U4/U5 and Central Asia


Posts: 794
Joined: Wed Mar 14, 2012 3:50 pm
Location: Anthrogenica
PostPosted: Sat Jun 30, 2012 3:05 pm
U2e is the European daughter of the South Asian U2. Soares estimated U2e as 16,700 years old, and Behar at 19,290 years old, either of which put it well before the appearance of farming. But of course it didn't die out after that. U2e appears in remains from archaeological cultures of west and east linked to the Indo-Europeans: Bell Beaker and Andronovo.

This map of the distribution of mtDNA haplogroup U2e from 23andMe shows patches of it in Central Asia. The highest level seems to be around Tajikistan, but the map does not give country boundaries. Another patch seems to be somewhere in/near the Tarim Basin.

MapDNA14.jpg
MapDNA14.jpg (13.24 KiB) Viewed 5139 times

Posts: 116
Joined: Tue Mar 20, 2012 3:38 am
PostPosted: Mon Jul 02, 2012 3:58 am
JeanM wrote:The oldest definite U5 from ancient DNA was from Erralla (Gipuzkoa), Spain - 12,300 BP, but the estimated date of origin from Behar et al 2012 is 30,248 BP, which seems reasonable to me. U5 appears to have arisen somewhere in Europe well before the Ice Age, because it appears to have sheltered in several of the Ice Age refugia and then spread out from there when the climate improved. U5b probably spread mainly from the Franco-Cantabrian refuge. Since the distribution of U5a is weighted towards Eastern Europe, it may have sheltered in an Ice Age refuge in south-eastern Europe.


There is a difference of about 6000 years in the ages estimates of Behar and Soares for U5, with Soares et al estimating U5 to be 36,000 (with an uncertainty ranges of 25300 to 47200). Behar et al have a larger sample than Soares, but we also have a large sample in the U5 project and my age estimates are close to those of Soares et al. The full genome sequence of a 7000 year old sample that is U5b2c1 seems to suggest the older date could be more plausible.

The difference of 6000 years is important when we consider possible histories during the last glacial maximum. If the Soares et al estimates are more nearly correct, U5a and U5b might have been part of a larger family of U5 daughters that lived in Europe before the LGM. If Behar's estimates are more accurate, U5a and U5b might have evolved in the refugia after the ice had already advance. Given the very slow growth in the U5 tree until U5a1, U5a2, U5b1, U5b2 and U5b3, I think it seems likely that Soares age estimates are more consistent with a contracting U5 population and then more rapid growth after U5 repopulated Europe.

It would be great to get mtDNA tests in Europe dating to from 20,000 to 30,000 years ago to confirm that U5 was in Europe and to better estimate the ages of U5a and U5b.

We have a fairly small number of Asian U5 full genome test results, and several of those are in "young" branches of U5 suggesting migrations to the east in the last 10,000 years. However there is also a U5b2 that seems to be unique to to India and perhaps that suggest some branches of U5 migrated to South Asia at an earlier date, perhaps during the last glacial maximum.

In any case, I assume that Sykes;s choice of "Delphi, Greece" as an origin of U5 was a work of speculative imagination designed to inspire interest in the subject.

Gail

Posts: 160
Joined: Sat Mar 31, 2012 2:36 pm
PostPosted: Mon Jul 02, 2012 12:47 pm
GailT wrote:
JeanM wrote:The oldest definite U5 from ancient DNA was from Erralla (Gipuzkoa), Spain - 12,300 BP, but the estimated date of origin from Behar et al 2012 is 30,248 BP, which seems reasonable to me. U5 appears to have arisen somewhere in Europe well before the Ice Age, because it appears to have sheltered in several of the Ice Age refugia and then spread out from there when the climate improved. U5b probably spread mainly from the Franco-Cantabrian refuge. Since the distribution of U5a is weighted towards Eastern Europe, it may have sheltered in an Ice Age refuge in south-eastern Europe.


There is a difference of about 6000 years in the ages estimates of Behar and Soares for U5, with Soares et al estimating U5 to be 36,000 (with an uncertainty ranges of 25300 to 47200). Behar et al have a larger sample than Soares, but we also have a large sample in the U5 project and my age estimates are close to those of Soares et al. The full genome sequence of a 7000 year old sample that is U5b2c1 seems to suggest the older date could be more plausible.

The difference of 6000 years is important when we consider possible histories during the last glacial maximum. If the Soares et al estimates are more nearly correct, U5a and U5b might have been part of a larger family of U5 daughters that lived in Europe before the LGM. If Behar's estimates are more accurate, U5a and U5b might have evolved in the refugia after the ice had already advance. Given the very slow growth in the U5 tree until U5a1, U5a2, U5b1, U5b2 and U5b3, I think it seems likely that Soares age estimates are more consistent with a contracting U5 population and then more rapid growth after U5 repopulated Europe.

