The DNA of the mesolithic hunter. Absent or present in northern European population?

epoch

epoch

Elite member
Messages
781
Reaction score
187
Points
0
Hi forum,

Recently genetic information from two mesolithic hunter-gatherer burials from La Brana in northern Spain has been published. I have read a number of articles about it and I am confused by the results. The main reason for my confusion lies in the fact that the la Brana skeleton seem related to northern Europe rather than the current day local Iberian population. The two carried mtDNA U5b2c1 which is not very abundant in current day Iberia - neither is any U5 - and the full sequence show relation to northern European population rather than local. The results are presented as evidence for a large population overhaul during the Neolithic transition in Iberia.

http://dienekes.blogspot.nl/2012/06/mesolithic-iberians-la-brana-arintero.html
http://eurogenes.blogspot.com.au/2013/12/brana-1-had-blue-eyes.html

Mesolithic and upper paleolithic mtDNA recovered until recently is mostly U4 and U5, and we do have quite a number of finds.

http://www.eupedia.com/europe/ancient_european_dna.shtml
http://www.ancestraljourneys.org/ancientdna.shtml


The autosomal genetic relation with northern Europe is all good and well were it that northern European population, which the la Brana autosomal DNA is clearly related to, would have carried lots U5. That is true for Finland, the Saami and the Baltic region - although in the latter U, U4 + U5 does not exceed the 21% - but it isn't true for most of the northwestern European culture, including Sweden, Denmark and Norway. There have been Farmers from the neolithic have been sequenced as well as hunter-gatherers from the Pitted Ware culture, found on Gotland. Mitochondrial DNA showed greet difference between the mainland Swedes and the hunter-gatherers, presented as evidence for.. you guessed it: a population overhaul in Sweden after the arrival of the neolithic.

http://ac.els-cdn.com/S096098220901...t=1387632483_de9e8b5f85e80f8936c82a48bacac689

However, autosomal DNA between the northerners with and those without a strong U5 presence hardly differs. Also, phenotypes are remarkably similar. And the autosomal admixture is mostly north European in both Finnish and Swedish samples. Even both the la Brana and the Gotland Pitted Ware samples show autosomical relation to Swedish and Saami rather than Saami.

http://1.bp.blogspot.com/-F9XoyN2wE-w/ToZBZZT7jdI/AAAAAAAAAmI/qwmGu64Xvwk/s1600/admixture.png
http://fennoscandia.blogspot.nl/2013/09/la-brana-2-and-modern-european-variation.html
http://fennoscandia.blogspot.nl/201...n-variation-ii.html?showComment=1383942706763

This is very strange. Autosomal evidence points to relation between mesolithic hunter-gatherers and current day northwest Europeans, whereas mtDNA points to a population overhaul. We pretty much know mesolithic and paleolithic hunter-gatherers had mostly mtDNA U (U, U4, U5) which is not very common nowadays.

How can this both be true? How can the northwest European population - a population which wikipedia describes as almost genetically indistinguishable - bear such genetic resemblance to hunter-gatherers if they aren't related to them, as the great difference in mtDNA suggests?

More links:

http://forwhattheywereweare.blogspot.nl/2010/11/ancient-mtdna-advance-maps.html
http://eurogenes.blogspot.nl/2013/02/post-mesolithic-population.html
 
epoch said:
The two carried mtDNA U5b2c1 which is not very abundant in current day Iberia - neither is any U5 -
Click to expand...
Where do you get that ? Because as far as I know, northern Iberia has one of the highest peaks of U5 in Europe (the Cantabria-Basque area), reaching near 20%
 
Wilhelm said:
Where do you get that ? Because as far as I know, northern Iberia has one of the highest peaks of U5 in Europe (the Cantabria-Basque area), reaching near 20%
Click to expand...
1


The area is a strategic route that was also used and fought and over during the migration period.
Im not claiming it a fact but a possibility for the origin.
 
