New mtDNA & Y-DNA frequencies for the Basques

[Maciamo]

I have added mtDNA frequencies for the Basques, based on this study featuring 615 samples. The Basques stand out from the rest of Europe by their exceptionally high frequency of haplogroup H (61.5%, including 44% of H1 and H3) and Europe's lowest percentage of haplogroup T (1%). They only have 2% for IWX combined, which is also the lowest frequency in Europe. This indeed suggest a very limited influx of maternal lineages from the Caucasus (where X might have originated), as well as the North Caucasus and Pontic-Caspian steppes (where W and T are especially common). Unfortunately I don't have the breakdown of U subclades, but I expect to find very little if any U2, U3 and U4, all of which are characteristic of the Black Sea region and the Pontic-Caspian steppes.

The paucity of T is especially telling. Almost every European population has approximately 10% of hg T. The three exceptions are the Finns (4%), the Bosniaks (4%), and the Basques (1%). The Finns and Bosniaks have the lowest percentage of R1b in Europe, both under 5%. That the Basques, who have the highest frequency of R1b in Europe, should have the lowest incidence of mtDNA T is surprising to say the least. Unless, as I have theorised a few years ago, the Basques were conquered by a group of R1b men, who killed most of the local men and procreated with their women. This would lead to a society where the vast majority of the male lineages are foreign (R1b) but almost all the maternal lineages remain indigenous. That also explains why the Basque kept their pre-IE language, as children are more likely to learn their mother tongue, well from their mothers...


I have also revised the Basques' Y-DNA frequencies using four different sources (Underhill et al., Adams et al., Iberianroots and the study in link above) totalling 597 samples. There are only a few changes, but important ones. I2a decreased from 9% to 5% to the profit of E1b1b (increase from 1% to 2.5%) and G2a, which had 0% and now has 1.5%. We now have a pre-IE admixture suggesting a considerable West Asian admixture, since the total of G2a, J1 and J2 is 4.5%, about the same as the Paleolithic I2a1 (5%). The big question mark is E1b1b (2.5%), which would be Paleolithic as well as Neolithic, or even an influence of neighbouring Cantabria.

There are a few more oddities among the Basques. Two studies for traces of haplogroup Q (3 samples in total, so 0.5%). Iberianroots listed an individual who belonged to hg H, while the study in link above has one hg L (I double-checked the STR values, and the haplogroup assignment is correct). H + L ? How did those South Asian haplogroups end up there ? Did Q come along through the same migration from the Levant (Phoenicians, Neolithic farmers) ?

[Wilhelm]

I have also revised the Basques' Y-DNA frequencies using four different sources (Underhill et al., Adams et al., Iberianroots and the study in link above) totalling 597 samples. There are only a few changes, but important ones. I2a decreased from 9% to 5% to the profit of E1b1b (increase from 1% to 2.5%) and G2a, which had 0% and now has 1.5%. We now have a pre-IE admixture suggesting a considerable West Asian admixture, since the total of G2a, J1 and J2 is 4.5%, about the same as the Paleolithic I2a1 (5%). The big question mark is E1b1b (2.5%), which would be Paleolithic as well as Neolithic, or even an influence of neighbouring Cantabria.

Makes me wonder if the basque samples included people with non full basque ancestry, which is today quite common.

There are a few more oddities among the Basques. Two studies for traces of haplogroup Q (3 samples in total, so 0.5%). Iberianroots listed an individual who belonged to hg H, while the study in link above has one hg L (I double-checked the STR values, and the haplogroup assignment is correct). H + L ? How did those South Asian haplogroups end up there ? Did Q come along through the same migration from the Levant (Phoenicians, Neolithic farmers) ?

There are also gypsies in the basque country, they speak their own langauge called Erromintxela. So, again, makes me doubt about the validity of such samples. Seems like these so-calles scientists are eagar for exoticness, instead of going to the pure racial stock of the people in question.

[Maciamo]

There are also gypsies in the basque country, they speak their own langauge called Erromintxela. So, again, makes me doubt about the validity of such samples. Seems like these so-calles scientists are eagar for exoticness, instead of going to the pure racial stock of the people in question.

