New map of mtDNA haplogroup T2

[Maciamo]

Mt-haplogroup T2 is a relatively widespread, yet mysterious haplogroup. Its distribution doesn't give much hint regarding its origin. The frequency of T2 varies widely within linguistic families or countries that otherwise share similar ancestry, and even between regions of a same country. The only consistent pattern is a fairly elevated percentage in all Balto-Slavic countries (6.5% to 8.5%).

It is not even clear when T2 first reached Europe. T2 has been found in all major Neolithic cultures in Europe (Starčevo, LBK, Cucuteni-Trypillian, Cardium Pottery, Atlantic Megalithic...), but like haplogroup J it could have been present in Southeast Europe since the Mesolithic.

T2 is divided in many subclades from T2a to T2l. The largest by far is T2b, which has 30 subclades of its own, and many more deep clades. The former subclades T3, T4 and T5 are now all listed under T2 (respectively T2c1a, T2a1b and T2e).

The highest frequencies of T2 are observed among the Udmurts (23.8%), an Uralic people from the Volga-Ural region, and the Chechens (12.5%) in the North Caucasus. In the western half of Europe T2 exceeds 10% only in the Netherlands, Iceland, Aragon, Sardinia, Calabria and the Adriatic coast of Italy. T2 is virtually absent from Cornwall and from the Saami population.

T2 is present across Central and North Asia. It has been found in Bronze Age sites linked to the Indo-Europeans in Siberia, Kazakhstan and Mongolia. It is likely that T2 originated in the Near East or Caucasus, first entered Southeast Europe and the Pontic Steppe during the Mesolithic and/or Early Neolithic, then spread to the rest of Europe, Central Asia and North Asia during the Neolithic and the Bronze Age. I don't have enough data on the subclade distribution outside Europe, but it seems that at least some subclades of T2b and T2e are linked to the Indo-Europeans.



UPDATE: a detailed page about the origins, history, distribution and subclades of haplogroup T is now available here.

[MOESAN]

a so diffuse HG can very well confuse us, by instance in the scarce samples - it does not seem linked to any autosomals component - the diffusion in the Great Isles and W-Norway and the Netherlands could check some historical relatively recent events but... perhaps some link to Italics too? surely more than an event and maybe drift?

[Sennevini]

Netherlands, you keep surprising me :) I should test my mt-line (and Y-line) once which is Hollandic, following the records.

Overall, T2 looks like a mystery to me.

[Nobody1]

Why a map only just for T2 and not all of T - like with the J map;

Only other frequency map of T (all T and J) that i know of is from:

Pala et al 2012 -
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3376494/

[Maciamo]

Why a map only just for T2 and not all of T - like with the J map;

T2 is alreay diverse enough. I wish I had data for various subclades. Unfortunately I could only compile separate percentages by country for T1 and T2. The T1 map will follow soon. I don't have enough data for J subclades.

Likewise I will not make a map for the whole hg H as it wouldn't be more informative than all hg U. I am planning to make a map of H5 though.

[MOESAN]

thanks Maciamo for the huge work you do for these threads

[Nobody1]

T2 is alreay diverse enough. I wish I had data for various subclades. Unfortunately I could only compile separate percentages by country for T1 and T2. The T1 map will follow soon. I don't have enough data for J subclades.

Likewise I will not make a map for the whole hg H as it wouldn't be more informative than all hg U. I am planning to make a map of H5 though.

Sounds great and i agree detailed is always clearer (if possible);
Maps are awesome;

[Joe17]

Pre Pottery Neolithic B Syria mtDNA=16, T2b=3 18.75%, one T2b from 7,400-6,700bc, two from 6,800-5,750bc. Yasinovatka, Ukraine Dnieper Donets culture mtDNA=10, T=2(5,616-5,482bc), LBK culture around 5,000bc Central Europe mtDNA=66, T=14 21.21%, seven had identified subclade all were T2. Epicardiel culture Avellaner cave, Catalonia, Spain 5,000bc mtDNA=7, T2b=2. One T2b had Y DNA haplogroup which was G2a.

