How come that Latvians are the most northern by autosomal genetics, but not Estonians or Finns?
https://i.imgur.com/jOKp7tr.jpeg
https://i.imgur.com/Rr2Rf6N.jpeg
Printable View
How come that Latvians are the most northern by autosomal genetics, but not Estonians or Finns?
https://i.imgur.com/jOKp7tr.jpeg
https://i.imgur.com/Rr2Rf6N.jpeg
Because that PCA is not ordered with respect to the cardinal points, even though it may seem so because it may resemble that idea we have about each town represented there.
Let PCA for a second, and look at the models.Quote:
Originally Posted by gixajo
Estonians can be modeled as 94% Latvians + 6% Maltese, but not the opposite case.
Seems Latvians are indeed the most northern.
These are the limitations inherent in the model and the instrument we are using to represent reality.Quote:
Originally Posted by Dušan
To make that plot cloud more in line with reality (which wouldn't be reality either), it should be rotated to the left until the NW references are at the same height as the NE ones.
NW folks are not much northern than Balts. Not at all.Quote:
Originally Posted by gixajo
Target: Icelandic
Distance: 6.4774% / 0.06477407
66.8 Latvian
33.2 Maltese
But in opposite case cannot be modeled.
Target: Latvian
Distance: 9.1645% / 0.09164497
100.0 Icelandic
Code:Icelandic,0.1326987,0.1336267,0.067819,0.0563097,0.0393663,0.0229619,0.0033488,0.0085188,0.0026418,6.07e-05,-0.0054806,0.0051702,-0.0063675,-0.0090831,0.0230725,0.0052924,-0.0116909,0.0019109,0.0017389,0.0038873,0.0092545,0.0028028,-0.0008422,0.0140179,0.0008282
Latvian,0.1354494,0.122473,0.0887742,0.089471,0.043208,0.0340246,0.0125966,0.0140762,-0.002168,-0.0368842,-0.0035728,-0.0138178,0.0232208,0.031378,-0.010559,0.0040572,0.0027118,-0.0003802,0.0016088,0.0043022,-0.0027952,-0.0057128,0.0090708,-0.0093264,0.0026346
Maltese,0.0919124,0.1464901,-0.0024985,-0.0434031,0.0132714,-0.0178141,-0.0044064,-0.0009231,0.0110952,0.0205244,0.0061504,0.0009739,-0.0028989,-0.0007226,-0.0042581,0.0008784,-0.0004074,-0.0037688,-0.0004871,-0.0023606,-0.0061454,-0.0025968,0.0028192,0.0011899,0.002904
Germanic-type admixture
Target: Estonian
Distance: 0.8785% / 0.00878467 | R3P
73.8 Latvian
21.2 Swedish
5.0 Mari
Target: Finnish_Central
Distance: 1.6539% / 0.01653862 | R3P
50.4 Latvian
35.6 Icelandic
14.0 Komi_B
Let's be practical; let's not start arguing about who's more northern than whom—it's pointless.Quote:
Originally Posted by Dušan
If you think that a Baltic-like NE component would be ideal for representing the North in Europe, you'd be better off basing on ancient references (Baltic BA or similar), to model modern ones.
I have made a 2-references model based in what you say in this thread.Quote:
Originally Posted by Dušan
Distance to: CDG_scaledQuote:
Originally Posted by Dušan
0.02186695 Icelandic
0.09926659 Latvian
0.13755828 Maltese
Target: CDG_scaled
Distance: 1.9471% / 0.01947080
93.2 Icelandic
6.8 Maltese
3 models (0 ms):
0.019 CDG_scaled = 93% Icelandic + 7% Maltese
0.022 CDG_scaled = 100% Icelandic + 0% Latvian
0.065 CDG_scaled = 62% Latvian + 38% Maltese
It´s just there are 2 tpes of northerners, Western and Eastern ones.(and 2 types of southerners, the easterners and the westerners.)Quote:
Originally Posted by Dušan
It's the typical "four corners of Europe" model that works so well, and of which I am proud to be the original creator.:p
It has to be taken this way, because both have old extras that add nuances that prevent one or the other from being seen as "the most representative of the north".
I think that's the fairest way to look at it. The PCAs and results of the G25 depend on and vary with the references we use; you see, if I use the references I'm interested in, I can become the most authentic Croatian on the forum.;)
Try it: Real Versus Fake Croatian(Scaled G25)
Since you're Balkanite, you'll get good distances. :rolleyes: