for a more interesting comparison.

The only thing is, like you, I don't love k36 espescially for Latinos. My favorite is MDLP k23b, so I think I'll make my updated chart (hopefully with as many samples as yours + regional info) using that calculator. I espescially like it because on their spreadsheet they have many Spanish groups to compare the Latino results with. I'm open to using another one though, if anyone has suggestions.

para decime esto en español q no entendi una mierda

Se puede ver lugar de nacimiento como?

[Taiguaitiaoghyrmmumin]

I've been away for a while so I've been busy, but now that I'm back I'll try to add more because 20 isn't that many.

I blacked out the names/gedmatch numbers of the people I found on gedmatch because I'm wary of violating people's privacy. I can PM you individual numbers if you're interested.

More or less similar to 23andMe results, with the bulk of unassigned seeming to go to non-Euro caucasoid categories.

whos the one that scored 20 percent Amerindian? Thats not normal its a bit outside the normal range for puerto ricans. Puerto ricans highest normal is like 17%
One of their parents likely had higher Amerindian then them in the 30s or high 20s