PDA

View Full Version : Upper Paleolithic/Mesolithic European mtDNA Haplogroups



J Man
12-19-2012, 06:57 PM
It now seems that numerous different mtDNA haplogroups were represented in the ancient Mesolithic populations of Europe. Not that long ago many people (including myself) thought that only a few mtDNA haplgoroups were truly representative of the Mesolithic populations of Europe but it now appears that there was more mtDNA genetic variation in Europe during the Mesolithic than the results indicated just a few years ago. I will list the mtDNA haplogroups below that have been confirmed in European Upper Paleolithic and Mesolithic remains.

Upper Paleolithic:

U2e
U
U5
HV or R0

Mesolithic (some of these results are from samples from the Neolithic period but they all came from hunter-gatherer cultures):

U5a
U5b
U4
U4d
H
C1
C5
D
T
T2b
V
Z1a

Here is where I got this information from.

http://www.buildinghistory.org/distantpast/ancientdna.shtml

If i missed anything please feel free to correct me.

It seems then that many of the haplogroups present in Europeans today such as U2, U4, U5, H, T, T2b and V are also represented in either Upper Paleolithic or Mesolithic Europeans remains. T2b may be a Neolithic entrant to Europe though as it has also been found in early Neolithic remains from Syria. It was found in one Pitted Ware culture sample from Gotland that was a very late European hunter-gatherer culture. This may represent admixture between farmers and foragers. Hard to really say though as this stage. It also looks like some X, H, J, K and T have been found in both Mesolithic and Neolithic samples from Greece. I am not sure exactly which haplogroups were found in the Mesolithic samples and which ones were found in the Neolithic samples when it comes to the Greek results though.

http://dienekes.blogspot.ca/2012/12/talk-by-christina-papageorgopoulou-on.html

From the looks of all of this then it seems that many Europeans today may be descended at least matrilineally from the early hunter-gatherers of Upper Paleolithic and Mesolithic Europe. Obviously more testing will provide much more information in the future.

mothergoddess
06-23-2014, 06:59 PM
I m new to this site and to the genetics topic, so please excuse me for any errors related to this forum site. I have recently been genotyped as H13a1awhich, according to 23 and Me, stated was probably Paleolithic hunter-gather from the Caucasus Mt area. On a forum including Redhaired that was speaking of European H, and U and RO, they were debating Paleo-dna that contributed to modern Europeans. My ancestry from my parents were English, Irish. Scot, and German- western European. If H13a1a is from that time period, then my Gravettian ancestors certainly did contribute to this modern woman--any thoughts???? thanks

yxc
06-23-2014, 10:51 PM
H13a1a, contributed . but not before neolithic. H&V is in majority in many places in Europe today , before neolithic they could had been in europe too, what is the exactly extant of europe ! in glacial period there was grassland disrupted only by broad rivers ,all the way between Wales and Sunghir. Caspian sea had higher surface elevation ,extanted further north , and as long as it drained in Black Sea cut off Caucasus from Europe .R0 , HV were rather in most eastern part and not widespread like U5 , which was ~40ka already in most western europe.

yxc
06-25-2014, 09:51 PM
not widespread like U5 , which was ~40ka already in most western europe.
U5 just~30ka , the older human remains perhaps were other. i.e. U6

J Man
06-29-2014, 11:48 PM
Some H definitely was present in Europe during the Upper Paleolithic and Mesolithic.

Jean M
06-30-2014, 12:19 AM
Some H definitely was present in Europe during the Upper Paleolithic and Mesolithic.

Where do you get that idea? Certainly not from my collation of aDNA results, which is what you cite. It would be very strange indeed if H was present in Europe in the Upper Palaeolithic. Its estimated date is 12,846 years ago (Behar 2012) and it pretty clearly arose in the Near East.

GailT
06-30-2014, 12:54 AM
U5 just~30ka , the older human remains perhaps were other. i.e. U6

There are two U5 samples at Dolni Vestonice dated at 31 kya (and one U8 sample) but U5 is very likely older than this in Europe. These samples set the lower limit for the age, and it is possible that U5 is at least a few thousand years older than the lower limit. The only older remain that has been tested is U2 at Kostenki. I'd guess that older (untested) remains in Europe might be precursors to U5, U8, U2, or perhaps more likely, other extinct U lineages.

The Dolni Vestonice KC521458 was used to define U5 in the most recent Phylotree update, and U5a'b now represents the only known surviving lineage of U5.

jeanL
06-30-2014, 01:12 AM
Where do you get that idea? Certainly not from my collation of aDNA results, which is what you cite. It would be very strange indeed if H was present in Europe in the Upper Palaeolithic. Its estimated date is 12,846 years ago (Behar 2012) and it pretty clearly arose in the Near East.

Has H been found in Upper Paleolithic Near Eastern remains? Has HV its precursor been found in the Middle East ancient remains? It(mt-DNA HV and R0) has been found in European Upper Paleolithic samples. The Behar.et.al.2012 estimate, is just that an "estimate". Here is a more recent, and more refined estimate using ancient dna data from Fu.et.al.2013:

http://2.bp.blogspot.com/-71JYGXNrukY/UUyZ07GOx5I/AAAAAAAAItk/g4kxvVCtEgc/s1600/mtdna_ages.png

J Man
06-30-2014, 09:21 PM
Where do you get that idea? Certainly not from my collation of aDNA results, which is what you cite. It would be very strange indeed if H was present in Europe in the Upper Palaeolithic. Its estimated date is 12,846 years ago (Behar 2012) and it pretty clearly arose in the Near East.

From David Reich the geneticist. He told me via email that one of the samples of the newly tested Samara >6000 years (Russia) hunter-gatherer remains that should be in a paper within the next year was H. The rest were U types.

Jean M
06-30-2014, 09:26 PM
From David Reich the geneticist. He told me via email that one of the samples of the newly tested Samara >6000 years (Russia) hunter-gatherer remains that should be in a paper within the next year was H. The rest were U types.

Interesting. Delighted to learn that mtDNA is on the way. However this is not the Upper Palaeolithic (50,000-10,000 years ago) or anytime close. :) It is 4000 BC. H had arrived in Europe earlier and was around to be married.

We have been round these particular houses a number of times. The fact that a group is hunter-gatherer does not stop them marrying the farmer's daughter next door, so to speak. In order to prove that H arrived in Europe before the Neolithic, you would have to actually find it in Europe before the Neolithic.

J Man
06-30-2014, 11:54 PM
Interesting. Delighted to learn that mtDNA is on the way. However this is not the Upper Palaeolithic (50,000-10,000 years ago) or anytime close. :) It is 4000 BC. H had arrived in Europe earlier and was around to be married.

We have been round these particular houses a number of times. The fact that a group is hunter-gatherer does not stop them marrying the farmer's daughter next door, so to speak. In order to prove that H arrived in Europe before the Neolithic, you would have to actually find it in Europe before the Neolithic.

Yes well of course there is still the possibility that this H sample from Samara came from intermarriage with farmers but there is also the case of the confirmed H sample from Mesolithic Karelia. That could also be from intermarriage but this seems to be making for at least a slightly stronger case of some H in pre-Neolithic Europe. Maybe not Upper Paleolithic but maybe Mesolithic. Time and more testing of ancient remains will tell.

J Man
06-30-2014, 11:57 PM
I emailed him to, asking for some detailed results. Based on his response I think we probably won't see the paper or hear of any results for probably a very long time.

His response.



It is very comforting that another bunch of European hunter gatherer mtDNA samples are showing that maternally Europe was very uniform and undiverse(nothing complicated that takes years to discover).

I have great news for you. You should look at Table S4. of this (http://www.sciencemag.org/content/suppl/2013/10/10/342.6155.257.DC1/Brandt.tablesS1-S17.xls), there are many many new results(Notably multiple U5b and U5a1a'g types from Pitted ware) you should put into your Ancestral Journeys site.

The HV1, HV2, and CR mutations listed in table S4. are according to RSRS and back mutations are not listed. So you have to manually compare them to Phyoltree(can't use predictors because none are set for RSRS and back mutations are not listed) it is time consuming but worth it. I found that the three Upper Palaeolithic Gravettian samples from Dolni Vestonice Czech republic are actually U8c(with no informative extra mutations so maybe U8c*), U5*, and U5*. Also, there is a Solutrean sample on your site that is U5 but was reported as just U.

When did these results come in? I thought all of the Pitted Ware U samples were U5a and not U5b.

Humanist
07-01-2014, 01:14 AM
I have posted the secret source a few times online but now I am going to delete the posts...

Are you referring to Brandt et al. (http://www.sciencemag.org/content/342/6155/257.abstract)?

J Man
07-01-2014, 01:17 AM
I have posted the secret source a few times online but now I am going to delete the posts, because many of the very intelligent(cough cough Maju) people who research this stuff are .....(don't want to get an infraction). I have the source saved on a Google Drive, I'll give you access if you give me your Gmail. Not only was plain U5b found in the Pitted Ware culture one of the samples could be broken down to U5b2b1a. I have not compared the mutations of the possible H results for some Iberian hunter gatherers, but I did take a glance and it seemed likely some did have H.

It should not be a surprise that U5b existed in Scandinavian hunter gatherers because the first settlers of Scandinavia and the entire Baltic region had some western(Loschbour-like) hunter gatherer origins(mainly U5b) which explains why their paternal lineages were mainly under I2a1-P37. Culturally Scandinavian hunter gatherers had always had links and common origins with Baltic and eastern European hunter gatherers(explains ANE ancestry and U5a), and U5b has already been found in Polish and Lithuanian hunter gatherers.

Damn for some reason it won't let me send you a PM. I had problems with the PM system on here before as well.

Barellalee
07-01-2014, 01:21 AM
Damn for some reason it won't let me send you a PM. I had problems with the PM system on here before as well.

So p37.2 was present in early Scandinavian/Nordic territories?? My P37.2+L233 is said to have originated in the Germanic area 3000 ybp, but I was always perplexed by it, due to the general infrequent occurrence of p37.2 in NW Europe.

Humanist
07-01-2014, 01:23 AM
I have great news for you. I'll email it to you if you give me your email. There are many many new results(Notably multiple U5b and U5a1a'g types from Pitted ware) you should put into your Ancestral Journeys site.

Are you referring to the results listed on this page: Mesolithic Western Eurasian DNA (http://www.ancestraljourneys.org/mesolithicdna.shtml)

Mehrdad
07-01-2014, 01:35 AM
Perhaps mtDNA I should also be included in the mix of European Paleolithic and Mesolithic populations?

Humanist
07-01-2014, 02:01 AM
Yeah, and for some reason I have not seen any mention of them anywhere online..

Is this Dienekes article (http://dienekes.blogspot.com/2013/10/ancient-central-european-mtdna-across.html) not discussing the paper in question?

Jean M
07-01-2014, 08:37 AM
That could also be from intermarriage but this seems to be making for at least a slightly stronger case of some H in pre-Neolithic Europe.

For those who want to believe maybe. ;)

Jean M
07-01-2014, 08:43 AM
@ Brennus

I know about the Brandt 2013 study. You can see it in the bibliography at http://www.ancestraljourneys.org/bibliography.shtml . Most of the results from it are entered in my tables. I didn't get around to finishing the job, as it is a massive amount of data, and other things came along which needed to be added. But Brandt 2013 sticks to Germany and starts in the Neolithic, for their own results anyway.

Looking at it again, I see what you are talking about. For comparison, they use data on Pitted Ware etc from other studies, but they have updated the haplogroup allocations. I hadn't noticed that. Thank you.


I have not compared the mutations of the possible H results for some Iberian hunter gatherers, but I did take a glance and it seemed likely some did have H.

The Chandler study is unreliable. It was done on museum collections at a time when proper precautions against contamination were not taken. I do list the results in http://www.ancestraljourneys.org/mesolithicdna.shtml but with a warning.

J Man
07-01-2014, 10:01 AM
For those who want to believe maybe. ;)

When do you think H came to Europe?

Ebizur
07-01-2014, 10:25 AM
Yes well of course there is still the possibility that this H sample from Samara came from intermarriage with farmers but there is also the case of the confirmed H sample from Mesolithic Karelia. That could also be from intermarriage but this seems to be making for at least a slightly stronger case of some H in pre-Neolithic Europe. Maybe not Upper Paleolithic but maybe Mesolithic. Time and more testing of ancient remains will tell.
Karelia is at the heart of the western (Baltic) Finnic territories. The Samara River is located near the southern Ural Mountains and the Ural River, just north of western Kazakhstan, and historically has been an important feature in the territory of various Turko-Mongol nomads (e.g. Bashkirs, Nogays, Kalmyks). The latter area in particular is only marginally European (probably it was not even considered to be part of Europe in ancient historical times), and Karelia obviously has experienced some sort of influence from the east at some time. In my opinion, finding some cases of mtDNA haplogroup H in ancient remains from those areas is not strong proof of the presence of that haplogroup in indigenous populations of Palaeolithic or Mesolithic Europe.

Generalissimo
07-01-2014, 10:38 AM
Karelia is at the heart of the western (Baltic) Finnic territories. The Samara River is located near the southern Ural Mountains and the Ural River, just north of western Kazakhstan, and historically has been an important feature in the territory of various Turko-Mongol nomads (e.g. Bashkirs, Nogays, Kalmyks). The latter area in particular is only marginally European (probably it was not even considered to be part of Europe in ancient historical times), and Karelia obviously has experienced some sort of influence from the east at some time. In my opinion, finding some cases of mtDNA haplogroup H in ancient remains from those areas is not strong proof of the presence of that haplogroup in indigenous populations of Palaeolithic or Mesolithic Europe.

The Samara region was part of the same biogeographical zone as the rest of Northern Europe during the Mesolithic and probably Neolithic, because farming came to this region from the Balkans, not directly from Asia.

That's why the hunter-gatherers and farmers from the Samara valley resemble other Northern European hunter-gatherers and farmers.


