>>9139893the source is the Global25 dataset made by Eurogenes, it's not "academic" in itself but it's pretty much in line with academic data and created with publicly available academic datasets and populations, it's nothing more than a PCA done with smartpca, a publicly available program used in almost all papers about the topic
here for comparison is an academic global PCA, the topology is the same, though it's confusing because there's so much populations included, while with the dataset I can plot whatever I like
