Talk:Origin of the Albanians/Archive 8

E-V13 Figure restored
I restored the figure because it was pertinent to the article. This section is called genetic studies on Albanians not on Europeans. Almost 40% of the Albanians belong to E V13 Haplogroup and having that picture makes more sense. Aigest (talk) 09:07, 18 December 2017 (UTC)




 * I deleted it as it was outdated and no longer considered correct. It presents E-V13 has having a simple peak in Central Albania and declining as you regularly with distance from there. But this isn't correct as among Albanians there are differences regarding E-V13 frequency. Overall Albanians are probably 30-37% but in Kosovo it peaks at 45% (Pericic et al) but much lower frequencies have been seen for other Albanian regions-- Battaglia had Tirana at only 24% (that's lower than some regions of Bulgarians, Romanians and Greeks) and Ferri had Tosks as a whole with 28% for all E1b1b (see here : [], I am not a fan of Dienekes' claims about Albanian genetics but he is still convenient for a link), which is not all E-V13, while Ghegs had 41%. So you see that among Albanians it is much more common among Ghegs especially in Kosovo, and declines as you go South (I think the Tirana sample may have been heavily Tosk). Macedonian Albanians, also Gheg in the sample, had 35%.
 * Meanwhile, the Tosks actually have less E1b1b than Greeks, especially Greeks from the South-- there is a secondary peak in Southern Greece where it even reaches up to 40%. Zalloua 2008 had it at 44% -- that's almost as high as Kosovo, though admittedly his sample size of 44 could've been bigger.
 * Also the map is horribly wrong for many parts of the Balkans especially concerning Bulgarians and Romanians. Here, you can see the distribution of hte haplogroup in Bulgaria -- [] . Although the map portrays Bulgarians as 5-10% E-V13 this is hilariously false-- Bulgarians overall are nearly 20%. Bulgarians from Razgrad, if we forgive the small sub-sample size, are 38% (!) percent. The map has E-V13 decreasing as you go east in Bulgaria but the opposite is true-- the western parts of Bulgaria have the least E-V13. As for Romania, I've blabbed a lot already, but at least Wallachia is also like 20% ([])-- not 5-10%. There's tons of other errors. Overall the map is badly outdated and misleading and I would be fine with keeping it if it were correct but it is very much not correct. --Calthinus (talk) 17:56, 18 December 2017 (UTC)


 * P.S. although not really relevant its badly wrong in Western Europe too. There's a peak in Marseille, France. The current explanation last I saw in the literature for that is that it's due to historical Greek not Albanian settlement in the area. Meanwhile there are random towns in Britain which have high percentages and this is attributed to "Illyrian soldiers". It seems like it is time for a new map. Maybe I'll make it myself. --Calthinus (talk) 18:02, 18 December 2017 (UTC)
 * While map may its own little problems, in general it is ok. This map is not about peak numbers but overall percentage on population. While there are cases of peak small areas (villages or cities) in total that would mean nothing. While in Albania the total percentage of people belonging to E-V13 is 33% (45% in Kosovo) in Bulgaria the total percentage of people belonging to E-V13 is 16 %, in continental Greece 18%, in Serbia 18%, in Italy 6%, in France 4% etc. The map shows that exactly. Bests Aigest (talk) 08:19, 19 December 2017 (UTC)
 * Well as you can see there is some wide variance on test results for Albanians in Albania (max 21.6% in Semino, 23.6% in Battaglia 2008, up to 33% in Samo 2015), and even more for Greeks. But I guess it's a question of whether you want to show the bigger picture clearly or accurately depict the sometimes patchy spread of the haplogroup per each region. I guess the map is trying to do the former. You can readd it if you want. I'm going to make an alternative this coming year -- let me know what you think when I do. --Calthinus (talk) 16:00, 19 December 2017 (UTC)