Wilhelm said:
Where do you get that ? Because as far as I know, northern Iberia has one of the highest peaks of U5 in Europe (the Cantabria-Basque area), reaching near 20%
Click to expand...

I got this from this

http://www.eupedia.com/europe/european_mtdna_haplogroups_frequency.shtml

However, I now see the map Maciamo made and indeed it shows up to 20% in some northern regions.

http://www.eupedia.com/forum/threads/29157-New-map-of-mtDNA-haplogroup-U5

The again, it hardly changes my point that current day U, U5 and U4 is only a minority of the mtDNA lineages in most of the rest of Spain and most of Northern Europe. Still, autosomically the la Brana samples show to be related to the northern Europeans. How can that be? Dienekes had an article on how the neolithic societies absorbed the mesolithic hunters, and claims that only a minority of current day maternal lines stem from mesolithic hunters. If that would be so, how come the Swedes, which hardly have any U5, show such strong affinity with mesolithic hunters?

http://dienekes.blogspot.nl/2012/03/neolithic-expansions-how-european.html

Mind you, it is very interesting what you said, since the genetic outlier in Iberia is the west Pyrenees, where the Basque live. A population that also shows more northern and north western affinity. Or am I missing something or misreading something?
 
epoch said:
I got this from this

http://www.eupedia.com/europe/european_mtdna_haplogroups_frequency.shtml

However, I now see the map Maciamo made and indeed it shows up to 20% in some northern regions.

http://www.eupedia.com/forum/threads/29157-New-map-of-mtDNA-haplogroup-U5

However, it hardly changes my point that current day U, U5 and U4 is only a minority of the mtDNA lineages in most of the rest of Spain and most of Northern Europe. However, autosomically the la Brana samples a related to the northern Europeans. How can that be? Dienekes had an article on how the neolithic societies absorbed the mesolithic hunters, and claims that only a minority of current day maternal lines stem from mesolithic hunters. If that would be so, how come the Swedes, which hardly have any U5, show such strong affinity with mesolithic hunters?

http://dienekes.blogspot.nl/2012/03/neolithic-expansions-how-european.html

Mind you, it is very interesting what you said, since the genetic outlier in Iberia is the west Pyrenees, where the Basque live. A population that also shows more northern and north western affinity. Or am I missing something or misreading something?
Click to expand...

The recent alpine paper for south Tyrol, trentino and northern veneto has over 200 Mtdna tested, with U5 being second in percentage after K1. Since this area is otzi area, we can guess the iberic burials might have the same markers
 
Sile said:
The recent alpine paper for south Tyrol, trentino and northern veneto has over 200 Mtdna tested, with U5 being second in percentage after K1. Since this area is otzi area, we can guess the iberic burials might have the same markers
Click to expand...

But Otzi was far less related to north Europe. Even the farmers at Gokhem showed more relation to local hunter-gatherers.
 
epoch said:
But Otzi was far less related to north Europe. Even the farmers at Gokhem showed more relation to local hunter-gatherers.
Click to expand...

Wasn't the Gokhem in scandinavia burial a female?
 
I red somwhere mt DNA was under influence of natural selection...? what is not amazing considering it this "guest" DNA has a metabolic function in cells
 
MOESAN said:
I red somwhere mt DNA was under influence of natural selection...? what is not amazing considering it this "guest" DNA has a metabolic function in cells
Click to expand...

There are finds of Pitted Ware burials, that lived alongside farmers. They show the same amount of U as mesolithic hunter-gatherers.
 
epoch said:
http://eurogenes.blogspot.com/2013/12/ancient-human-genomes-suggest-three.html?showComment=1387851513616632050721459316

There you have it. North Western admixture is a mesolithic inheritence. Or signal. Or whatever you want to call it. This finding may also vindicate the Germanic Substrate theory and explain the continuity seen in Ertebolla and Funnel Beaker Culture.
Click to expand...

