Wednesday, December 25, 2013

My Eurogenes K15 results

For more information about Polako's tool :

Eurogenes K15 now at GEDmatch

 

 - Admix Results (sorted) :

#PopulationPercent
1 Atlantic 36.45
2 West_Med 22.18
3 North_Sea 18.29
4 Baltic 10.14
5 East_Med 8.70
6 Eastern_Euro 2.23
7 Red_Sea 1.01
8 Amerindian 1.00

Too many components to be honest, the only studies in which I'm interested are those that divide one's genomes into components which really amount for a true migratory wave (for instance, hunter-gatherer vs farmer).

To compare with : Population averages


 - Oracle results  :

a) Using 1 population approximation :


1 Southwest_French @ 5.588
2 Spanish_Aragon @ 6.086
3 Spanish_Valencia @ 6.877
4 Spanish_Cantabria @ 7.812
5 Spanish_Castilla_La_Mancha @ 7.914
6 Spanish_Castilla_Y_Leon @ 8.905
7 Spanish_Andalucia @ 9.086
8 Spanish_Cataluna @ 9.362
9 Spanish_Murcia @ 9.956
10 Spanish_Extremadura @ 11.328

The novelty is that Polako has included a sample from SW France that I suppose originates from a 2012 study which included Gascon samples from Bigorre, Béarn and Chalosse.

Pre-Roman genetic structure has persisted in modern Basque populations

Unsurprisingly enough, my best Oracle guess are my fellow SW French people that should precisely be dubbed Gascon as we have yet to see Languedocian samples for instance.

Then the usual Iberians, in a somehow similar order from Dienekes' project : Cantabrians, Aragonese people and Valencians.

Polako's Eurogenes K15 project doesn't include Spanish Basques : they were my best guesses in DODECAD K12 runs.

It is being confirmed the classical French Basque sample used by most studies and also used by Polako is very specific, maybe inbred as my distance seems to be quite important to them, not even in the top 10.

 

b) Using 2 populations approximation :


1 50% Southwest_French +50% Spanish_Aragon @ 5.231
2 50% Southwest_French +50% Spanish_Valencia @ 5.463
3 50% Southwest_French +50% Southwest_French @ 5.588
4 50% French_Basque +50% Spanish_Castilla_Y_Leon @ 5.769
5 50% French_Basque +50% Spanish_Cataluna @ 5.813
6 50% Spanish_Aragon +50% Spanish_Aragon @ 6.086
7 50% Southwest_French +50% Spanish_Castilla_La_Mancha @ 6.156
8 50% French_Basque +50% Spanish_Murcia @ 6.200
9 50% Spanish_Aragon +50% Spanish_Valencia @ 6.200
10 50% French_Basque +50% Portuguese @ 6.245

The French Basque sample appears once the algorithm is forced into finding another sample to decypher me.

 
c) Using 4 populations approximation :

1 French_Basque + Southwest_French + Spanish_Castilla_Y_Leon + Spanish_Valencia @ 4.854
2 French_Basque + Southwest_French + Spanish_Valencia + Spanish_Valencia @ 4.940
3 French_Basque + Southwest_French + Spanish_Cataluna + Spanish_Valencia @ 4.954
4 French_Basque + Southwest_French + Southwest_French + Spanish_Castilla_Y_Leon @ 4.984
5 French_Basque + Southwest_French + Southwest_French + Spanish_Valencia @ 5.004
6 French_Basque + Southwest_French + Southwest_French + Spanish_Murcia @ 5.078
7 French_Basque + Southwest_French + Southwest_French + Spanish_Cataluna @ 5.086
8 French_Basque + Southwest_French + Spanish_Aragon + Spanish_Castilla_Y_Leon @ 5.111
9 French_Basque + Southwest_French + Spanish_Castilla_Y_Leon + Spanish_Castilla_Y_Leon @ 5.140
10 French_Basque + Southwest_French + Spanish_Murcia + Spanish_Valencia @ 5.145
11 French_Basque + Southwest_French + Spanish_Aragon + Spanish_Cataluna @ 5.163
12 French_Basque + Southwest_French + Spanish_Castilla_Y_Leon + Spanish_Cataluna @ 5.188
13 French_Basque + Spanish_Castilla_Y_Leon + Spanish_Valencia + Spanish_Valencia @ 5.196
14 French_Basque + Southwest_French + Spanish_Aragon + Spanish_Valencia @ 5.212
15 Southwest_French + Southwest_French + Spanish_Aragon + Spanish_Aragon @ 5.231
16 French_Basque + Southwest_French + Spanish_Andalucia + Spanish_Castilla_Y_Leon @ 5.250
17 Southwest_French + Southwest_French + Southwest_French + Spanish_Aragon @ 5.254
18 Southwest_French + Southwest_French + Spanish_Aragon + Spanish_Valencia @ 5.261
19 French_Basque + Spanish_Cataluna + Spanish_Valencia + Spanish_Valencia @ 5.284
20 French_Basque + Southwest_French + Spanish_Andalucia + Spanish_Cataluna @ 5.285

A side note on "Ancient human genomes suggest three ancestral populations for present-day Europeans"

You might have read the newest article about the peopling of Europe.


New data has emerged : unsurprisingly enough, a Neolithic farmer from Stuttgart clusters with other Neolithic farmers (Otzi and so on) whereas a Mesolithic hunter-gatherer from modern-day Luxembourg clusters with La Braña samples in Iberia, from Mesolithic times as well.

Figure B

We now know that modern-day Europeans were born from the mixing of aboriginal hunter-gatherers and farmers coming from the Levant. The novelty of this article is that the authors believe a third layer is to be identified that they dub "Ancient North Eurasians" : the mixing may have appeared prior to the arrival of farmers.

I'm not competent enough on that issue to tell you whether they are right or not.I don't doubt ancient hunter-gatherers mixed with Siberians but I'm rather clueless about everything else.

What triggered my curiosity was the presence of a "French_South" sample popping for the first time in an official scientific study.

I checked in tthe supplementary information : that sample is made of 7 individuals (Lat : 43.4 Long : -0.6) and the source is : David Comas. Well, that's Béarn for you, my home area.

It explains the ADMIXTURE results quite well and why those "southern French" individuals appear to be so similar to Basque people (lacking in the yellow component at its maximum in the Caucasus). I believe dubbing such specific sample "French_South" is thus quite misleading.