Rossen culture, Oberweiderstedt, Germany 4,625-4,250bc mtDNA=10 , T2=2. Mezzocorona, Trentino-South Tyrol Italy 4,444-4,326bc mtDNA=1: T2b3a=1, Navarre Spain mtDNA=37, T2=1(4,185-3,185bc). Camıi de Can Grau Granollers, Barcelona Spain 3,500-3,000bc mtDNA=11, T2=2.

mtDNA T2 was definitely very in Neolithic Europe and Middle east. Out of 101 mtDNA samples from Neolithic Germany 17.8% had T every single one with identified subclade was T2. In Neolithic, copper age, bronze age, and iron age mtDNA from Europe T is about at the rate it is today with T2 as the dominate subclade and most T2 being T2b and all T1 with subclade identified is T1a. To see how all mtDNA is distributed just look at FTDNA they have public results for many countries. It shows deep deep subclades of every haplogroup and there are many countries in Europe with 100's of samples some are very close to 2,000. So far I am shocked by the huge similarity in mtDNA haplogroups even deep subclades acroos Europe. Even though in Y DNA and autosomal DNA there are big differences between different Europeans. mtDNA from Neloithic-medival Europe shows pretty good constituency in my opinion with modern Europeans. But how did such deep subclade of for example mtDNA H1 and H3 become so widespread from Finland-Iberia-Greece-Russia-Ireland and some pretty deep subclades have already been found in around 7,000 year old Neolithic samples. It seems like modern Europeans mtDNA so maternal lineages is overwhelmingly from Neolithic farmers. It doesn't make sense though to say all those deep pretty unique European subclades of H, T, J, K, X, W, I, V all came from the Near east starting only about 9,000 years ago.


Even though so many argue that H1 and H3 are totally Palaeolithic Iberian. Europeans from Finland-Iberia I have seen in FTDNA share way to many deep subclades at around the same percentages for it to have spread before Neolithic. And the lack of almost any mtDNA H from European hunter gathers no matter it be from Palaeolithic, Mesolithic, Neolithic, or bronze age and H1 and H3 constantly popping up in Neolithic European farmers is even more evidence of a Neolithic spread. This is strange since Finland and Baltic have the highest amount of North Euro in Dodecade globe13 and K12b and highest Atlantic Baltic in K7b which all were dominate in European hunter gather samples La Brana(7,000ybp northwest Spain), Ajv52(4,000-4,800ybp Gotland, Sweden), AJV72(4,000-4,800ybp Gotland, Sweden). Therefore the most pre Neolithic ancestry in Europe but it seems they have mainly Neolithic maternal lineages like everyone else in Europe including Sardinia who have nearly identical results in K12b, globe13, and K7b to samples from Neolithic and copper age European farmers. Y DNA G2a and E1b1b V13 both already found in Neolithic farming European samples are almost completely absent from all of Scandinavia, Finland, and Baltic.

It really doesn't make any sense North Euro component in autosomal DNA is very popular in Europe and is from pre Neolithic European hunter gathers so far ancient DNA has shown were nearly 100% U5, U2e, and U4. Sardinia people who have like I said nearly identical results to Neolithic and copper age European farmer samples have very similar mtDNA haplogroup percentages as Finnish who have overwhelmingly hunter gather ancestry in autosomal DNA. Why are farmer maternal lineages so popular but hunter gather ancestry still very high.

[Joe17]

Maciamo I don't see why you try to link everything to Indo Europeans? Even when there are samples of haplogroups you say are connected with Indo Europeans from non Indo European Neolithic cultures or even Mesolithic and Palaeolithic. Also why do you generalize Indo Europeans? They didn't come from Mars they derived from humans on earth. I am sure different early Indo European speakers had very different ancestry and since they were farmers probably more typical mtDNA haplogroups of European farmers.

How do we know your maps are accurate? I see the spreadsheets from Dodecade you make your autosomal DNA maps off of. There is not nearly enough info to put all of the detail you have. It seems you predict how groups are distributed with no results to prove it. The spreadsheets don't say what part of the country the results came from. I don't know how you make your Y DNA maps I know at least some of it is just predictions. I do believe your maps are basically accurate and as accurate as possible with the data available. It seems sometimes you try to match your maps with certain theory's for example Gedorsian especially in Scandinavia it looks like you tried to match it with R1b map.

