The following K runs were used by Dienekes to uncover the various admixture components:
K=3 uncovered the West Eurasian, Sub-Saharan, and East/South Asian components.
K=4 uncovered a split of Caucasoid into a component centered on Europe and West/South Asia.
Result: Established sufficient distance between South Asia and Sub-Saharan Africa.
K=5 uncovered the Mediterranean component.
K=6 uncovered the North African component.
K=7 uncovered the Southwest Asian component. [Caucasoid component in East Africa and Arabia] [Recent introduction to Northern Europe]
K=8 uncovered the South Asian component. This component removes the common variable found in South Asians and East Asians. The South Asian component represents the Caucasus. The rest of the Caucasus not included is split between West Asian and North/Central European components.
K=9 uncovered the Sardinian and Basque components of the West Mediterranean component.
K=10 uncovered the Northwestern and Northeastern divide of the North/Central European component. This Northwest component is a poor marker as it's highest in Orcadians, the British Isles, Scandinavia, France, Iberia, and Italy. Furthermore, balanced ratios of this Northwest/Northeast European dividing component is found among Greeks, Lezgins, Turks, and Iranians.
K=11 uncovered an Ethiopian/East African component.
Unfinished categorization:
- Uncover a component that puts more distance between Mediterranean and Southwest Asian.
- Establish a Western Mediterranean group that does not include the Sardinian component.
- Establish a component based on commonalities found in Northwest Africa and Iberia. [Berber]
- Establish a component based on commonalities found in North Africa and Sicily. [Saracen]
- Establish a component that puts more distance between the K=10 NW Europe and South Asia [Caucasus]. There is clearly an overlap that shows as NW European in Iran.
- Rename the South Asia and Southwest Asia components to better express their Caucasoid affinity.
Pretty pathetic message again. The tittle gives an idea how absurd is the claim all runs were made "to uncover ancestries". You must have serious issues to say such thing, it's time to be resonable and stop acting like insane when all academic comunity is telling you are wrong.
First of all, the lower K's are simply minor experiments to separate populations in a simple way, nothing significant in regards of full ancestry. Dienekes' precisely mentioned what I say in the 600 member milestone run, and he always explains the objective of a run...no need to use much brain to know wich analysis are relevant in terms of full ancestry, and wich are simply experiments with a concrete objective, very different from trying to find an accurate definition for individuals. So first point, don't be demagogic with the meaning of the lower K's, and don't forget the introduction of a new cluster does not mean this was the objective of the run.
And second, the higher K's, specially K=10, did not say any lie while scoring Northern European and Southern European, the data was quite aproximate. The separation between Northeast and Northwest does not imply a huge change, the same as the Mediterranean (Southern Europe). It is only a bit more regional, the same as the other K=12 with the Sardinian and Basque components, with very similar total European/African/Asian scores, but just in different regional categories. Slightly different results is what you obtain going from K=10 to K=12's (both higher K's, not minor experiments), nothing drastic as you dishonestly try to show alluding the more or less regional character of the analysis (determined by the Nº of clusters, the distances, and the way components are named).
Very soon, there'll be a K=14 run at Eurogenes (some people has made the DIY Teaser, me included) with the components already described in the blog. I guess you'll have a very hard work trying desperately to show masked ancestries there. Come on, it's time to move.
PD:Now some comments:
Uncover a component that puts more distance between Mediterranean and Southwest Asian.
- Actually the distance is enough to note what Mediterranean really means. West Asian is closer to Southwest Asain in the last run, so the pretension to put more distance has been acomplished comparing with the previous K=10 run. No significant change in any European population, except for Sardinians, who were nearly 98% European in the K=10 run (almost entirely Southern European), and now they score 85% (-13%). It clearly means this Mediterranean is a very good representation of the Southern European ancestry as whole, much better than the previous K=10. Absolutely nothing to argue here.
Establish a component based on commonalities found in Northwest Africa and Iberia. [Berber]
- That's nothing but your sick agenda, any normal researcher would think to do so, since the North African admixture between Iberians is perferctly listed and easy to see how low it is. The same is valid for Sicily, the distinction between African and other ancestries has been clarified exceedingly well. Perfectly clear wich are the common aspects, since actually North Africans as whole have more Euro and Asian (Caucasoid) ancestry than any European has African ancestry. Or perhaps you simply ignore the fact ethnic Berbers belong almost entirely to the Northwest African component, remaining distant from all Europeans. Plain silly pretension and, again, trying to go against all stated by clustering studies.
· Houston, we have a problem...LOL