The mtDNA composition of the steppe population is primarily West Eurasian, in contrast with northwest Russian samples of this period (Der Sarkissian et al. PLoS Genetics 2013) where an East Eurasian presence is evident.

Samara experienced major population turnovers over time: early samples (>6000 years) belong primarily to mtDNA haplogroups U4 and U5, typical of European hunter-gatherers but later ones include haplogroups W, H, T, I, K, J.

Link (https://mcidublin.conference-services.net/reports/template/onetextabstract.xml?xsl=template/onetextabstract.xsl&conferenceID=3958&abstractID=814052)

The Turks came much later from Asia.

Jean M
07-01-2014, 10:49 AM
@ Brennus. I have now updated, thanks. http://www.ancestraljourneys.org/mesolithicdna.shtml

Jean M
07-01-2014, 11:08 AM
When do you think H came to Europe?

It is starting to feel like we have had this conversation about a dozen times, but probably that is just me feeling a bit limp and lifeless. :D

It looks as though a bunch of mtDNA haplogroups, including H, arrived in Europe with farmers from the Near East. I cover this in AJ, so all the citations are there. One geneticist after another has come to that conclusion. I'm flowing with the tide. Here for example is an image from Brandt 2013:

2011

Click to enlarge.

Ebizur
07-01-2014, 11:47 AM
The Samara region was part of the same biogeographical zone as the rest of Northern Europe during the Mesolithic and probably Neolithic, because farming came to this region from the Balkans, not directly from Asia.

That's why the hunter-gatherers and farmers from the Samara valley resemble other Northern European hunter-gatherers and farmers.

The Turks came much later from Asia.
In that case, it should be easy to find a sample of haplogroup H mtDNA among Palaeolithic or Mesolithic remains from an authentically European place (Germany, France, Italy, etc.). Please let me know if you are aware of any such sample.

Generalissimo
07-01-2014, 11:59 AM
In that case, it should be easy to find a sample of haplogroup H mtDNA among Palaeolithic or Mesolithic remains from an authentically European place (Germany, France, Italy, etc.). Please let me know if you are aware of any such sample.

It probably won't be easy, because it seems like mtDNA H arrived in the Samara Valley during the Neolithic, with farmers from the Balkans. Before that this area was home to typical Northern European hunter-gatherers carrying mtDNA U4 and U5, although as per the link above, probably with much higher levels of ANE admixture than contemporaneous Central Europeans (close to 0% ANE) and even Scandinavians (15-19% of ANE).

Karelia and nearby regions of Fennoscandia and Northwestern Russia appear to have had a different population history than the rest of Northern Europe and the European steppe, with non-trivial frequencies of East Eurasian mtDNA haplogroups, and probably also clear levels of genome-wide East Eurasian or ENA (as opposed to ANE) admixture. So it's difficult to say where that Karelian Mesolithic mtDNA H lineage came from.

Ebizur
07-01-2014, 12:10 PM
It probably won't be easy, because it seems like mtDNA H arrived in the Samara Valley during the Neolithic, with farmers from the Balkans. Before that this area was home to typical Northern European hunter-gatherers carrying mtDNA U4 and U5, although as per the link above, probably with much higher levels of ANE admixture than contemporaneous Central Europeans (close to 0% ANE) and even Scandinavians (15-19% of ANE).
How is your hypothesis of an arrival of haplogroup H to the Samara river basin during the Neolithic, introduced by farmers from the Balkans relevant to determining which haplogroups were present among Upper Palaeolithic and Mesolithic Europeans as per the title of this thread? Are you suggesting that haplogroup H was present among Palaeolithic or Mesolithic peoples of SE Europe despite an apparent lack of this haplogroup in contemporary populations of Central or Western Europe?

Generalissimo
07-01-2014, 12:21 PM
How is your hypothesis of an arrival of haplogroup H to the Samara river basin during the Neolithic, introduced by farmers from the Balkans relevant to determining which haplogroups were present among Upper Palaeolithic and Mesolithic Europeans as per the title of this thread? Are you suggesting that haplogroup H was present among Palaeolithic or Mesolithic peoples of SE Europe despite an apparent lack of this haplogroup in contemporary populations of Central or Western Europe?

It's not relevant, but I felt it was useful to point out that the Samara Valley was not part of Asia in terms of genetic affinities until historic times. In fact, the people living there during the Mesolithic and Neolithic were in some ways more European than those in Western Europe at the time, because they had more indigenous European hunter-gatherer ancestry and also carried ANE, which was missing in Western Europe until the metal ages, but is now an important genetic component amongst Europeans.

J Man
07-01-2014, 01:10 PM
Well whatever you do dont go limp and lifeless on us. :p...As you already know a number of PPNB samples from the Near East have been tested for mtDNA and have been shown to be mainly part of haplogroup K. This of course is only one small group of early Neolithic farmers.

J Man
07-01-2014, 01:14 PM
It probably won't be easy, because it seems like mtDNA H arrived in the Samara Valley during the Neolithic, with farmers from the Balkans. Before that this area was home to typical Northern European hunter-gatherers carrying mtDNA U4 and U5, although as per the link above, probably with much higher levels of ANE admixture than contemporaneous Central Europeans (close to 0% ANE) and even Scandinavians (15-19% of ANE).

Karelia and nearby regions of Fennoscandia and Northwestern Russia appear to have had a different population history than the rest of Northern Europe and the European steppe, with non-trivial frequencies of East Eurasian mtDNA haplogroups, and probably also clear levels of genome-wide East Eurasian or ENA (as opposed to ANE) admixture. So it's difficult to say where that Karelian Mesolithic mtDNA H lineage came from.

Well David Reich told me that there is a single H sample among the early Samara hunter-gatherer samples. This plus the H sample from Mesolithic Karelia seem to add to the possibility of some H being present in Mesolithic Europe.

yxc
07-01-2014, 01:39 PM
Perhaps mtDNA I should also be included in the mix of European Paleolithic and Mesolithic populations?
N1 & I ? my guess is mtdna X instead.

Generalissimo
07-01-2014, 01:53 PM
Well David Reich told me that there is a single H sample among the early Samara hunter-gatherer samples. This plus the H sample from Mesolithic Karelia seem to add to the possibility of some H being present in Mesolithic Europe.

Did he say if there was any East Eurasian mtDNA among the hunter-gatherer Samara samples, like among the Mesolithic Karelians?

J Man
07-01-2014, 03:27 PM
Did he say if there was any East Eurasian mtDNA among the hunter-gatherer Samara samples, like among the Mesolithic Karelians?

No he did not.

Shaikorth
07-01-2014, 03:32 PM
I think it's safe to assume that the >6000 years old samples were the most "hunter-gatherer"-like and they were just U with one H. The mesolithic NW Russian samples were U, C and one H, the presence of C most likely implying differentiation between mesolithic (and perhaps general post-LGM) steppe and forest/arctic zone in Eastern Europe.

Jean M
07-01-2014, 04:06 PM
As you already know a number of PPNB samples from the Near East have been tested for mtDNA and have been shown to be mainly part of haplogroup K.

It doesn't matter what the percentages of H were in the PPNB samples http://www.ancestraljourneys.org/wasianneolithicdna.shtml . The mere fact that H was present does away with any idea that it was born in European Mesolithic. It's right there in the Near Eastern Neolithic and is not there in European Mesolithic (properly tested samples, guarding against contamination, etc.)

I'm really very busy J Man, so could I be excused now please? Pretty please? :angel:

J Man
07-01-2014, 08:36 PM
It doesn't matter what the percentages of H were in the PPNB samples http://www.ancestraljourneys.org/wasianneolithicdna.shtml . The mere fact that H was present does away with any idea that it was born in European Mesolithic. It's right there in the Near Eastern Neolithic and is not there in European Mesolithic (properly tested samples, guarding against contamination, etc.)

I'm really very busy J Man, so could I be excused now please? Pretty please? :angel:

There is no need for you to be rude. I know I asked the question before but that was quite some time ago. No need for that attitude. BTW there is also a U* sample from the PPNB results. BY that logic you could also say then that U was not born in Europe.

Jean M
07-01-2014, 09:20 PM
BTW there is also a U* sample from the PPNB results. BY that logic you could also say then that U was not born in Europe.

Of course U wasn't born in Europe. We find U6 in North Africa, and U2 types in India. It is likely that U was born somewhere around what I call the Asian crossroads. It seems likely that U5 was born in Europe, and U3 in the Near East.

2020

Click to enlarge.

J Man
07-01-2014, 09:53 PM
Of course U wasn't born in Europe. We find U6 in North Africa, and U2 types in India. It is likely that U was born somewhere around what I call the Asian crossroads. It seems likely that U5 was born in Europe, and U3 in the Near East.

2020

Click to enlarge.

Some subclades of H may have been born in Europe as well then.

Jean M
07-01-2014, 10:03 PM
Some subclades of H may have been born in Europe as well then.

That's pretty well certain I'd say, especially as we get into more recent ones. I'm not an expert on mtDNA H, so I won't start delving into detail. You can find plenty of discussion on the H forum http://www.anthrogenica.com/forumdisplay.php?80-H .

jeanL
07-01-2014, 10:51 PM
It doesn't matter what the percentages of H were in the PPNB samples http://www.ancestraljourneys.org/wasianneolithicdna.shtml . The mere fact that H was present does away with any idea that it was born in European Mesolithic. It's right there in the Near Eastern Neolithic and is not there in European Mesolithic (properly tested samples, guarding against contamination, etc.)

I'm really very busy J Man, so could I be excused now please? Pretty please? :angel:

H has been found in European samples: Karelia, La Chora(Upper Paleolithic), La Pasiega(Upper Paleolithic), now the Samara sample. So, why do you keep this charade up? Again, more and more evidence points to the presence of mt-DNA H in Europe pre-Neolithic. What's more, why haven't you contacted the authors of Hervella.et.al.2012 since you are after all misreporting their data. Anyone who sees your website would see La Chora reported as some rare form of U?, when it is in fact H6. Are you implying the authors falsified the data. The only thing ancient DNA has shown is that mt-DNA U was the dominant one on Hunter Gatherers samples, whereas mt-DNA H was a minority haplogroup. I have yet to see you arguing that mt-DNA C was a Neolithic arrival in Europe. I hate to get into the personal stuff, because it adds nothing to the scholarly debate, but you are outright denying evidence in favor of special pleading.

According to you: mt-DNA H arrived in the Neolithic in Europe, but its parent R0 or HV was found in La Pasiega, assuming that it is R0 or HV, and not H as it was tested with the RFLP. At the same time La Chora is some unknown type of U, and it is not H6. Karelia dating to 7500 ybp is because of a farmer who ended up way up there. Now Samara is because yet another farmer. Then we also have the parental clade of H found in Paglicci Cave dating 24,000 ybp with mt-DNA R0 or HV and CRS in HVR1, 7025 AluI, 00073A, 11719G, 12308A from Caramelli.et.al.2003. But once more, never mind the parental clade, H came from the Middle East.

Yet we have an article that shows that farmers and hunter gatherers living side by side in Germany did not mix for at least 2000 years, but you are damn sure that the Karelian and Samaran Hunter Gatherers sequences are of farmer origin (http://www.sciencedaily.com/releases/2013/10/131010142704.htm).

I'm the one that has grown rather tired of the same discussion.

PS: One has to love how you don't question the U5 from Eralla, Guipuzcoa reported by Hervella.et.al.2012 with 16270 HVR-I, but when it comes to La Chora and La Pasiega there is just no way that they could be H and H6. Double standard much???

GailT
07-01-2014, 11:52 PM
H has been found in European samples: Karelia, La Chora(Upper Paleolithic), La Pasiega(Upper Paleolithic), now the Samara sample. So, why do you keep this charade up?

There are no confirmed H samples that are free of contamination. Every year that goes by with more U and no H makes it more difficult to sustain the faith that H was present in Europe before the Neolithic. We can't rule out the possibility, but there is no evidence to support it.

parasar
07-02-2014, 12:00 AM
There are no confirmed H samples that are free of contamination. Every year that goes by with more U and no H makes it more difficult to sustain the faith that H was present in Europe before the Neolithic. We can't rule out the possibility, but there is no evidence to support it.

And how about the middle-east prior to the Neolithic?

jeanL
07-02-2014, 02:36 AM
There are no confirmed H samples that are free of contamination. Every year that goes by with more U and no H makes it more difficult to sustain the faith that H was present in Europe before the Neolithic. We can't rule out the possibility, but there is no evidence to support it.

That's a pretty bold statement considering:




Authenticity of the results

We have successfully analyzed 49 out of 54 individuals recovered from the nine prehistoric sites analyzed in the present study; four sites correspond to hunter-gatherer groups (La Chora, La Pasiega, Erralla and Aizpea), another four to the Neolithic period (Los Cascajos, Paternanbidea, Marizulo and Fuente Hoz) and the last one to the Bronze Age (Urtiaga). Out of these 49 individuals, 30 (61% of the individuals) were analysed in duplicate in our laboratory; in addition, a third sample from each of 22 individuals (44% of the individuals) was replicated at an independent laboratory (some in the University of La Laguna and others in the INTCF of Madrid, Spain).

The number of molecular targets (mtDNA copy number) was quantified for each extract by means of quantitative real-time PCR (qPCR). The results showed that the number of molecules per μl in the extracts ranged between 100–12,000 (Table S1). These values are within the limits proposed for reproducibility in aDNA studies [28].

In addition, a total of 69 PCR products from 49 individuals were cloned. A minimum of ten clones per PCR product were amplified and sequenced. The results were used to determine the coincidence between the consensus sequence obtained from the clones and the sequence obtained by direct sequencing of the initial PCR product. A mean of 4.20 mutations per fragment cloned (~100 bp) were rejected, as these mutations were found in single clones, possibly as consequence of post-mortem damage to the aDNA molecules (Table S2).

In order to identify any possible contamination that might have occurred in the different stages of the laboratory work, at least two extraction controls and several PCR negative controls were included in each amplification. If any contamination was detected, the results obtained were discarded.