Well, actually the H and L samples do not come from official studies by geneticists, by Iberianroots (commercial sample) and the doctoral thesis of Kristin Leigh Young in link in the OP. In the latter, the paper says "DNA samples were collected in mountain villages throughout the four Basque provinces of northern Spain by Dr. A. Apraiz, under the support of a National Geographic Society Grant to the University of Kansas Laboratory of Biological Anthropology." The L sample came from Gipuzkoa. However I am not aware of the presence of haplogroup L among the Gypsies. They normally belong exclusively to hg H1a (or European haplogroups).

[Knovas]

I said in the other post G2a wasn't probably absent between the Basques, and It seems I was right. 1.5% is slightly higher than expected (0.5%-1%), so I'm pretty sure the values will decrease a bit as more samples come. I also think I2a will come up,at least, to 7%. The actual percent it's too low.

About the rest, just minimal changes (meaningless), although the maternal linages are a great improve and very interesting. Considering all the information as whole about the Basques we know today, it doesn't surprise me they are the most purest Europeans.

Fascinating population.

[Asturrulumbo]

The much lesser amount of I2*/I2a seems to confirm to me that the original centre of Basque was more along the central Pyrenees (around today's Aragon, see map), which have a much higher concentration of I2*/I2a. That may be a boost to the theory that Iberian was related to Basque... Of course, y-DNA percentages in Navarre could help confirm that.

[Knovas]

Today it's quite obvious non Basque Iberians are related to them. The Dienekes' K=12 with Sardinian and Basques components, did not show less than 20% Basque ancestry between Spaniards. Focussing in the Northeast side, the percent was higher than 30%, and probably if we could check more Catalans (depending on the region), Aragonese, Navarrese, Riojans, Cantabrians...even Asturians, we could see very high percents.

We don't need an Y-DNA table. I share genomes with a Navarran woman at 23andme and she almost clusters with Basques. They are the most similar people.

[Wilhelm]

Some parts of Catalonia have hotspots of I2, specially in the Pyrenees. (9.5% in Bortzerriak, Navarra; 9.7% in Chazetania, Aragon; 8% in Val d'Aran, Catalunya; 2.9% in Alt Urgell, Catalunya; and 8.1% in Baixa Cerdanya, Catalunya)

[Knovas]

About haplogroup U, I supose Basques belong almost enterely to U5, more likely U5b subclades. Will see when the information comes in. I also wouldn't be surprised to find substantial U2e (Paleolithic) in comparison with the rest.

It's very interesting this source has collected 6% K, when for example 23andme says 4% (well, of course it's possible). I always wondered if my subclade could be quite present between the Basques, since it's very widespread in all Europe and perhaps they show higher levels than the average.

[Maciamo]

Behar et al. published The Basque Paradigm: Genetic Evidence of a Maternal Continuity in the Franco-Cantabrian Region since Pre-Neolithic Times in March 2012. The abstract says :

"We identified six mtDNA haplogroups, H1j1, H1t1, H2a5a1, H1av1, H3c2a, and H1e1a1, which are autochthonous to the Franco-Cantabrian region and, more specifically, to Basque-speaking populations. We detected signals of the expansion of these haplogroups at ∼4,000 years before present (YBP) and estimated their separation from the pan-European gene pool at ∼8,000 YBP, antedating the Indo-European arrival to the region. Our results clearly support the hypothesis of a partial genetic continuity of contemporary Basques with the preceding Paleolithic/Mesolithic settlers of their homeland."


The supplemental data contains the list of all the complete mtDNA tested with their haplogroup assignments and deep subclades for 420 hg H samples. The samples are not only Basque, but also Gascon and Castillan.

It is remarkable though that out of 129 H subclades, they could only associate 6 of them with the pre-Neolithic inhabitants of the Franco-Cantabrian region, especially since the samples contained 54 H1 subclades and 22 H3 subclades. It would be surprising if the other H1 and H3 subclades didn't originate in the region as well. I suppose that they meant that those six subclades already existed when Mesolithic people resettled western and central Europe after the last glacial period, while the others might only have emerged from the native H1* and H3* in the last 8000 years. New mutations happen all the time, so it is possible that some deep subclades only developed only in the last centuries, while others didn't change at all for thousands of years.