[ElHorsto]

Pre Pottery Neolithic B Syria mtDNA=16, T2b=3 18.75%, one T2b from 7,400-6,700bc, two from 6,800-5,750bc. Yasinovatka, Ukraine Dnieper Donets culture mtDNA=10, T=2(5,616-5,482bc), LBK culture around 5,000bc Central Europe mtDNA=66, T=14 21.21%, seven had identified subclade all were T2. Epicardiel culture Avellaner cave, Catalonia, Spain 5,000bc mtDNA=7, T2b=2. One T2b had Y DNA haplogroup which was G2a.

Rossen culture, Oberweiderstedt, Germany 4,625-4,250bc mtDNA=10 , T2=2. Mezzocorona, Trentino-South Tyrol Italy 4,444-4,326bc mtDNA=1: T2b3a=1, Navarre Spain mtDNA=37, T2=1(4,185-3,185bc). Camıi de Can Grau Granollers, Barcelona Spain 3,500-3,000bc mtDNA=11, T2=2.

mtDNA T2 was definitely very in Neolithic Europe and Middle east. Out of 101 mtDNA samples from Neolithic Germany 17.8% had T every single one with identified subclade was T2. In Neolithic, copper age, bronze age, and iron age mtDNA from Europe T is about at the rate it is today with T2 as the dominate subclade and most T2 being T2b and all T1 with subclade identified is T1a. To see how all mtDNA is distributed just look at FTDNA they have public results for many countries. It shows deep deep subclades of every haplogroup and there are many countries in Europe with 100's of samples some are very close to 2,000. So far I am shocked by the huge similarity in mtDNA haplogroups even deep subclades acroos Europe. Even though in Y DNA and autosomal DNA there are big differences between different Europeans. mtDNA from Neloithic-medival Europe shows pretty good constituency in my opinion with modern Europeans. But how did such deep subclade of for example mtDNA H1 and H3 become so widespread from Finland-Iberia-Greece-Russia-Ireland and some pretty deep subclades have already been found in around 7,000 year old Neolithic samples. It seems like modern Europeans mtDNA so maternal lineages is overwhelmingly from Neolithic farmers. It doesn't make sense though to say all those deep pretty unique European subclades of H, T, J, K, X, W, I, V all came from the Near east starting only about 9,000 years ago.


Even though so many argue that H1 and H3 are totally Palaeolithic Iberian. Europeans from Finland-Iberia I have seen in FTDNA share way to many deep subclades at around the same percentages for it to have spread before Neolithic. And the lack of almost any mtDNA H from European hunter gathers no matter it be from Palaeolithic, Mesolithic, Neolithic, or bronze age and H1 and H3 constantly popping up in Neolithic European farmers is even more evidence of a Neolithic spread. This is strange since Finland and Baltic have the highest amount of North Euro in Dodecade globe13 and K12b and highest Atlantic Baltic in K7b which all were dominate in European hunter gather samples La Brana(7,000ybp northwest Spain), Ajv52(4,000-4,800ybp Gotland, Sweden), AJV72(4,000-4,800ybp Gotland, Sweden). Therefore the most pre Neolithic ancestry in Europe but it seems they have mainly Neolithic maternal lineages like everyone else in Europe including Sardinia who have nearly identical results in K12b, globe13, and K7b to samples from Neolithic and copper age European farmers. Y DNA G2a and E1b1b V13 both already found in Neolithic farming European samples are almost completely absent from all of Scandinavia, Finland, and Baltic.

It really doesn't make any sense North Euro component in autosomal DNA is very popular in Europe and is from pre Neolithic European hunter gathers so far ancient DNA has shown were nearly 100% U5, U2e, and U4. Sardinia people who have like I said nearly identical results to Neolithic and copper age European farmer samples have very similar mtDNA haplogroup percentages as Finnish who have overwhelmingly hunter gather ancestry in autosomal DNA. Why are farmer maternal lineages so popular but hunter gather ancestry still very high.