Finally, the mtDNA HVR-I segment from the researchers and archaeologists that handled the samples was sequenced in order to rule out any possible contamination (Table S3). A comparison between the haplotypes of the researchers/archaeologists and those obtained in the aDNA samples (Table 2) produced two cases of coincidence, one with the rCRS haplotype, which is the most frequent one in European populations, and the other with haplogoup K (ht18: 92-224-311), which is widely present not only throughout the whole of the Iberian Peninsula [29] and the Canary Islands [30], but also Europe [9], [12], [13] and North Africa [31]. HVR-II sequencing was performed on those prehistoric samples that coincided with the haplotypes of the researchers and/or archaeologists in order to discard a complete coincidence between them.

Ancient DNA from Hunter-Gatherer and Farmer Groups from Northern Spain Supports a Random Dispersion Model for the Neolithic Expansion into Europe (http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0034417)




Authentication of the mtDNA data

Strict precautions were taken in order to minimize the risk of contamination by modern DNA and detect artefactual mutations arising from contamination and aDNA degradation. Seven criteria support the authenticity of the mtDNA data presented here.
1.Pre-PCR DNA work was carried out at the ACAD, a purpose-built laboratory dedicated to aDNA studies. The laboratory is under positive air-pressure and physically isolated from any molecular biology laboratory amplifying DNA. Routine decontamination of the laboratory surfaces and instruments involves exposure to UV radiation and thorough cleaning using bleach, decon90 (decon) and ethanol. In order to protect the laboratory environment from modern human DNA, researchers are required to wear protective clothes consisting of a whole body suit, a face-mask, a face-shield, gumboots, and three pairs of surgical gloves that are changed between individual working steps.

2.Blank controls (one extraction blank for every five ancient samples and two PCR/GenoCoRe22 blank controls for every six reactions) allowed monitoring and controlling large-scale and systematic contamination within the laboratory or in the reagents. In addition, haplotypes similar to those of the users of the laboratory could not be observed from aDNA extracts. Mitochondrial DNA data from the archaeologists and anthropologists involved in the collection of the samples was not available. However, we estimate as rather low the probability that contamination by a few modern-day individuals would generate the diversity and specific patterns of mtDNA lineage distribution observed in the ancient populations under investigation.

3.Multiple replications of HVR-I amplification and direct sequencing were performed in order to detect artefactual sequences due to contamination, DNA degradation or jumping PCR events. When possible, two teeth were collected for each individual and DNA was extracted independently from each sample (i.e., a minimum of two extractions per individual). For each extract, each PCR fragment and each GenoCore22 SNP position was genotyped from at least two independent PCR products (i.e., a minimum of four independent PCRs per fragment and four GenoCoRe22 reactions per individual). This strategy was chosen over cloning for most of the individuals examined here. In low-template conditions, clone sequences can represent the small population of highly degraded starting DNA templates that were exponentially amplified by the one single PCR. In our opinion, a hierarchical replication strategy based on multiple independent amplifications is a powerful alternative to cloning in order to detect artefactual mutations and provides confidence about the authenticity of our DNA sequences.

4.The independent replications of DNA extraction/amplification/direct sequencing carried out at the Johannes Gutenberg University confirmed the diagnostic mutations initially identified at the ACAD in the six selected individuals: UZOO-77, BOO57-1, BOO72-1, BOO72-4, BOO72-7, and BOO72-15 (Figure S3).

5.Sequencing of cloned PCR products for six individuals (individuals UZOO-77, BOO57-1, BOO72-1, BOO72-4, BOO72-7, and BOO72-15) allowed the corresponding haplotypes to be verified. The sequences showed nucleotide positions modified by post-mortem damage as inconsistent cytosine to thymine or guanine to adenosine base changes (Figure S3). For one individual (BOO57-1), independent replications and cloning did not allow allelic resolution at np 16390R. At this position, double peaks (A/G) were observed in direct sequencing, and alleles A and G showed an equal distribution among clones (Figure S3). This position might be heteroplasmic in the BOO57-1 individual, as np 16390 has been described as a mutational hotspot [86], and therefore might as well be a hotspot for post-mortem DNA damage exhibiting a high rate of post-mortem cytosine deamination.

6.The amount of template mtDNA molecules for two fragments of different sizes (133 bp and 179 bp) was estimated and compared in order to test whether they were consistent with low concentrations of recent human mtDNA contaminants in six selected aDNA extracts (UZOO43, UZOO74, BOO72-1, BOO79-9, and two ancient co-extracts from a related study; data not shown). The size distribution of endogenous aDNA molecules was previously shown to be skewed towards smaller fragment sizes due to post-mortem damage, i.e. DNA fragmentation [87]–[90]. Here, the Shapiro-Wilk W test was first used to verify that the number of copies for each fragment followed a normal distribution (p = 0.2215 for the 133 bp short fragment and p = 0.5381 for the 179 bp long fragment). A significantly larger number of copies for the shorter (133 bp) compared to the larger (179 bp) fragment was statistically confirmed by a one-tailed paired t-test (p = 0.04337) in R version 2.12 (R Development Core Team, http://www.R-project.org). Quantitative PCR results suggest a low level of contaminating DNA molecules, the presence of which would have been detected by higher copy-number of longer (less fragmented) DNA molecules.

7.The phylogenetic consistency of the haplotypes and matching hgs assignments of both HVR-I data and coding region SNPs, were indicative of the robustness of the mtDNA typing approach presented here.

Ancient DNA Reveals Prehistoric Gene-Flow from Siberia in the Complex Human Population History of North East Europe (http://www.plosgenetics.org/article/info:doi/10.1371/journal.pgen.1003296)

So how exactly are the samples from La Chora, La Pasiega or Karelia contaminated??? Need I remind you UZOO-77 is the one that turned out to be mt-DNA H.

J Man
07-02-2014, 03:48 AM
There are no confirmed H samples that are free of contamination. Every year that goes by with more U and no H makes it more difficult to sustain the faith that H was present in Europe before the Neolithic. We can't rule out the possibility, but there is no evidence to support it.

I doubt that the Mesolithic H sample from Karelia is the result of contamination. Same goes for the new Samara H sample.

Jean M
07-02-2014, 09:52 AM
I doubt that the Mesolithic H sample from Karelia is the result of contamination. Same goes for the new Samara H sample.

I agree. But as I have said earlier on this thread, these are samples from hunter-gatherers after the Neolithic reached Europe. They are not from pre-Neolithic Europe.

Jean M
07-02-2014, 10:06 AM
And how about the middle-east prior to the Neolithic?

We have no ancient DNA from the Middle East prior to the Neolithic. However the estimated date of H is 12,846 years ago (Behar 2012). The start of the Near Eastern Neolithic was c. 11,000 years ago. H shows a pattern of huge expansion during the Near Eastern Neolithic i.e. a star-cluster of subclades.


2024

Its expansion in Europe coincides with the arrival of the Neolithic. Fu, Rudan, Pääbo and Krause 2012 took European samples of mtDNA U to represent European hunter-gatherers and H to indicate farmers. Then they counted the number of mutations in these sequences. European H-type mtDNAs have on average six differences in their coding region, while U-type mtDNAs have on average 18 differences. That suggests a much older population expansion in U than in H.

2025

Jean M
07-02-2014, 10:29 AM
One has to love how you don't question the U5 from Eralla, Guipuzcoa reported by Hervella.et.al.2012 with 16270 HVR-I, but when it comes to La Chora and La Pasiega there is just no way that they could be H and H6. Double standard much???

Believe me I feel your pain. I am distressed for Basques generally as they grapple with a new ethnic reality. It can't be easy. And believe it or not, there are British archaeologists who are in shock and just can't believe it either. They cherished the idea that the Basques are 100% palaeolithic relic every bit as much as any Basque. I loved the idea myself when I first came across it about 40 years ago. My ideas have had to change as I follow the data.

It does indeed look highly unlikely that the samples from La Chora and La Pasiega were H of any variety. I'm terribly, terribly sorry to have to say this to you, when you find it so upsetting, believe me. If I could find a way to make all Basques happy that was actually in accordance with the scientific process, then I would be delighted to do so.

Humanist has started a thread on Basque mtDNA http://www.anthrogenica.com/showthread.php?2791-Basque-mtDNA

J Man
07-02-2014, 01:04 PM
Anyways it is pretty cool that some U5b has been found among the Pitted Ware samples from Sweden.

jeanL
07-02-2014, 01:17 PM
We have no ancient DNA from the Middle East prior to the Neolithic. However the estimated date of H is 12,846 years ago (Behar 2012). The start of the Near Eastern Neolithic was c. 11,000 years ago. H shows a pattern of huge expansion during the Near Eastern Neolithic i.e. a star-cluster of subclades.


2024

Its expansion in Europe coincides with the arrival of the Neolithic. Fu, Rudan, Pääbo and Krause 2012 took European samples of mtDNA U to represent European hunter-gatherers and H to indicate farmers. Then they counted the number of mutations in these sequences. European H-type mtDNAs have on average six differences in their coding region, while U-type mtDNAs have on average 18 differences. That suggests a much older population expansion in U than in H.

2025

The key words here are population expansions, not origin. Needless to say that given the observed frequencies(Mostly mt-DNA U with little mt-DNA H) of mt-DNA haplogroups in pre-Neolithic Europeans is more than obvious that mt-DNA H expanded later. At the expense of coming out blunt, because I cannot sugar coat the statements. You keep quoting a single scientific study which showed that mt-DNA H was 12,846 years old(Behar 2012), Behar and colleagues have published other studies that show mt-DNA H to be older. All other studies show mt-DNA H to be ~20 kya. I just freaking quoted a study earlier in this thread (http://www.anthrogenica.com/showthread.php?533-Upper-Paleolithic-Mesolithic-European-mtDNA-Haplogroups&p=44079&viewfull=1#post44079) Fu.et.al.2013 which calibrated using ancient DNA and found mt-DNA H to be 20kya+ old. Why do you keep ignoring the evidence???? You do realize that what you are doing is using the fringe area that is due to randomness and claiming it to be true. Let me make it very clear, what you are doing is taking the one study which likely due to randomness yielded a different estimate, and fixating on it.

jeanL
07-02-2014, 01:28 PM
Believe me I feel your pain. I am distressed for Basques generally as they grapple with a new ethnic reality. It can't be easy. And believe it or not, there are British archaeologists who are in shock and just can't believe it either. They cherished the idea that the Basques are 100% palaeolithic relic every bit as much as any Basque. I loved the idea myself when I first came across it about 40 years ago. My ideas have had to change as I follow the data.

It does indeed look highly unlikely that the samples from La Chora and La Pasiega were H of any variety. I'm terribly, terribly sorry to have to say this to you, when you find it so upsetting, believe me. If I could find a way to make all Basques happy that was actually in accordance with the scientific process, then I would be delighted to do so.

Humanist has started a thread on Basque mtDNA http://www.anthrogenica.com/showthread.php?2791-Basque-mtDNA

Ok now I'm angry!! You are projecting yourself here, you are the one with the clear agenda, not me. How does the fact that my background is Basque or not change the presence of mt-DNA R0/HV in Paglicci Cave dating 24,000 ybp, how does it make La Chora and La Pasiega not H. You see the freaking scientific process has shown that mt-DNA H was indeed in pre-Neolithic Europe. If you want to sell your ideas in your book fine!!! Now don't you dare go around accusing people of being biased, when all you do is manipulate data. Once more, instead of dodging the questions, or going on the tangent on how Basques are these or that, prove how are the Karelian, Samarans, Cantabrians mt-DNA H finding inconsistent with mt-DNA H being pre-Neolithic in Europe. This is about science, and the scientific process of learning and the unethical practice of bringing opinions that have nothing to do with the argument by people like you just destroys it, and makes our discussion decay into a "I'm right and you are wrong!!" argument. So what is it? You can't refute the data, so let's just claim we are all hurt that the Basques cannot be Paleolithic, and call it day. Poor little Basque boy wants Basques to be 100% Paleolithic that's why he is making up the facts. No madam!! I'm showing you scientific studies, the fact that you keep ignoring the Fu.et.al.2013 age estimates, the fact that you have to go into special pleading about the presence of mt-DNA H in Mesolithic Karelians 7500 ybp, and now seems Hunter Gatherers from Samara, says a lot about you and your agenda, not me. What are the odds that the single farming lineage that makes it into the Mesolithic Karelians and Samarans is mt-DNA H, when there is a study also posted earlier showing that farmers and hunter gatherers living side by side in Germany did not mix for over 2000 years!!! Moreso, the frequency of mt-DNA H is <20% in LBK, and other nonSouthWestern European samples. Why are we seeing mt-DNA H in the Hunter Gatherers from Karelia, and say not N, or T. The EEF Sttutgart was not mt-DNA H. I've had enough, to me this is demeaning, and just shows what kind of person you are. I'm out of this discussion because you cannot keep it in a scholarly manner, and there is a clear agenda behind your assertions.

parasar
07-02-2014, 02:25 PM
We have no ancient DNA from the Middle East prior to the Neolithic. However the estimated date of H is 12,846 years ago (Behar 2012). The start of the Near Eastern Neolithic was c. 11,000 years ago. H shows a pattern of huge expansion during the Near Eastern Neolithic i.e. a star-cluster of subclades.

...

Its expansion in Europe coincides with the arrival of the Neolithic. Fu, Rudan, Pääbo and Krause 2012 took European samples of mtDNA U to represent European hunter-gatherers and H to indicate farmers. Then they counted the number of mutations in these sequences. European H-type mtDNAs have on average six differences in their coding region, while U-type mtDNAs have on average 18 differences. That suggests a much older population expansion in U than in H.


U is phylogenically equivalent to R0, not H.

We do have the precursor for H present in Europe al least 24000ybp - Paglicci-25: HVR1 CRS, +7025 AluI, 73A, 719G, and 12308A.

