2
IDK if you really can call me a "Romanian user", but what I wanted to note was the people usually point to one amateurishly made PCA by davidski based on muh k15 calc. It basically has little to no scientific value because the way it's made is also sketchy. Usually PCA's are made by plotting the biggest genome difference at genes where there are a big difference..... Yeah, I won't even try to explain it since I myself don't fully understand it, but what's important to note is that it's;
1. Made from a genome
2. Never accurate due to minor differences being not included
K15 PCA is sorta like plotting a genome into a ADMIXTURE-like K=15 algorithm, then placing the 15 components on a PCA (IDK how they came up with the locations for these), then you plot in a calculator the amount of all the different components and then you get the co-ordinates from the calculations. I'm not a mathematician so IDK the formula, but I think you get the gist of it.
Now, even though professionally made PCA are not perfect (Behar is infamous to Romanians for including at least two non-ethnic Romanians as "Romanians" in the genetic study), they are still far better, and let's just look at some.....
Now, yeah they are more southern plotting because their northern neighbours the magyars were shifted northeast and later northwest perhaps by magyar then germanic and west slavic admixture. I'm basically saying that if you want to cover the diversity of Romanians (all of them including Moldovans) it should be that some are even south of the Bulgarian average, some are near Croats, some are west of Serbs, some are northeast of Serbs. I'm not trying to say that some Romanians are like Hungarians or anything, but we can't really know since no one has done any real proper research on autosomal Romanian DNA. Davidski amateur calcs. don't cut if for me, those calcs. can be manipulated a lot more than you think, some Romanians get Bulgarian closest on K15, but Croats closest on k36 for example.
Bookmarks