Interesting post, but very difficult to comprehend.
I bet that H1 and H3 started to spread by mesolithic or epipaleolithic hunter-gatherers from the mediterranean or near-east who were autosomally different from the actual "Cro-Magnon" La Brana, Ajv etc. They were autosomally more close to the neolithic farmers who came subsequently and who replaced most of the H1 and H3 in the east mediterranean, leaving the most H1 + H3 remnants on the atlantic coast up to the baltic shores alive today. These two hunter-gatherer autosomals would have mixed most in western europe.

[Nobody1]

T2 is alreay diverse enough. I wish I had data for various subclades. Unfortunately I could only compile separate percentages by country for T1 and T2. The T1 map will follow soon. I don't have enough data for J subclades.

Likewise I will not make a map for the whole hg H as it wouldn't be more informative than all hg U. I am planning to make a map of H5 though.

there is data (not sure if enough) on J sub-clades J1 and J2 (+ more) in Pala et al 2012 -
Supplemental Data - Document S2 (Table S3/S4)
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3376494/

[MOESAN]

for me mt DNA seems less subject to drift than Y-DNA, but nevertheless could underwent some drift, very more than global autosomals - so, even if rare, sometimes, a population could show AND Y-DNA AND mt-DNA different for what say autosomals even if it does not correspond to evidence or intuition - at prehistoric times, populations wee very scarce and then under hazard influence more than today populations - and mt DNA is also under selection pressure, if what I red is true; even this last observation is sometimes without worth to explicate local differences - mt T seems to me neolithical but who knows?
all the way mt DNA seems vey mixed in European populations, often shared at very low frequencies and submitted to hazard sometimes: here a subclade is strong when an other of same phylum is weak, just in the region beside it is just the contrary, as by statistical "compensation", when History does not show any evident difference so...

[Angela]

for me mt DNA seems less subject to drift than Y-DNA, but nevertheless could underwent some drift, very more than global autosomals - so, even if rare, sometimes, a population could show AND Y-DNA AND mt-DNA different for what say autosomals even if it does not correspond to evidence or intuition - at prehistoric times, populations wee very scarce and then under hazard influence more than today populations - and mt DNA is also under selection pressure, if what I red is true; even this last observation is sometimes without worth to explicate local differences - mt T seems to me neolithical but who knows?
all the way mt DNA seems vey mixed in European populations, often shared at very low frequencies and submitted to hazard sometimes: here a subclade is strong when an other of same phylum is weak, just in the region beside it is just the contrary, as by statistical "compensation", when History does not show any evident difference so...

I completely agree. When I first became interested in population genetics, it was all about drift...then it went out of fashion, and everything could be explained by migrations. Well, after seeing more and more results, it's my personal opinion that mtDNA and yDNA are indeed much more volatile and subject to drift (and yDNA more than mtDNA) than autosomal dna, and mtDNA is also subject to selective pressures. We may also find out that there are selective pressures involving the y-chromosome, perhaps having to do with sperm counts or motility, or, based on some research I've seen, a certain combination of yDNA and mtDNA may result in greater fertility.

That's why I think some of these interpretations of the mtDNA data from central Europe which posit huge population replacements based on changes in the mtDNA are overblown, not to mention that the data doesn't even support that in my opinion.

Just my two cents.:)

[Bollox79]

My maternal group was assigned as T2b2b per National genographic Geno 2.0... haven't had all the mtDNA testing yet... but according to FTDNA origins and their kits that are T2b2b - it's 3 times as common in Ireland as anywhere else (2nd is Scotland) and that lines up well with my maternal line, which is from somewhere in the SW or W of Ireland . The grandmother had a surname that was also a Gaelic forename... or so I read about it. So I suppose the sub groups all have different stories!

[alef37]

Awesome map.

This is the strangest thing I found on taking my DNA tests.

I have a heritage since the '500s from the same small region in Northern Italy. My Y haplogroup is the R-U152 very common throughout NorthWestern Italy, Switzerland and France.

My maternal haplogroup is T2c1d1. I read everything about those concentrations in Russia and in the Caucasus. That's quite strange! Of course, I can't trace back my ancestry on a genealogical level before the middle-ages.. this haplogroup still is quite a mystery to me.

[Gilgamesh]

Maciamo, many thanks for this useful Eupedia page, i am carrier of a quite rare T2 subclade (T2f1b TMRCA :3300 ybp), defined recently, looks like mostly North East Slavic subclade.