Hervella et al for their H determination went through a number of steps, both to account for potential contamination and to establish haplogroup:

In order to identify any possible contamination that might have occurred in the different stages of the laboratory work, at least two extraction controls and several PCR negative controls were included in each amplification. If any contamination was detected, the results obtained were discarded.
...

to classify the mitochondrial variability of the individuals analyzed in this study, we proceeded to amplify 11 markers, which are required for defining the 10 Caucasian haplogroups described [63 [http://www.sciencedirect.com/science/article/pii/S0002929707616772]].

DMXX
07-02-2014, 05:37 PM
Let's try and refrain from personalising the discussion, ladies and gents. Any further postings will be edited in accordance to our rules.

J Man
07-02-2014, 07:53 PM
It would be interesting to see the subclade of H from Samara. I think one of the Iberian once from the Hervella study is H6 and the one from Mesolithic Karelia is H2a. Maybe some H subclades were present in Europe during the Upper Paleolithic and Mesolithic and then some came later with the Neolithic. This seems likely actually.

Generalissimo
07-03-2014, 12:06 AM
What are the odds that the single farming lineage that makes it into the Mesolithic Karelians and Samarans is mt-DNA H, when there is a study also posted earlier showing that farmers and hunter gatherers living side by side in Germany did not mix for over 2000 years!!!

That's actually a very good point, although we have seen mtDNA T2b and K1a among Pitted Ware Scandinavian hunter-gatherers, but no H.

So perhaps with larger sample sizes we'll see Mesolithic European hunter-gatherers all the way from Iberia to the Samara, and beyond, carrying mtDNA U5 and U4, but also with very minor frequencies of H, T, K and J?

GailT
07-03-2014, 12:38 AM
Maybe some H subclades were present in Europe during the Upper Paleolithic and Mesolithic and then some came later with the Neolithic. This seems likely actually.

It seems possible but not very likely, and we don't have any evidence to suggest this occurred. We debated the Fu et al. paper more than a year ago, and it does not seem very useful to repeat that debate as views are deeply entrenched. But quoting from Fu et al.


All but one of the ancient modern human sequences from Europe belonged to mtDNA hg U, thus confirming previous findings that hg U was the dominant type of mtDNA before the spread of agriculture into Europe.

The one exception was T2b1 and when it was properly dated, it was found to be 800 years old, not 30,000 years old.

The history of this debate is informative. Originally people assumed H was Paleolithic in Europe because H is the dominant haplgroup in Europe today, and we know that people never migrated very much after Europe was first populated by modern humans, therefore, the first modern humans in Europe must have been H. The same argument was made for Y-DNA R1b. Of course that was always a nonsensical argument and bad science.


Fu et al. do not describe how they derived a date for haplogroup H. The focus of the paper was the dating the Africa/non-African date and using ancient samples to better calibrate the dating. All of the ancient samples were U, none were H, so they obviously did not use ancient H samples to calibrate H. My guess is that they did not do a careful analysis of H. Until someone shows how they derived the date, I consider it dubious, and the careful analysis of Behar is much more reliable.

We also have many modern H samples that are plain H with no extra mutations. This is extremely unlikely if H were much older than 14,000 years. The only way you can justify an older date for H is to claim it has a slower mutation rate, which some people have tried to do, but which seems to be special pleading to avoid the obvious.

alan
07-03-2014, 12:47 AM
I have an open mind of H. There are several factors that make me feel this way

1. The impulses from the near east into Europe are common and exist in the Palaeolithic and Neolithic. It is possible that the Gravettian is one of those.

2. The refugia with its long bottlenecking etc may have had different proportions come the end of the LGM and movement out. The main movement out of a refuge was the Magdellenian one and it is possible that it could have altered the mtDNA of hunter gatherer Europe where it spread into areas that previously had been settled. Other groups may have brought Siberian genes in from the east at a similar or slightly later period. The third and less expansive but most in-situ group in the east and SE of Europe were epi-Gravettians. Again they were probably bottled up in refugia. So there may not have been a single pattern and we need to be more specific about what group of hunters we are talking about.

Jean M
07-03-2014, 09:44 AM
1. The impulses from the near east into Europe are common and exist in the Palaeolithic and Neolithic. It is possible that the Gravettian is one of those.


Gravettian certainly seems to start in Eastern Europe, so you never know. The problem here is the date. Even if we suppose that new people arrived 32,000 years ago, that is much too early for them to be carrying mtDNA H.

I recall that I found the date of H a puzzle even back in the days when I was mainly going along with Sykes and IIRC it was dated somewhat earlier. How could it have got to Europe in the middle of the LGM? Who would be mad enough to go north in the LGM? So I was a bit baffled. Now that the date of H is post-LGM, the obvious pulse into Europe is the Neolithic. It all makes sense.

However ancient DNA has brought surprises over and over again. No doubt it will carry right on doing that. With an improved climate, there would be nothing stopping hunter-gatherers from wandering northwards from the Levant into Eastern Europe after about 10,000 BC. Not that we have any evidence of it. The main input into the re-peopling of northern Europe in the Mesolithic was from the Franco-Cantabrian refuge.

2027

alan
07-03-2014, 11:21 AM
I think Jean is probably right in that H was at least very rare in pre-farming Europe. I dont really know the ins and outs of the dating of H and the 20+ option is a bit vague. Anyway, there are a few possible ways new genes of any type could have spread into Europe within the c.20000-10000BC time range from the LGM to late pre-farming times. Its worth chewing over them.

1 There is the somewhat mysterious Solutrean of SW Europe. Much debated but local origin from Gravettian seems likely and although possible the Gibraltar route of contacts remains unproven. BTW dont google this culture - the net is totally spammed up with the Clovis type nonsense.

2. There is the Badegoulian c. 21000BC which some see as an eastern movement into western Europe that led to the Magdalenian culture c. 18500BC. Its hotly debated though. Its more likely to be an impulse from east-central Europe though so technically unlikely to bring new non-European genes in and perhaps unlikely that H was involved. However, it seems to me that the eastern part of Europe in the LGM would not be entirely cut off from the area to the south so who knows. I understand that even NW Turkey had strong late upper Palaeolithic and epi-Palaeolithic links with groups on the west and north Pontic areas.

3. There is the poorly understood (due to it being under water today) Med. shore route of the LGM which in theory could have allowed pre-farming movement along its length, perhaps including via north Africa. Perhaps the most likely but least well understood route for pre-farming H?

4. There is of course the Magdalenian movement itself which if somehow H had made it by some route (the Med.? ) to its core area in Franco-Cantabria by c. 18000BC or so, was the main post-LGM movement across Europe reaching from SW Europe through central Europe and along the northern European zone as far east as and including the Swiderian groups of the Baltic etc. It main zone of little impact was the areas like Italy, SE Europe and Ukraine where epi-Gravettain groups remained.

5. Then there are the post-Swiderian and other groups with pressure microblades arriving after 9000BC in northern and eastern Europe. Seems an unlikely source of H though by any stretch.

On balance its not obvious how H would have made its way into Europe from SW Asia in LGM or post-LGM pre-farming times but the route from the warmer south and/or along the Med. seems to me to hardly be a closed one during the LGM so its not impossible there could have been gene flow.

Jean M
07-03-2014, 12:20 PM
I've been searching for my previous comments on the Bollongino 2013 paper, referred to above. Found them. http://www.anthrogenica.com/showthread.php?755-Ancestral-Journeys-The-Peopling-of-Europe-from-the-First-Venturers-to-the-Vikings&p=30721&viewfull=1#post30721


It suggests a strange situation, with a late hunting group not mixing with the farmers, but the conclusion that they lived local parallel lives for 2000 years is not actually supported by the paper. That is an assumption based on the idea that the hunting group was a local survival from the Mesolithic. In fact hunters and fishermen carried right on hunting and fishing around the Baltic and along the North Sea coast after the arrival of farmers. So a fishing family could have arrived at the far inland Blätterhöhle site when the evidence actually proves that they were there i.e. 3922 ± 60 [BC] and kept to themselves for nearly 500 years, which is the maximum indicated by the paper.

I found this study* puzzling, as two separate groups would not normally share a burial place. A better explanation just occurred to me. The fishing people could have been just occasional visitors from the coast, delivering fish. Occasional deaths in situ would be dealt with by the locals burying the body in the local burial spot that they used themselves.

* http://www.sciencemag.org/content/342/6157/479.abstract

Jean M
07-03-2014, 12:27 PM
We also have many modern H samples that are plain H with no extra mutations. This is extremely unlikely if H were much older than 14,000 years.

Gail - I'm thrilled to bits! I never thought that my own boring, uninformative H* would enter into any deductions about anything. :)

Seriously, that is a brilliant piece of reasoning.

Jean M
07-03-2014, 12:34 PM
I don't really know the ins and outs of the dating of H

As Gail says, anything older than 14.000 years ago = 12,000 BC seems unlikely. The most recent dating effort is that by Behar 2012. He has

H 12,846 y.a. = 10,846 BC.
H1 9888 y.a.
H2 11,905 y.a.
H3 8919 y.a.
H4 10 617 y.a.
H5 9877 y.a.
H6 10,945 y.a.
H7 8890 y.a.
H8 8341 y.a.
H9 6460 y. a.
H10 8594 y.a.
H11 8425 y.a.
H31 3965 y.a.

etc. You see the expansion in the Neolithic.

alan
07-03-2014, 01:48 PM
If that dating is close to real then yes I agree although I would leave open the possibility of H bleeding into the European population in the pre-farming era c. 12-7000BC As I posted there are for example connections between Turkey and the west/north Pontic epigravettian in the late/epi-Palaeolithic sort of period. However, admittedly that culture wouldnt provide a scenario to take take it very far west in Europe - certainly not west of Italy.

I doubt a movement from SW Asia through the Caucasus or east happened because of the whole Pontic-Caspian-Aral cascade after the LGM forming a huge barrier unless it was very late pre-farming.

However the lack of any major physical barrier stopping movement west from SW Asia in the post-LGM pre-farming period is interesting as a possibility and most of that shoreline is under water. I cannot say though that there is any obvious remains of such a movement on dry land.



As Gail says, anything older than 14.000 years ago = 12,000 BC seems unlikely. The most recent dating effort is that by Behar 2012. He has

H 12,846 y.a. = 10,846 BC.
H1 9888 y.a.
H2 11,905 y.a.
H3 8919 y.a.
H4 10 617 y.a.
H5 9877 y.a.
H6 10,945 y.a.
H7 8890 y.a.
H8 8341 y.a.
H9 6460 y. a.
H10 8594 y.a.
H11 8425 y.a.
H31 3965 y.a.

etc. You see the expansion in the Neolithic.

Jean M
07-03-2014, 02:25 PM
If that dating is close to real then yes I agree

What has actually been building up over the last few years from archaeology and genetics is a really interesting picture. Instead of one one single migration into Europe in the Palaeolithic, it is starting to look like several. The fascinating bit for me is that we can actually link the movement through the Caucasus to mtDNA U2, which then seems to have hung around in eastern Europe. U2 must be a separate (and later) movement from the "Asian Crossroads" than U*, because U2 also appears in South Asia. The Behar 2012 date for U2 is 42,805 y.a.


2028

Where I'm keeping an open mind is over the possibility of R0 [Behar date 39,960 years ago] and/or HV [Behar date 21,905 years ago] entering Europe before the LGM. R0 would fit within the wanderings of the Aurignacian.

parasar
07-03-2014, 03:10 PM
What has actually been building up over the last few years from archaeology and genetics is a really interesting picture. Instead of one one single migration into Europe in the Palaeolithic, it is starting to look like several. The fascinating bit for me is that we can actually link the movement through the Caucasus to mtDNA U2, which then seems to have hung around in eastern Europe. U2 must be a separate (and later) movement from the "Asian Crossroads" than U*, because U2 also appears in South Asia. The Behar 2012 date for U2 is 42,805 y.a.


2028

Where I'm keeping an open mind is over the possibility of R0 [Behar date 39,960 years ago] and/or HV [Behar date 21,905 years ago] entering Europe before the LGM.
Many kinds of U appear in South, not just U2. U2 has just undergone a greater expansion especially in the north.

U1, U2, U3, U4, U5, U6, U7, U8
Fig. 3. Phylogenetic network of haplogroup R and its sub-haplogroups among Indian populations
Locations are in Fig. 2 http://www.krepublishers.com/02-Journals/IJHG/IJHG-08-0-000-000-2008-Web/IJHG-08-1-2-001-256-2007-Abst-PDF/IJHG-08-1-2-085-08-336-Maji-S/IJHG-08-1&2-085-08-336-Maji-S-Tt.pdf

U1, U2, U4, U5, U7, U8, K, U9
http://www.biomedcentral.com/1471-2156/9/86/table/T5
http://www.biomedcentral.com/1471-2156/9/86/table/T3

Jean M
07-03-2014, 03:31 PM
Many kinds of U appear in South, not just U2. U2 has just undergone a greater expansion especially in the north.

Perhaps I should have put that differently. U2 has its own subclades in South Asia, which do not (generally) appear in Europe. They are of an age to suggest that they developed from U2 among hunter-gathers in South Asia e.g. U2a 22,693 years ago; U2b 29,253 years ago.

The European subclade is U2e, which appears to have arisen there.

parasar
07-03-2014, 04:21 PM
Perhaps I should have put that differently. U2 has its own subclades in South Asia, which do not (generally) appear in Europe. They are of an age to suggest that they developed from U2 among hunter-gathers in South Asia e.g. U2a 22,693 years ago; U2b 29,253 years ago.

The European subclade is U2e, which appears to have arisen there.

Europe is possible, as is North Asia, though not certain. U2e forms a clade (shared T152C!) with U2c which is quite common in South Asia which has some U2e too.