It would be great to get mtDNA tests in Europe dating to from 20,000 to 30,000 years ago to confirm that U5 was in Europe and to better estimate the ages of U5a and U5b.

We have a fairly small number of Asian U5 full genome test results, and several of those are in "young" branches of U5 suggesting migrations to the east in the last 10,000 years. However there is also a U5b2 that seems to be unique to to India and perhaps that suggest some branches of U5 migrated to South Asia at an earlier date, perhaps during the last glacial maximum.

In any case, I assume that Sykes;s choice of "Delphi, Greece" as an origin of U5 was a work of speculative imagination designed to inspire interest in the subject.

Gail


Is that U5b2 branch found mainly in NW South Asia or somewhere more South? What about in Central Asia? And what ydnas/population movement would have brought U5b2?

Posts: 794
Joined: Wed Mar 14, 2012 3:50 pm
Location: Anthrogenica
PostPosted: Mon Jul 02, 2012 12:53 pm
GailT wrote:
In any case, I assume that Sykes's choice of "Delphi, Greece" as an origin of U5 was a work of speculative imagination designed to inspire interest in the subject.


Sorry I missed that mention of Sykes earlier and did not respond to it. I couldn't agree more. All of the speculation on "clan" origins in Sykes, The Seven Daughters of Eve (2001) is worthless frankly, apart from the recognition that mtDNA haplogroup J appears to have entered Europe from the Near East with farmers.

Posts: 794
Joined: Wed Mar 14, 2012 3:50 pm
Location: Anthrogenica
PostPosted: Mon Jul 02, 2012 1:01 pm
newtoboard wrote:Is that U5b2 branch found mainly in NW South Asia or somewhere more South? What about in Central Asia?


You can see here the results of the Mitochondrial DNA Haplogroup U5 Family Tree DNA Project. There is also a map, which reveals four project members in India (one of whom is U52b) and four in Central Asia. You can load the map with all members, then click on each marker to see the details of that member. The U52b member in India is alone in U5b2* Group D - proposed new subclade. The oldest known female line ancestor is given as Shamani - 1888-1980, location Jetpur, Gujarat.

Posts: 160
Joined: Sat Mar 31, 2012 2:36 pm
PostPosted: Mon Jul 02, 2012 1:58 pm
JeanM wrote:
newtoboard wrote:Is that U5b2 branch found mainly in NW South Asia or somewhere more South? What about in Central Asia?


You can see here the results of the Mitochondrial DNA Haplogroup U5 Family Tree DNA Project. There is also a map, which reveals four project members in India (one of whom is U52b) and four in Central Asia. You can load the map with all members, then click on each marker to see the details of that member. The U52b member in India is alone in U5b2* Group D - proposed new subclade. The oldest known female line ancestor is given as Shamani - 1888-1980, location Jetpur, Gujarat.


The "Central Asian" matches seem to be in China with distant ancestors having European names.

What ydna came with the South Asian U5b2 though?

Posts: 160
Joined: Sat Mar 31, 2012 2:36 pm
PostPosted: Mon Jul 02, 2012 10:10 pm
So regarding mtdna U4 in the Kalash it is also accompanied by mtdna J2(j2b?). The Mansi and Khanty people have this too. Is this evidence of Uralic admixture? As far as I know neither Keltiminar and Botai were Uralic and Uralic speakers were always restricted to the area North of Kazakhstan. I don't find the theory of pre IE Uralic hunter gatherers in the area since South Central Asia probably had a R2 ydna populations followed by a Neolithic one from Iran and the Northern hunter gatherers who might have migrated to Kazakhstan migrated along the steepe corridor and would have been more similar to eastern Europeans not south from the Taiga.

Posts: 116
Joined: Tue Mar 20, 2012 3:38 am
PostPosted: Tue Jul 03, 2012 6:43 pm
Thanks, Jean, for linking to the map. The 4 markers in India are correctly located, but the 4 in China are actually people of European ancestry who incorrectly entered their map coordinates, I'll try to get them to correct it.

The samples in the project are mostly from people with European ancestry so the map does not reflect the true geographic diversity of U5. When we look at GenBank test results we see a wider distribution to the east.

Newtoboard - we cannot link y-DNA and mtDNA haplogroups at those early dates. We will need ancient y-DNA testing which is not available at older dates.

Gail
Previous

Return to Haplogroup U (mtDNA)

Who is online

Users browsing this forum: No registered users and 1 guest