The supplemental material from the recent Auton et al. article, which I blogged about here, is now available for download free of charge. It includes a wide variety of STRUCTURE diagrams, PCAs and tables featuring the POPRES and HapMap datasets.

One of the most interesting is a global STRUCTURE analysis showing the results for each of the European countries sampled (see below). At first glance it's really hard to make out anything of note within this closely related region, but some patterns and differences emerge if you look close enough.

It seems the third and fourth bars from top (K4 and K5) best indicate inter-continental admixture, both ancient and more recent. Light green is South/Central Asian and shows up in Turks, Cypriots, Italians, Romanians, Yugoslavs and apparently even the Czechs. The few dark green smudges amongst the Europeans are obviously from West Africa, and it appears the Portuguese carry the highest amount of this element, relatively speaking. Or maybe I'm just imagining things there?

More light green shows up at K6 (second bar from bottom), and then also varying amounts of purple at K7. This is either noise, or it could be the regional clusters breaking up and showing substructure. In any case, it's certainly not admixture across the major continental groups. The most likely reason why none of that purple appears in the HapMap samples (CEU, CHB, JPT and YRI) is because they were genotyped on a different platform to the POPRES dataset.

Here's the global diagram plus another one termed "Mexican admixture", which focuses on the genetic relationship between Eurasia and Mexico. The files are as clear as I could get them, even more so than in the original PDF, so hopefully you'll be able to make out at least some detail without going blind (click to enlarge).



Sample codes: 1 = Albania; 2 = Australia; 3 = Austria; 4 = Belgium; 5 = Bosnia-Herzegovina; 6 = Bulgaria; 7 = Canada; 8 = Croatia; 9 = Cyprus; 10 = Czech Republic; 11 = Denmark; 12 = Finland; 13 = France; 14 = Germany; 15 = Greece; 16 = Hungary; 17 = Ireland; 18 = Italy; 19 = Kosovo; 20 = Latvia; 21 = Macedonia; 22 = Netherlands; 23 = Norway; 24 = Poland; 25 = Portugal; 26 = Romania; 27 = Russia; 28 = Scotland; 29 = Serbia; 30 = Slovakia; 32 = Spain; 33 = Sweden; 34 = Swiss French; 35 = Swiss German; 36 = Swiss Italian; 37 = Switzerland; 38 = Turkey; 39 = Ukraine; 40 = UK; 41 = USA; 42 = Former Yugoslavia; a = Bengali; b = English; c = Gujarati; d = Hindi; e = language unknown; f = Punjabi; g = Pushto; h = Urdu; i = Konkani; j = Tamil; k = Malayalam; l = Sinhalese.




source: http://eurogenes.blogspot.com/2009/0...be-take-2.html