U2c:
http://www-tc.pbs.org/wnet/finding-your-roots/files/2012/05/U2cMap-631x413.png

The Lankan paper you had posted in the New Papers thread has some interesting data on the Vedda.
http://www.nature.com/jhg/journal/v59/n1/full/jhg2013112a.html


… Vedda, believed to be the most indigenous people on the island …

… the oldest skeletal remains of anatomically modern man (Homo sapiens) reported from the South Asian region, and dated tentatively to 37 000 YBP, were discovered from the cave site, Fahien-lena,8 on the island, with their association with the present-day Vedda people proposed on a comparative anatomical ground …

… Vedda population has the lowest proportion of shared haplotypes among their subgroups (63%) indicating their greater genetic diversity among subgroups …

… Vedda people had the lowest frequency of haplogroup M (17.33%). It is quite astonishing to see such a lower frequency of M haplogroup in the Vedda population when compared with southern Indian tribal groups (70–80%) as well as southern Indian caste populations (65%) …

… Vedda people and Low-country Sinhalese showed relatively high frequencies of haplogroup R (45.33 and 25%, respectively) which was contributed mainly by sub-haplogroup R30b (38.67 and 20%). The haplogroup was less frequent in Up-country Sinhalese, Sri Lankan Tamils and Indian Tamils. Haplogroup U was mostly found in Vedda (29.33%) and Up-country Sinhalese (23.33%), with highest contribution from sub-haplogroups U1a’c (12 and 5%, respectively) and U7a (13.33 and 11.67%, respectively).


U1, U2, U5, U6, U7, U4'9
http://www.nature.com/jhg/journal/v59/n1/extref/jhg2013112x7.pdf

Jean M
07-03-2014, 05:05 PM
Europe is possible, as is North Asia, though not certain.

The Asian side of the Urals is entirely possible as a birthplace of U2e. The modern boundary of the Urals scarcely seems to have bothered hunter-gatherers. Somewhere around the Urals would be my guess, though we shall never know for sure. It certainly seems to have travelled with Indo-Europeans, which would explain its appearance in South Asia.


U2e forms a clade (shared T152C!) with U2c

Some misconception here I think. Here is the tree of U : http://www.phylotree.org/tree/subtree_U.htm

T152C! crops up in various places, including haplogroup K. Sharing just one mutation does not make a clade. I find it best to just follow PhyloTree.

parasar
07-03-2014, 05:24 PM
...
Some misconception here I think. Here is the tree of U : http://www.phylotree.org/tree/subtree_U.htm

T152C! crops up in various places, including haplogroup K. Sharing just one mutation does not make a clade. I find it best to just follow PhyloTree.

That is true, and that had me initially thinking the that U2b2 may form a clade with U2e, but that would make the tree convoluted.
The Phylotree is showing common descent under T152C!:

U2a
U2b
T152C!
.........U2c'd C16234T
........................ U2c C5790a T14935C A15061G
.........................U2d T199C G263A T471C T1700C C4025T A8938G A11893G A14926G T16189C! C16294T
.........U2e A508G A3720G A5390G T5426C C6045T T6152C A10876G T13020C T13734C A15907G G16129c T16189C! T16362C

Jean M
07-03-2014, 07:27 PM
The Phylotree is showing common descent under T152C!:


You are right! Don't know why I missed that. My apologies. Now that really is an interesting little puzzle.

GailT
07-03-2014, 10:25 PM
I don't think that the hyper-Hypervariable region markers like 152 are useful for phylogentics, especially very deep in the tree, but even if you do assume that T152C! has some value for indicating a closer relationship among U2c, U2d and U2e, it might suggest that they share a common ancestor around 40 kya instead of 42 kya, so T152C! doesn't seem to add value.

I think it seems very likely that U2 was widespread from Russia to south Asia around 40 kya, given that we have 5 surviving lines that date from time, and also the Kostenki U2* sample.

U2e is interesting because it undergoes a bottleneck from 40 kya to 20 kya before it begins to expand in Asia and Europe. It seems possible that U2e expanded with Indo-European speakers given that it is found today from Europe to India, but that it also arrived in Europe as early as 11 kya, based on the Blätterhöhle sample. I would guess that U2e expanded from southwest Asia into south Asia after 20 kya and into Europe after the LGM, and then there might have been additional waves of expansion of U2e from SW Asia during the Neolithic. But we need more ancient mtDNA. I don't think we can reconstruct the complexity of multiple waves of migration from present day distributions because the time resolution ability of mtDNA is very poor.


There are 4 lineages of U2e:

U2e1 - found from Europe to India.
U2e2 - found from Europe to India.
U2e3 - just a few FMS samples found in Italy, Ireland and the Middle East
U2e* - a single sample from India

parasar
07-04-2014, 03:52 AM
I don't think that the hyper-Hypervariable region markers like 152 are useful for phylogentics, especially very deep in the tree, but even if you do assume that T152C! has some value for indicating a closer relationship among U2c, U2d and U2e, it might suggest that they share a common ancestor around 40 kya instead of 42 kya, so T152C! doesn't seem to add value.

I think it seems very likely that U2 was widespread from Russia to south Asia around 40 kya, given that we have 5 surviving lines that date from time, and also the Kostenki U2* sample.

U2e is interesting because it undergoes a bottleneck from 40 kya to 20 kya before it begins to expand in Asia and Europe. It seems possible that U2e expanded with Indo-European speakers given that it is found today from Europe to India, but that it also arrived in Europe as early as 11 kya, based on the Blätterhöhle sample. I would guess that U2e expanded from southwest Asia into south Asia after 20 kya and into Europe after the LGM, and then there might have been additional waves of expansion of U2e from SW Asia during the Neolithic. But we need more ancient mtDNA. I don't think we can reconstruct the complexity of multiple waves of migration from present day distributions because the time resolution ability of mtDNA is very poor.

...



11kybp is the latest. U2e could have been present in Europe much earlier - in the Kostenki 14 timeframe.

Jean M
07-04-2014, 08:57 AM
U2e could have been present in Europe much earlier - in the Kostenki 14 timeframe.

Not according to the date given by Behar 2012, which dates U2e as 19,290 years ago. Kostenki 14 is plain U2.

Jean M
07-04-2014, 09:04 AM
I would guess that U2e expanded from southwest Asia into south Asia after 20 kya and into Europe after the LGM.

I hadn't thought of that, I must admit. I was assuming that it was a descendant of the Kostenki 14 sample, so to speak. But of course that is not necessarily so.


It seems possible that U2e expanded with Indo-European speakers

It is more than possible, I'd say. It has actually been found in samples from cultures generally seen as IE-speaking. Obviously it is earlier than Proto-Indo-European (4000 BC), so there is plenty of time to account for before then.

Behar estimated dates:

U2e123 - 17,873 years ago = 15,000-16,000 BC
U2e1 - 14,898 years ago = 12,000-13000 BC
U2e2 - 12,724 years ago = 10,000-1100 BC
U2e3 - 1047 years ago = AD 900-1000

GailT
07-04-2014, 02:16 PM
I hadn't thought of that, I must admit. I was assuming that it was a descendant of the Kostenki 14 sample, so to speak. But of course that is not necessarily so.

Kostenki has 4 extra mutations (at markers 542, 711, 13269, & 15262) so it forms its own subclade parallel to U2a, U2b, U2e etc. So this suggests that U2 was widespread, perhaps from Russia to India, with at least 6 different lineages surviving ca 35 kya. The ancestor of U2e was one of those lineages, but the MRCA of U2e probably lived around 20 kya almost certainly somewhere in southwest Asia.



It is more than possible, I'd say. It has actually been found in samples from cultures generally seen as IE-speaking. Obviously it is earlier than Proto-Indo-European (4000 BC), so there is plenty of time to account for before then.

So that seems to confirm multiple waves of migration of U2e into Europe. We know that U2e was in Europe by 11 kya, but U2e almost certainly did not arrive in Europe until after the LGM (unless the age estimate for U2e is significantly wrong).

It would be very helpful to have the full sequence for the Blätterhöhle sample to see if it is extinct or if it is ancestral to any of the U2e subclades in Europe. My guess is that U2e1, U2e2 and U2e3 all originated in southwest Asia and spread with Indo-European migrations. They could have also spread before that time, but it seems likely that this would have been a major event in in the spread of U2e.

The U2e* samples from India is AY714026. It is difficult to say anything from a single sample. This could represent a U2e subclade that migrated from southwest Asia at any time after 20 kya. The fact that it has not yet been found outside of India might indicate a more ancient migration rather than an Indo-European migration.

edit: I'll post some additional info on U2e in the U2 dicussion forum.

parasar
07-04-2014, 03:03 PM
Not according to the date given by Behar 2012, which dates U2e as 19,290 years ago. Kostenki 14 is plain U2.

Kostenki 14 is a parallel line.
As Kostenki 14 was four mutations down from 16051, I put the split of the U2e line from the U2c'd line in the same timeframe. U2c'd (T152C! 16234) is two mutations down from 16051.

Jean M
07-04-2014, 04:23 PM
I'll post some additional info on U2e in the U2 dicussion forum.

That would be very helpful Gail. I suspect I need to revise some text somewhere.

jeanL
07-04-2014, 06:16 PM
Fu et al. do not describe how they derived a date for haplogroup H. The focus of the paper was the dating the Africa/non-African date and using ancient samples to better calibrate the dating. All of the ancient samples were U, none were H, so they obviously did not use ancient H samples to calibrate H. My guess is that they did not do a careful analysis of H. Until someone shows how they derived the date, I consider it dubious, and the careful analysis of Behar is much more reliable.


The Fu.et.al.2013 age estimate are by far much more reliable than the Behar.et.al.2012 age estimates because they use branch shortening since they had actual ancient DNA. Dienekes explains it very good in his post:

The paper uses the idea of branch shortening to infer dates for the mtDNA phylogeny. Briefly, if one counts differences between a present-day sample P and an ancient sample A, one can get a sense of when their most recent common ancestor X, lived. But PX and PA are not equal: PA is shorter, because A has "missed" a few thousand years of evolution (depending on its age). If we know the age of A (and this can be reliably known by direct dating for many samples), then we can infer the time of X. (http://dienekes.blogspot.com/2013/03/revised-timescale-of-human-mtdna.html)

So there is no need for there to have been any ancient mt-DNA H used in their analysis, since they probably obtained the right mutation rate using calibrated data from ancient DNA. What's is more the Behar.et.al.2012 is the only age estimate that differs significantly from any other age estimate for mt-DNA H(~20 kya).

http://2.bp.blogspot.com/-71JYGXNrukY/UUyZ07GOx5I/AAAAAAAAItk/g4kxvVCtEgc/s1600/mtdna_ages.png

GailT
07-04-2014, 07:24 PM
The Fu.et.al.2013 age estimate are by far much more reliable than the Behar.et.al.2012 age estimates because they use branch shortening since they had actual ancient DNA.

The key question is what haplogroup H database did Fu et al. use in their analysis? If they used the Behar database, they need to explain why their results were consistent with Behar for U5 but very different for H.

Behar used the Soares et al. mutation rate, 1.665 x10-8 for the whole genome. Fu et al. estimated a rate of 2.67 x10-8. It's not clear what effect the rate has in their BEAST simulations and their ML tree. It is not a linear dependence.

Behar's age estimates also differed from Soares, presumably because Behar had a larger database.

Given that all of the branch shortening samples were in U, it is possible that they could have used these samples with BEAST to do a refined analysis for U. I don't know if they did this. But it certainly was not possible for H. We've had quite a few ancient H samples reported since 2013 so this would be an interesting analysis to do.

In any case, I consider Fu et al. far less reliable than Soares and Behar, When you compare average mutations in U5 and H, it is completely implausible that they are differ in age by only 6000 years. This only works if you assume that H has a slower mutation rate than U5.

parasar
07-04-2014, 08:23 PM
... When you compare average mutations in U5 and H, it is completely implausible that they are differ in age by only 6000 years. This only works if you assume that H has a slower mutation rate than U5.

How many more mutations is U5* downstream from a common node compared to H*?

jeanL
07-04-2014, 10:27 PM
Behar used the Soares et al. mutation rate, 1.665 x10-8 for the whole genome. Fu et al. estimated a rate of 2.67 x10-8. It's not clear what effect the rate has in their BEAST simulations and their ML tree. It is not a linear dependence.


What do you mean by their ML tree? You are going to have to be a little be more explicative about their Bayesian vs. Classical analyses. What are they inferring in here? TMRCA, mutation rate, current distribution? I was reading about BEAST and it seems that they use MCMC, but I couldn't see what their sampling techniques are.Anyhow, if you could please clarify what exactly was being inferred by their ML or Bayesian estimates, that would be great?

It seems Beast can be used in several different ways, one of the is a coalescent analysis with splitting. The other one is with a fixed mutation rate. I'm trying to think that perhaps it was the choice of database for mt-DNA H what produce a different result. In any, given what I said that almost all but the Behar.et.al.2012 study have produced the same age for mt-DNA H, then it is likely the Behar.et.al.2012 that due to randomness produce a type II error. Or it could be that they used PAML instead of Beast, that could definitely make a hell of a difference. Never mind, that is likely what produced the difference between the two.

Here is what is really unnerving about the Behar.et.al.2012 age estimate for mt-DNA H, for once no mention of the sample size, yet they produce an age estimate of 12846 with an SD of 773.4 years. The total sample size was less than 10,000, that's a pretty darn good SD, which seems almost unreal. For twice, their age estimate for mt-DNA D4h3a was 12953.8 years with an SD of 2580.2 years, now we know that Anzick kid (http://www.nature.com/nature/journal/v506/n7487/extref/nature13025-s1.pdf), who was date to 12,722 to 12,590 calibrated ybp was already D4h3a, that's cutting it close, yet with the more ample SD I'm not so skeptical to accept that age estimate.

GailT
07-04-2014, 11:25 PM
What do you mean by their ML tree? You are going to have to be a little be more explicative about their Bayesian vs. Classical analyses. What are they inferring in here? TMRCA, mutation rate, current distribution? I was reading about BEAST and it seems that they use MCMC, but I couldn't see what their sampling techniques are.Anyhow, if you could please clarify what exactly was being inferred by their ML or Bayesian estimates, that would be great?

