3 ms·
This is interesting, and yet I'm not quite sure what this map is portraying. Perhaps a read of the original paper is in order, but as usual there is no link. Ir
by mechanical_fish 18y ago
This is interesting, and yet I'm not quite sure what this map is portraying. Perhaps a read of the original paper is in order, but as usual there is no link. Ironically, this time it's not the journal's fault, because the full text is online here:
http://www.current-biology.com/content/article/fulltext?uid=PIIS0960982208009561 http://www.current-biology.com/content/article/fulltext?uid=...
The web celebrated its seventeenth birthday one week ago. Anyone want to bet how much longer it will be before the NYT editors figure out how it works?
Unfortunately, I don't have time to thoroughly read the original article, nor do I really have the stats skills to understand it all. I will note, though, that the plot on the left really does seem to represent the two most significant principal components of the genetics data. The fact that it resembles the map of Europe is not a deliberately arranged feature of the plot's design -- except to the extent that the authors oriented the axes so that Italy's group was toward the bottom and Britain's was to the left. The plot resembles the map of Europe because the clusters of related genes resemble the map of Europe.
Cool.
Interesting features: English and French genetics overlap a lot less than the languages (I guess that's what happens when English citizens learn the language of their French aristocratic rulers). However, no amateur linguist would be surprised to learn that English and Dutch people have much genetic overlap. Hungary's poor overlap with Austria is interesting, as is the huge genetic distance between Finland and Hungary, whose languages are famously related to each other and famously unrelated to most other languages. Makes me want to study Finnish history. And what is up with Romania?
[After I finished the article I went to change this link to a no-login link, courtesy of Aaron Swartz's NYT link converter, but I found it seemed to already be converted. I'm so confused.]