Soares used several different approaches to estimate ages of haplogroups, including evaluation of rho (the average number of mutations for the whole genome) and a maximum likelihood (ML) approach using rho. And yes, it would be very helpful if someone summarized and simplified what Soares did and explained exactly how it compares to Behar and to Fu et al (who don't provide enough information to determine how they arrived at their estimates).

I listed below the age estimates from Soares for U5, H and the ratio of U5/H. These values are from Table S5 which is based on ML using the complete genome. Soares reported an estimate of about 30,000 years for U5 using the rho approach. Behar used the Soares recommended ML approach, using PAML software but reports younger ages for both U5 and H, but the ratio of U5/H is similar to Soares. Fu has an age estimate for U5 similar to Behar but doubles the age estimate for H, and the Fu ratio of U5/H is very different from both Soares and Behar.

It doesn't seem plausible that Fu used ancient U samples to develop a better analysis, but then derives the same age for U5 as Behar and doubles the age of H. Fu doesn't say anything about the H age estimate, and I wonder if they even noticed that they have a very different estimate than Behar and Soares. I would have expected them to commented on this.



study U5 H U5/H
Soares 36000 18600 1.94
Behar 30248 12846 2.35
Fu 29600 23900 1.24

palamede
07-06-2014, 08:37 PM
How many more mutations is U5* downstream from a common node compared to H*?

I did a calculation but I counted the number of mutations only one time and I hope I didn't do bad errors.
http://eng.molgen.org/viewtopic.php?f=24&t=1098


The common ancestor are from older to more recent
Eve mitochondrial__new mutations_________cumul mutations
L1-6_____________8(5)___________________8(5)____8 is the total number of mutatins and 5 the number of coding mutations
L2-6_____________5(4)__________________13(9)
L2’3’4’6_________11(7)__________________24(16)
L3’4’6__________2(2)___________________26(18)
L3’4____________5(3)___________________31(21)
L3_____________3(2)___________________34(23)
N_____________5(5)____________________39(28)
R_____________2(1)____________________41(29)

Branch H
R0____________2(1)____________________43(30)
HV____________1(1)____________________44(31)
H_____________2(2)____________________46(33)
H1____________1(1)____________________47(34)

Therefore, H is 6 mutations (included 5 coding mutations) below R .

Branch U
U_____________3(3)____________________44(32)
Therefore, U is 3 mutations (included 3 coding mutations) below R, therefore U is older than H for 2 mutations (including 1 coding mutation)
U1____________7(5)____________________51(37)
U5____________5(3)____________________49(35)
Therefore, U5 is 8 mutations (included 6 coding mutations) below R, therefore U5 is younger than H for 2 mutations (including 1 coding mutation) and younger than H1 for 1 mutation (including 1 coding mutation).
U5a___________2(1)____________________51(36)
U5b___________3(2)____________________52(37)
U6____________2(1)____________________46(33)
U2'3'4'7'8'9____1(1)_____________________45(33)
U2___________1(0)_____________________46(33)
U3___________5(3)_____________________50(36)
U4'9__________3(1)_____________________53(37)
U4___________6(5)_____________________59(42)
U9___________4(4)_____________________57(41)
U7___________11(8)____________________56(41)
U8___________1(1)_____________________46(34)
U8a__________6(5)_____________________52(39)
U8b'K________3(3)______________________49(37)
U8b__________3(0)_____________________53(37)
K____________5(3)_____________________54(40)

You can understand I am very sceptikal about all theses erudite calculations. In more the basis of their time calculations is wrong because all serious anthropological works find a split between Homo and Chimpanzee dated between 9M and 15M which shows the genetic dates are subdated from about a rapport of two. Things which seem confirmed by archeological works from the Skhul and Qafzeh skeletons until the recent discoveries in South Arabia and elsewhere .
http://en.wikipedia.org/wiki/Kafzeh read the objective information and neglect the hypotheses .

GailT
07-06-2014, 10:30 PM
Therefore, H is 6 mutations (included 5 coding mutations) below R .

Therefore, U is 3 mutations (included 3 coding mutations) below R, therefore U is older than H for 2 mutations (including 1 coding mutation)

Therefore, U5 is 8 mutations (included 6 coding mutations) below R, therefore U5 is younger than H for 2 mutations (including 1 coding mutation) and younger than H1 for 1 mutation (including 1 coding mutation).


Counting mutations on each branch is not a reliable approach to estimate the age of haplogroups because the accumulation of mtDNA mutations is highly variable. For example, there are living people in haplgroup H with zero extra mutations and people with more than 10 extra mutations.

Instead, you use all of the samples in each subclade to determine the average number of mutations and use this to estimate the age, or better yet, use a Bayesian MCMC (http://en.wikipedia.org/wiki/Markov_chain_Monte_Carlo) modelling approach to determine the maximum likelihood age estimate for each branch.

When you look at the entire database of H and U5 samples, the average number of mutations in U5 is larger than in H.

Here is a more concrete example: JQ324622 is H with one extra mutation. JN573371 is H with 12 extra mutations. If these two people each had descendants that were to define subclades H200 and H201 at some time in the future, counting the mutations that define the subclades would be a very unreliable approach to estimate the age of each subclade. Say that 1000 years from now they each had 500 descendants, and the average number of mutations among the descendants was about 0.4 for both H200 and H201. You would correctly estimate that each subclade was about 1000 years old. H200 would be defined by one mutation and H201 would be defined by 12 mutations, but that would not provide useful information to predict the age of each subclade.

jeanL
07-07-2014, 01:11 AM
Counting mutations on each branch is not a reliable approach to estimate the age of haplogroups because the accumulation of mtDNA mutations is highly variable. For example, there are living people in haplgroup H with zero extra mutations and people with more than 10 extra mutations.

Instead, you use all of the samples in each subclade to determine the average number of mutations and use this to estimate the age, or better yet, use a Bayesian MCMC (http://en.wikipedia.org/wiki/Markov_chain_Monte_Carlo) modelling approach to determine the maximum likelihood age estimate for each branch.

When you look at the entire database of H and U5 samples, the average number of mutations in U5 is larger than in H.


If I recall correctly Bayesian analyses assume that the input is a random variable, whereas maximum likelihood modeling assumes is not, simply that it has some noise, but that it is drawn from a Gaussian distribution of unknown mean and standard deviation, which using the central limit theorem can be approximated as the sample mean and standard deviation. So I'm not sure how they are using Bayesian MCMC to determine do an ML estimation? In fact a maximum likelihood estimator coincides with the most probable Bayesian estimator given a uniform prior distribution on the parameters. What is more using the central limit theorem to determine the minimum sample size required to draw a meaningful estimate will often time yield numbers that are way outside of the current samples that we have.

One thing that I'm greatly concerned is the very very small standard deviation of 773.4 years produce by Behar.et.al.2012, which translates roughly into 0.213 mutations using Soares.et.al.2009 rates. Given a mean of roughly 3.54 mutations for the H sample, I cannot help but wonder how big is the sample size Behar.et.al.2012 used for mt-DNA H. Because that standard deviation is very, very small. By comparison the Fu.et.al.2013 standard deviation for the mt-DNA H sample was 4200 years, or roughly 1.86 mutations using their rate of 2.67 x10-8, and a mean of 10.56 mutations in the whole molecule. Now something doesn't quite add up in here, does it now? How can Fu.et.al.2013 H sample have an average of 10.56 mutations in the whole molecule, Soares.et.al.2009 assume the TMRCA of 18600 and their rate yielded 5.13 mutations on average, yet Behar.et.al.2012 yielded 3.54 mutations with an SD of 0.213, this means that Soares.et.al.2009 mean would be off the chart, because it would be a whopping 7.46 standard deviations away from Behar.et.al.2012 mean estimates. Whereas relative to Fu.et.al.2013 estimate and SD, the Soares.et.al.2009 5.13 mutations average is 2.91 standard deviations away from the mean or in the 0.18 percentile (Using the law of large numbers, central limit theorem.) Again how does comparing something against the RSRS vs the rCRS make 7 mutations disappear, is beyond me.

parasar
07-07-2014, 04:03 AM
Counting mutations on each branch is not a reliable approach to estimate the age of haplogroups because the accumulation of mtDNA mutations is highly variable. For example, there are living people in haplgroup H with zero extra mutations and people with more than 10 extra mutations.

Instead, you use all of the samples in each subclade to determine the average number of mutations and use this to estimate the age, or better yet, use a Bayesian MCMC (http://en.wikipedia.org/wiki/Markov_chain_Monte_Carlo) modelling approach to determine the maximum likelihood age estimate for each branch.

When you look at the entire database of H and U5 samples, the average number of mutations in U5 is larger than in H.

Here is a more concrete example: JQ324622 is H with one extra mutation. JN573371 is H with 12 extra mutations. If these two people each had descendants that were to define subclades H200 and H201 at some time in the future, counting the mutations that define the subclades would be a very unreliable approach to estimate the age of each subclade. Say that 1000 years from now they each had 500 descendants, and the average number of mutations among the descendants was about 0.4 for both H200 and H201. You would correctly estimate that each subclade was about 1000 years old. H200 would be defined by one mutation and H201 would be defined by 12 mutations, but that would not provide useful information to predict the age of each subclade.

I think this is not a good solution. I would rather count mutations between nodes the way palamede did and calibrate with ancient dna. Over long periods of time mutational rate differences should average out, and for nodes such as U2 we can easily calibrate with ~40000ybp Kostenki 14 to get the age of 16051.

Jean M
07-07-2014, 10:55 AM
I've been searching for my previous comments on the Bollongino 2013 paper ...

I found this study puzzling, as two separate groups would not normally share a burial place. A better explanation just occurred to me. The fishing people could have been just occasional visitors from the coast, delivering fish. Occasional deaths in situ would be dealt with by the locals burying the body in the local burial spot that they used themselves.


Had a nagging feeling that I was remembering the paper incorrectly. I was. The later bunch of foragers were eating freshwater fish. This still distinguishes them from the earlier Mesolithic burials at the Blätterhöhle, who were meat-eaters. The DNA of the fishermen is also too different from that of the earlier Mesolithic group for confidence of direct descent, though the paper was keen not to rule it out:


At first sight, this would seem to cast doubt on a direct genetic continuity between them. However, coalescent simulations do not reject a model of population continuity under a very wide range of demographic parameters..

Also there were U5b2a2 individuals among both the farmers and fishermen. So intermarriage had actually taken place, though the paper plays this down. (There was also a U5b2a5 among the farmers, though not matched among the fishermen.)

So what we have now looks to me like a case of in-laws arriving periodically, or perhaps being cared for by daughters in old age/injury. That would explain the burial in the farmers' usual burial site. It would also explain the complete lack of any sign of Mesolithic life surviving in the surrounding area. The original contact could still have come from trade, but not of marine fish.

J Man
07-08-2014, 03:17 AM
I have a quick question here about mtDNA. Sorry if it is off topic.

Why aren't control region mutations listed on mtDNA results from FTDNA, what range of markers(eg, HVR1 16024-16569) does FTDNA test in each one of its tests, and why are some samples that only tested HV1 or HV1+2, listed as a haplogroup(ex. H) that only have control region mutations?

Humanist
07-08-2014, 04:00 AM
I have a quick question here about mtDNA. Sorry if it is off topic.

Why aren't control region mutations listed on mtDNA results from FTDNA...

Control region mutations are listed, if so configured by the project administrator. However, coding region mutations are not viewable to the public.

FTDNA


What are the parts of the mitochondrial DNA (mtDNA)? What are HVR1 and HVR2? What is the Coding Region?

Mitochondrial DNA (mtDNA) has two major parts, the control region and the coding region.

The control region is often called the hypervariable region (HVR). Hypervariable means fast changing. In mitochondrial DNA, the control region is the fast changing part. The control region may be further divided into two Hypervariable regions, HVR1 and HVR2.

HVR1 runs from nucleotide 16001 to nucleotide 16569.
HVR2 runs from nucleotide 00001 to nucleotide 00574.
The coding region (CR) is the part of your mtDNA genome that contains genes. Because it does contain some genes, the coding region is believed to be slower mutating than the control region. Often, the mutations that are found in the coding region are used to define haplogroups.

The coding region runs from nucleotide 00575 to nucleotide 16000.

GailT
07-08-2014, 04:57 AM
I think this is not a good solution. I would rather count mutations between nodes the way palamede did and calibrate with ancient dna. Over long periods of time mutational rate differences should average out, and for nodes such as U2 we can easily calibrate with ~40000ybp Kostenki 14 to get the age of 16051.

It averages out if you take the average of all of the descendants of the haplogroup (the approach used by every published study). There is no averaging when you simply count the mutations between each node in the tree because each node represent a single individual.

parasar
07-08-2014, 12:09 PM
It averages out if you take the average of all of the descendants of the haplogroup (the approach used by every published study). There is no averaging when you simply count the mutations between each node in the tree because each node represent a single individual.

If I were calculating backwards from the present I would do the same, but we now have calibration reference points from the past.

mothergoddess
09-14-2016, 02:47 AM
If H was found in Paglacci 23 in the Paleolithic, and H1 and H2 are in Europe in the LGM, did Europeans move back to the Caucasus and upper Euphrates area around 8,000 BC? I saw H13 c around Gobekli Tepe and found H13 in the Minoans. So how did H start in Europe and wind up in the Caucasus. I read H13a1a is a very old and diverse haplogroup. If it originated in the Caucasus or n.Iran 25,000 ya and migrated to Europe and left some clan members behind where H13 descended, where are her predecessors in the Caucasus? This haplogroup confuses me like none other! Help please!

J Man
09-14-2016, 02:53 AM
If H was found in Pelagacci 23 in the Paleolithic, and H1 and H2 are in Europe in the LGM, did Europeans move back to the Caucasus and upper Euphrates area around 8,000 BC? I saw H13 c around Gobekli Tepe and found H13 in the Minoans. So how did H start in Europe and wind up in the Caucasus. I read H13a1a is a very old and diverse haplogroup. If it originated in the Caucasus or n.Iran 25,000 ya and migrated to Europe and left some clan members behind where H13 descended, where are her predecessors in the Caucasus? This haplogroup confuses me like none other! Help please!

The Paglicci 23 sample does not belong to mtDNA haplogroup H.

mothergoddess
09-14-2016, 03:09 AM
J man, I have read in several places that it did, but I am new to this. Can you explain H13 a1a and if H 1,2 are European did H mama go back to the Caucasus after the LGM? Thanks R16263 in Paglicci 23 is an R group and H is neither ruled out or in per Wiki. Just read it.

J Man
09-14-2016, 03:11 AM
J man, I have read in several places that it did, but I am new to this. Can you explain H13 a1a and if H 1,2 are European did H mama go back to the Caucasus after the LGM? Thanks

It is not. It is most likely some sort of U.

parasar
09-14-2016, 03:43 AM
http://www.anthrogenica.com/showthread.php?4246-19000-year-old-mt-DNA-Haplogroup-H-from-the-Franco-Cantabrian-site-of-el-Miron!!!!&p=77663&viewfull=1#post77663

Sammy Andrews
09-14-2016, 04:40 AM
http://www.anthrogenica.com/showthread.php?4246-19000-year-old-mt-DNA-Haplogroup-H-from-the-Franco-Cantabrian-site-of-el-Miron!!!!&p=77663&viewfull=1#post77663

It's contamination. H1 IMO can be explained by Eastern European EEF. Most of the EEF ancestors of Northern and SouthEstern Europeans probably lived around Ukraine and Romania and a signifcant but lesser amount for SouthWestern Europeans did. We have little DNA from Neolithic people of those countries. Once we do get their mtDNA it might be rich in H. We do have some and it is rich in H, and also lacks signatures of LBK mtDNA.

GailT
09-17-2016, 05:53 AM
Behar et al. did the most comprehensive age estimates for H and its subclades Their age age estimate for H is about 13,000 years and for H13a1a about 8500 years. The papers that claim that H was present in Europe before the LGM were not based on sound science, rather, they were based on the mistaken belief that people did not migrate very much, and therefore the dominant haplogroup in Europe today most also have been in Europe before the LGM.

Sammy Andrews
09-17-2016, 06:41 AM
Behar et al. did the most comprehensive age estimates for H and its subclades Their age age estimate for H is about 13,000 years and for H13a1a about 8500 years. The papers that claim that H was present in Europe before the LGM were not based on sound science, rather, they were based on the mistaken belief that people did not migrate very much, and therefore the dominant haplogroup in Europe today most also have been in Europe before the LGM.

Do you agree H with Behar et al's age estimate for H? Personally I wouldn't be surprised if it's over 30,000 years old.

GailT
09-18-2016, 04:25 AM
Do you agree H with Behar et al's age estimate for H? Personally I wouldn't be surprised if it's over 30,000 years old.

Yes - I think the Behar estimate is accurate. I'm a volunteer administrator for the FTDNA mtDNA H project, and there are large numbers of people who are H with no additional coding region mutations. It is not plausible that so many people would have no mutations in over 30,000 years. But the more compelling argument is this: H expanded in diversity extremely rapidly shortly after it's origin, i.e., it was part of a population that was growing very rapidly. The Neolithic revolution in the Near East is almost certainly the cause of the rapid expansion in H.

We know from testing of ancient remains that U5 is more than 30,000 years old and was present in Europe both before and after the LGM, and unlike H, U5 experienced a population bottleneck with extremely slow expansion in diversity from 30,000 to about 18,000 years ago. It is only after the LGM that U5 begins to expand in diversity, although at a much slower rate than H. If H were over 30,000 years old, we would expect it to have expanded in diversity slowly, more similar to U5, until the Neolithic Revolution.

Sammy Andrews
09-18-2016, 04:37 AM
Yes - I think the Behar estimate is accurate. I'm a volunteer administrator for the FTDNA mtDNA H project, and there are large numbers of people who are H with no additional coding region mutations.

That's a legitimate reason. However the H32 in a Natufian, H32 in Iran Neolithic, and H13c in Mesolithic Georgia don't support this idea(All close or over 10ky). My guess is your hypothesis H expanded with Near Eastern farmers began when most here thought there was a single Near Eastern farming population. Today we know there were at least two distinct early farming populations, we also know both carried H.

In the above quote you give a legitimate piece of evidence from modern mtDNA, but I just don't trust we can get many insights from modern mtDNA. I can't imagine H, which isn't that far away from early Eurasian R, could be so young. Already ancient Middle Eastern mtDNA is supporting an older age.

GailT
09-18-2016, 05:40 AM
That's a legitimate reason. However the H32 in a Natufian, H32 in Iran Neolithic, and H13c in Mesolithic Georgia don't support this idea(All close or over 10ky). My guess is your hypothesis H expanded with Near Eastern farmers began when most here thought there was a single Near Eastern farming population. Today we know there were at least two distinct early farming populations, we also know both carried H.

In the above quote you give a legitimate piece of evidence from modern mtDNA, but I just don't trust we can get many insights from modern mtDNA. I can't imagine H, which isn't that far away from early Eurasian R, could be so young. Already ancient Middle Eastern mtDNA is supporting an older age.

I agree that more testing of ancient samples is needed and this will provide a more precise date. The Behar estimates could be off by 10 to 30%, but probably not much more than that for major subclades with large numbers of samples.

The Georgia H13c samples is dated at 7940-7600 calBCE, or about 9800 years ago. Behar has age estimates of about 12,500 years for H13 and 11,400 years for H13c.

jeanL
09-18-2016, 04:27 PM
Behar et al. did the most comprehensive age estimates for H and its subclades Their age age estimate for H is about 13,000 years and for H13a1a about 8500 years. The papers that claim that H was present in Europe before the LGM were not based on sound science, rather, they were based on the mistaken belief that people did not migrate very much, and therefore the dominant haplogroup in Europe today most also have been in Europe before the LGM.

http://www.anthrogenica.com/showthread.php?8477-aDNA-from-Natufian-Hunters-to-Bronze-Age-farmers-(14-000-3-400-BP)&p=187028&viewfull=1#post187028

Sounds science(as in written in stone) and processes that are inherently random are quite mutually exclusive! Your Behar.et.al religion(because at this point is a religion, much like string theory for some physicists!) requires/makes Natufians and Iranian Neolithic quite the oddity placing them 2+ SD from the mean estimates.

jeanL
09-18-2016, 04:31 PM
It's contamination. H1 IMO can be explained by Eastern European EEF. Most of the EEF ancestors of Northern and SouthEstern Europeans probably lived around Ukraine and Romania and a signifcant but lesser amount for SouthWestern Europeans did. We have little DNA from Neolithic people of those countries. Once we do get their mtDNA it might be rich in H. We do have some and it is rich in H, and also lacks signatures of LBK mtDNA.

Neither H1 nor H3 have been found in Anatolia nor the Middle East, but we do have Portuguese Neolithic being H rich, we have early Navarra Neolithic being H(Including H3) rich and lacking mt-DNA Hg N. We have an mt-DNA H from Upper Paleolithic Vestonice 42. So all in all, the Middle East looks more like a recipient or perhaps the birth places of certain subclades of mt-DNA H. Which adds to the notion that mt-DNA H is quite old, not as old as U like the apples to oranges comparison that some here want to throw at us, but slightly younger than U5.

http://i1133.photobucket.com/albums/m582/jeanlohizun/1-s2_0-S1040618215000580-gr1_zpsjbxla2xu.jpg (http://s1133.photobucket.com/user/jeanlohizun/media/1-s2_0-S1040618215000580-gr1_zpsjbxla2xu.jpg.html)

http://i1133.photobucket.com/albums/m582/jeanlohizun/de.la.rua.et.al2015-Table-1_zpsyxcsojlp.jpg (http://s1133.photobucket.com/user/jeanlohizun/media/de.la.rua.et.al2015-Table-1_zpsyxcsojlp.jpg.html)

kingjohn
09-18-2016, 07:28 PM
you are right
Cardial Portugal Galeria da Cisterna, Almonda Cave [G 21] M 5330-5230 BC
H3
Olalde 2015

and burgundi before the celts gauls and franks
there was grandma ...... :)

RRBP France Gurgy Les Noisats [GLN 227, 246, 269]
[4900-4500 BC]

H3 3 samples 2706A 6776C 7028C 11719G 12705C 14766C Rivollat 2015
RRBP France Gurgy Les Noisats [GLN 214]
[4900-4500 BC]

H3 2706A, 6776C, 7028C, 11719G, 12705C, 14766C, 16235G Rivollat 2015
\

Epicardial Spain Avellaner cave, Catalonia [Ave 03] M 5000 BC G2a M287-, P15+, Giv in STR table H3 T6776C Lacan 2011b


rosen caulture germany no h3 but h1
and others

Rossen Germany Oberweiderstedt - Schrammhöhe [OSH 1]
4625-4250 BC

H16 G73A, C146T, C195T, A247G, A769G, A825t, A1018G, G2706A, A2758G, C2885T, T3594C, G4104A, T4312C, T7028C, G7146A, T7256C, A7521G, T8468C, T8655C, G8701A, C9540T, C10394T, G10398A, T10664C, A10688G, C10810T, C10873T, C10915T, A11719G, A11914G, T12705C, G13105A, G13276A, T13506C, T13650C, T14766C, A16129G, T16187C, C16189T, T16223C, G16230A, T16278C, C16311T Adler 2012; Brotherton 2013; Brandt 2013

Rossen Germany Oberweiderstedt [OSH 2]
4625-4250 BC

H89 G73A, C146T, C152T, C195T, A247G, A769G, A825t, A1018G, G2706A, A2758G, C2885T, T3594C, G4104A, T4312C, A6932G, T7028C, G7146A, T7256C, A7521G, C8068T, T8468C, T8655C, G8701A, C9540T, G10398A, T10664C, A10688G, C10810T, C10873T, C10915T, A11719G, A11914G, T12696C, T12705C, G13105A, G13276A, T13506C, T13650C, T14766C, A16129G, T16187C, C16189T, T16223C, G16230A , T16278C, C16311T Adler 2012; Brotherton 2013; Brandt 2013

Rossen Germany Oberweiderstedt [OSH 3]
4625-4250 BC
H1

also lbk

LBK Germany Karsdorf [I0797 / KAR 16] M
5500-4775 BC 95,833 T1a M70+
H46b

regards
Adam

Sammy Andrews
09-18-2016, 11:04 PM
Neither H1 nor H3 have been found in Anatolia nor the Middle East, but we do have Portuguese Neolithic being H rich, we have early Navarra Neolithic being H(Including H3) rich and lacking mt-DNA Hg N. We have an mt-DNA H from Upper Paleolithic Vestonice 42. So all in all, the Middle East looks more like a recipient or perhaps the birth places of certain subclades of mt-DNA H. Which adds to the notion that mt-DNA H is quite old, not as old as U like the apples to oranges comparison that some here want to throw at us, but slightly younger than U5.


There are H1s from the ancient Middle East; Bronze-age H1s from Armenia, but that's insignificant, because they weren't ancestral as far as we know to modern Europeans. Anyways I'm thinking H1 specfically was popular in Neolithic Eastern Europe. Today H3 is important only for Atlantic Europe, there's as much H1 in Lithuania as in Spain but essentially no H3, so where else could Lithuanian H1 be from except Neolithic Eastern Europe? I'm not very confident in this theory it's just something I'm considering.

Vestonice 42 was probably U, they gave two results one U and one H. Because of other UP mtDNA results, it is safe to assume Vestonice 42 had U. The Middle East definitly looks like the source of H not a receptor and not a source of only some H subclades. H's closest relatives; R0a, HV(xH), are today mostly rooted in the Middle East(xNorth Africa) and already well documented in the Stone age Middle East. So, we should expect H's origins to be somewhere in the Middle East.

J Man
09-19-2016, 12:37 AM
There are H1s from the ancient Middle East; Bronze-age H1s from Armenia, but that's insignificant, because they weren't ancestral as far as we know to modern Europeans. Anyways I'm thinking H1 specfically was popular in Neolithic Eastern Europe. Today H3 is important only for Atlantic Europe, there's as much H1 in Lithuania as in Spain but essentially no H3, so where else could Lithuanian H1 be from except Neolithic Eastern Europe? I'm not very confident in this theory it's just something I'm considering.

Vestonice 42 was probably U, they gave two results one U and one H. Because of other UP mtDNA results, it is safe to assume Vestonice 42 had U. The Middle East definitly looks like the source of H not a receptor and not a source of only some H subclades. H's closest relatives; R0a, HV(xH), are today mostly rooted in the Middle East(xNorth Africa) and already well documented in the Stone age Middle East. So, we should expect H's origins to be somewhere in the Middle East.

Are you still going to be active with your blog called mtDNA Atlas?

jeanL
09-19-2016, 01:51 AM
There are H1s from the ancient Middle East; Bronze-age H1s from Armenia, but that's insignificant, because they weren't ancestral as far as we know to modern Europeans. Anyways I'm thinking H1 specfically was popular in Neolithic Eastern Europe. Today H3 is important only for Atlantic Europe, there's as much H1 in Lithuania as in Spain but essentially no H3, so where else could Lithuanian H1 be from except Neolithic Eastern Europe? I'm not very confident in this theory it's just something I'm considering.

You said it, the H1 in ancient Middle East are much younger than those found in Europe, so they cannot be ancestral! We agree there.


Vestonice 42 was probably U, they gave two results one U and one H. Because of other UP mtDNA results, it is safe to assume Vestonice 42 had U. The Middle East definitly looks like the source of H not a receptor and not a source of only some H subclades. H's closest relatives; R0a, HV(xH), are today mostly rooted in the Middle East(xNorth Africa) and already well documented in the Stone age Middle East. So, we should expect H's origins to be somewhere in the Middle East.

You are thinking Vestonice 13, who was U5b when the analysis was restricted to damaged fragment, and H7 when it wasn't. Vestonice 42 was mt-DNA H. See here:

http://www.anthrogenica.com/showthread.php?7057-The-genetic-history-of-Ice-Age-Europe&p=155986&viewfull=1#post155986

jeanL
09-19-2016, 02:31 AM
BTW if we are to use Ancient DNA as calibration points, since H32 was found in:

I1690 NAT6 32.65 35.067 M Natufian Israel_Natufian This study (S1690.E1.L1) half 1240K.capture 11840-9760 BCE Raqefet Cave Israel 0.014 16866 mt-DNA HG: H32

Per the supplementary info of Behar.et.al.2012b (http://www.cell.com/cms/attachment/2024885281/2044552800/mmc1.pdf)

Page-80 gives the age of mt-DNA Haplogroup H as 12846.0 ybp with a standard deviation of 773.4 years.


Haplogroup------mean-age-(years)------SD (years)

H--------------------12846---------------773.4

This is the age that GailT used to argue against the authenticity of every mt-DNA Haplogroup H found in pre-Neolithic Europe.

Page-100 gives the age of mt-DNA Haplogroup H32

Haplogroup------mean-age-(years)------SD (years)

H32--------------------3437.0---------------3244.4

Page-76 gives the age of mt-DNA Haplogroup V

Haplogroup------mean-age-(years)------SD (years)

V----------------------9739-----------------1357

We have ancient sample:

I1069 Nat5 32.65 35.067 M Natufian Israel_Natufian This study (S1069.E1.L1) half 1240K.capture 11840-9760 BCE Raqefet Cave Israel 0.009 10288 mt-DNA HG: V

So let us start with mt-DNA Hg V:

Now comparing age differences we find that average age of mt-DNA H per Behar compares to average age of mt-DNA V as:

Average: H age/V age=12846 ybp/ 9739 ybp=1.32 times older.

Lower bound: H age/V age=(12846 ybp-1SD)/(9739+1SD)=(12846 ybp-773.4)/(9739+1357)=12072.6 ybp/11096 ybp=1.09 times older

Upper bound: H age/V age=(12846 ybp+1SD)/(9739-1SD)=(12846 ybp+773.4)/(9739-1357)=13619.4 ybp/8382 ybp=1.62 times older

Thus the age of Hg H per Behar.et.al.2012b compared to the age of age of Hg H32 is 1.32 times older with the +-1SD bounds being from 1.09 to 1.62 times older.

Lower bound estimate:

9760 BCE = 11710 ybp age of mt-DNA Hg V.

Thus the using the mean average value we will find age of mt-DNA Hg H is:

11710 ybp *1.32 =15457 ybp


Upper bound estimate:

11840 BCE = 13790 ybp age of mt-DNA Hg V.

Thus the using the mean average value we will find age of mt-DNA Hg H is:

13790 ybp *1.32 =18202 ybp


Now let us calibrate using mt-DNA H32 found in Natufians:

Now comparing age differences we find that average age of mt-DNA H per Behar compares to average age of mt-DNA H32 as:

Average: H age/H32 age=12846 ybp/ 3437.0 ybp=3.73 times older.

Lower bound: H age/H32 age=(12846 ybp-1SD)/(3437+1SD)=(12846 ybp-773.4)/(3437+3244.4)=12072.6 ybp/6681.4 ybp=1.81 times older

Upper bound: H age/H32 age=(12846 ybp+1SD)/(3437-1SD)=(12846 ybp+773.4)/(3437-3244.4)=13619.4 ybp/192.6 ybp=70.71 times older

Thus the age of Hg H per Behar.et.al.2012b compared to the age of age of Hg H32 is 3.73 times older with the +-1SD bounds being from 1.81 to 70.7 times older.

Lower bound estimate:

9760 BCE = 11710 ybp age of mt-DNA Hg H32.

Thus the using the mean average value we will find age of mt-DNA Hg H is:

11710 ybp *3.73 =43678 ybp

Using the lowest possible bound of +1SD for H32 and -1SD for mt-DNA H we find

11710 ybp*1.81 =21195 yb

Upper bound estimate:

11840 BCE = 13790 ybp age of mt-DNA Hg H32.

Thus the using the mean average value we will find age of mt-DNA Hg H is:

13790 ybp *3.73 =51437 ybp

Using the lowest possible bound of +1SD for H32 and -1SD for mt-DNA H we find

13790 ybp*1.81 =24960 ybp

Conclusions:

No way in hell that mt-DNA H is only 13000 ybp old. Enough said!!!!

Sammy Andrews
09-19-2016, 04:52 AM
Are you still going to be active with your blog called mtDNA Atlas?

I've been completely inactive for 3 months or so. I'll probably get back in it. I will probably post once a month because I'm in college now.

Sammy Andrews
09-19-2016, 04:53 AM
Are you still going to be active with your blog called mtDNA Atlas?

I've been completely inactive for 3 months or so. I'll probably get back in it. I will probably post once a month because I'm in college now.

Gravetto-Danubian
09-19-2016, 05:00 AM
I've been completely inactive for 3 months or so. I'll probably get back in it. I will probably post once a month because I'm in college now.

Congrats, what You studying ( if its not top secret)

What do you know about H47 ?

vettor
09-19-2016, 06:05 AM
you are right
Cardial Portugal Galeria da Cisterna, Almonda Cave [G 21] M 5330-5230 BC
H3
Olalde 2015

and burgundi before the celts gauls and franks
there was grandma ...... :)

RRBP France Gurgy Les Noisats [GLN 227, 246, 269]
[4900-4500 BC]

H3 3 samples 2706A 6776C 7028C 11719G 12705C 14766C Rivollat 2015
RRBP France Gurgy Les Noisats [GLN 214]
[4900-4500 BC]

H3 2706A, 6776C, 7028C, 11719G, 12705C, 14766C, 16235G Rivollat 2015
\

Epicardial Spain Avellaner cave, Catalonia [Ave 03] M 5000 BC G2a M287-, P15+, Giv in STR table H3 T6776C Lacan 2011b


rosen caulture germany no h3 but h1
and others

Rossen Germany Oberweiderstedt - Schrammhöhe [OSH 1]
4625-4250 BC

H16 G73A, C146T, C195T, A247G, A769G, A825t, A1018G, G2706A, A2758G, C2885T, T3594C, G4104A, T4312C, T7028C, G7146A, T7256C, A7521G, T8468C, T8655C, G8701A, C9540T, C10394T, G10398A, T10664C, A10688G, C10810T, C10873T, C10915T, A11719G, A11914G, T12705C, G13105A, G13276A, T13506C, T13650C, T14766C, A16129G, T16187C, C16189T, T16223C, G16230A, T16278C, C16311T Adler 2012; Brotherton 2013; Brandt 2013

Rossen Germany Oberweiderstedt [OSH 2]
4625-4250 BC

H89 G73A, C146T, C152T, C195T, A247G, A769G, A825t, A1018G, G2706A, A2758G, C2885T, T3594C, G4104A, T4312C, A6932G, T7028C, G7146A, T7256C, A7521G, C8068T, T8468C, T8655C, G8701A, C9540T, G10398A, T10664C, A10688G, C10810T, C10873T, C10915T, A11719G, A11914G, T12696C, T12705C, G13105A, G13276A, T13506C, T13650C, T14766C, A16129G, T16187C, C16189T, T16223C, G16230A , T16278C, C16311T Adler 2012; Brotherton 2013; Brandt 2013

Rossen Germany Oberweiderstedt [OSH 3]
4625-4250 BC
H1

also lbk

LBK Germany Karsdorf [I0797 / KAR 16] M
5500-4775 BC 95,833 T1a M70+
H46b

regards
Adam

LBK Germany Karsdorf [I0795 / KAR 6] M
5140-5050 BC T1a M70+
H1


Is this H1 older than iberian H1 ?

kingjohn
09-19-2016, 11:16 AM
i dont know just wanted to show
that some h subclades were present in neolithic europea from portugal -france -and inside to germany :)
the lbk guy who belong to mtdna h1 his y haplogroup T like you :)
regards
adam

kingjohn
09-19-2016, 06:29 PM
the karelia dude
i know most of them are U but still there is a single H
that should not be ignored {UZOO-77}
but as jean say this H COULD BE GENEFLOW TO HUNTERS FROM FARMERS SEEMS LIKELY
11742

and another food for a thought mtdna H in magdalenian cantabaria {mtdna h6}
11743

regards
adam

parasar
09-19-2016, 06:41 PM
the karelia dude
i know most of them are U but still there is a single H
that should not be ignored {UZOO-77}
but as jean say this H COULD BE GENEFLOW TO HUNTERS FROM FARMERS SEEMS LIKELY
11742

and another food for a thought mtdna H in magdalenian cantabaria {mtdna h6}
11743

regards
adam

And from nearby Upper Dvina too.

kingjohn
09-19-2016, 06:59 PM
the link to the cantabrian dude
http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0034417
if it is real ? jean m should put it here http://www.ancestraljourneys.org/mesolithicdna.shtml
dear parsar can you give me the source to the other H in the nearby upper dvina ?
regards
Adam

Sammy Andrews
09-19-2016, 07:16 PM
Congrats, what You studying ( if its not top secret)

What do you know about H47 ?

My major is History Education, I might change it. I know nothing about H47. It's one of the rare basal H subclades which hasn't been studied.

parasar
09-19-2016, 09:42 PM
the link to the cantabrian dude
http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0034417
if it is real ? jean m should put it here http://www.ancestraljourneys.org/mesolithicdna.shtml
dear parsar can you give me the source to the other H in the nearby upper dvina ?
regards
Adam
"Table 3. Results of genotyping of the material from archaeological sites of Upper Dvina region."
"The existence of mtDNA haplogroup H can be reliably traced in the forest zone of Eastern Europe —
in Upper Dvina region, basing on the materials of the site Serteya VIII (border of V-IV mill. BC), which testifies the appearance
of communities-bearers of this marker in post-glacial period.
https://www.academia.edu/9452168/Archaeology_of_lake_settlements_IV-II_mill._BC_Mazurkevich_A._Polkovnikova_M._Dolbuno va_E._ed

GailT
09-20-2016, 04:53 AM
Obviously the Behar estimates are much less reliable for subclades like H32 that had an extremely small number of samples when the paper was published in 2012.

U5b1c2 is another good example of an inaccurate date based on small sample size - Behar had an age estimate of about 4000 years,but we now have two ancient samples of U5b1c2 dated at 7000 years and 8600 years. This does not mean that the Behar estimate of U5 is off by a factor of two. In fact, the Behar estimates for U5 seem pretty accurate based on subsequent ancient samples.

In general the Behar dates are more reliable for older subclades that have large numbers of samples. I do think it is possible that H could be 10% to 30% older than the Behar estimate. I've never claimed that the Behar estimates are 100% accurate so please, enough with the "religion" insults. Eventually more testing of ancient remains will provide a more accurate estimate. I'll be surprised if any H samples are found that are older than 17,000 years, but I'll be also be glad if such samples are found because it would provide data that can be used to improve on the Behar estimates.

jeanL
09-20-2016, 01:28 PM
@ GailT How exactly are you deriving the 10% to 30% older estimate. Moreover; there was no long ago where I showed in this very board that to come up with such a narrow standard deviation like Behar.et.al did for mt-DNA H you needed a sample size of like 50,000. (http://www.anthrogenica.com/showthread.php?533-Upper-Paleolithic-Mesolithic-European-mtDNA-Haplogroups&p=44533&viewfull=1#post44533) So when you say you do think, please (for us folks from the scientific community) do tell us how do you derived those conclusions. Moreover, you might want to start using the verb feel, since it seems most of those thoughts are deriving from the limbic system not the frontal cortex.


I provided age estimates using the age-ratio from H/H32 and H/V, notice that mt-DNA Hg V doesn't have a small sample size, and it still produces a much older age that 13000 ybp for mt-DNA Hg H. See here:

http://www.anthrogenica.com/showthread.php?533-Upper-Paleolithic-Mesolithic-European-mtDNA-Haplogroups&p=187650&viewfull=1#post187650

Using mt-DNA Hg V found in Natufians as a calibration point we get the age of mt-DNA H as being between 15500 ybp to 18000 ybp. The lower bound is 32% higher than the age of mt-DNA H given by Behar, so your 10% to 30% doesn't stand its ground against facts!

The Behar.et.al.2012b age estimates are flawed for mt-DNA Hg H, that's the conclusion here. No amount of safety factor you might want to throw at it is going to fixed it. These are random processes(MCMC derived processes to be more specific), and they don't follow your standard 15% safety factor rule.

Finally: Vestonice42, its mt-DNA is clearly marked as Hg-H, it was not(coincidentally perhaps, perhaps there is something more) included in the main study because it is very low quality; but its mt-DNA passed all kinds of contamination test, and I believe it is one of the supplementary tables that list it as mt-DNA Hg H. Now to save us some time, this isn't Vestonice14 who was H7 but did not pass contamination estimates and then changed to U5b when restricted to damaged fragments. This is this fellow right here:

http://i1133.photobucket.com/albums/m582/jeanlohizun/Fu.et.al.2016-Table-S2.1_zpsikj8yojx.png (http://s1133.photobucket.com/user/jeanlohizun/media/Fu.et.al.2016-Table-S2.1_zpsikj8yojx.png.html)

So yes, it was found? Are you surprised already? Or is the cognitive dissonance still messing you up?


...I'll be also be glad if such samples are found because it would provide data that can be used to improve on the Behar estimates.

So glad you can't wait to use the estimates I provided up there to recalibrate the age of mt-DNA Hg H32 and mt-DNA Hg V as well as that of H right?

GailT
09-20-2016, 03:38 PM
you do think[/B], please (for us folks from the scientific community) do tell us how do you derived those conclusions. Moreover, you might want to start using the verb feel, since it seems most of those thoughts are deriving from the limbic system not the frontal cortex.


I'll be happy to respond when you commit to stop making offensive comments like those above. I don't insult you, and I expect the same treatment from you.