From Eurogenes...


Loschbour - Y-DNA I2a1bMotala 2 - Y-DNA I
Motala 3 -Y-DNA I2
Motala12 - I2a1b

Looks like the samples are I2a1b, putting a dent in the idiotic "i2a1b-din is only 2000 years old" nonsense that Albanians and Ken N. parroted for years.
 
james stock said:
From Eurogenes...


Loschbour - Y-DNA I2a1bMotala 2 - Y-DNA I
Motala 3 -Y-DNA I2
Motala12 - I2a1b

Looks like the samples are I2a1b, putting a dent in the idiotic "i2a1b-din is only 2000 years old" nonsense that Albanians and Ken N. parroted for years.
Click to expand...

You don't know what you're talking about. Neither sample was I2a-Din. Loschbour was confirmed I2a1b* L621- L161-. The L621- means not I2a-Din, because I2a-Din is L621+. Nordtvedt places I2a1b as a whole at about 16k years old, so no contradiction. Similarly, Motala12 was I2a1b L621-, and hence not I2a-Din. It could have been I2a-Isles L161+ in theory, but could also have been I2a1b* like Loschbour. Again, no contradiction.
 
You don't know what you're talking few put. Neither sample was I2a-Din. Loschbour was confirmed I2a1b* L621- L161-. The L621- means not I2a-Din, because I2a-Din is L621+. Nordtvedt places I2a1b as a whole at about 16k years old, so no contradiction. Similarly, Motala12 was I2a1b L621-, and hence not I2a-Din. It could have been I2a-Isles L161+ in theory, but could also have been I2a1b* like Loschbour. Again, no contradiction.
Click to expand...

Yes, I2a1b l178 was found amonst the samples. I2a1b l178/m423 is dated at 8k years +/- 4k years. We now have samples at 8k years that confirms l178 is present so the lower range of l178 of 4-8k years is now thrown out. The same can probably be said about i2a1b din.

Similarily , it's only a matter of time until the models are irrelevant because we have accumulated enough data through fossils to actually date each haplogroup.
 
james stock said:
Yes, I2a1b l178 was found amonst the samples. I2a1b l178/m423 is dated at 8k years +/- 4k years. We now have samples at 8k years that confirms l178 is present so the lower range of l178 of 4-8k years is now thrown out. The same can probably be said about i2a1b din.
Click to expand...

Who dates M423/L178 at 8k years? Nordtvedt dates the TMRCA of all known branches at ~12k and its overall clade age (which is the important one for testing in this case) at ~16k, as I said. See here.

james stock said:
Similarily , it's only a matter of time until the models are irrelevant because we have accumulated enough data through fossils to actually date each haplogroup.
Click to expand...

It will even be possible to validate existing dating models once we have enough ancient DNA. As is, any validation gives too large error bars.
 
sparkey said:
Who dates M423/L178 at 8k years? Nordtvedt dates the TMRCA of all known branches at ~12k and its overall clade age (which is the important one for testing in this case) at ~16k, as I said. See here.



It will even be possible to validate existing dating models once we have enough ancient DNA. As is, any validation gives too large error bars.
Click to expand...

It doesn't seem that you understand how models work. If we create a model that is completely predictive of all known data, it will not give us certainty that that same model will be predictive of unknown data. Even if the model predicts unknown data, it is not proof that the model will predict further unknown data. If a model predicts all unknown data that became known, it is not proof that the model will predict future data. It is therefore impossible to validate a model.
 
james stock said:
It doesn't seem that you understand how models work. If we create a model that is completely predictive of all known data, it will not give us certainty that that same model will be predictive of unknown data. Even if the model predicts unknown data, it is not proof that the model will predict further unknown data. If a model predicts all unknown data that became known, it is not proof that the model will predict future data. It is therefore impossible to validate a model.
Click to expand...

You're reaching an absurdity because your definition of "validation" is too strict. We wouldn't be looking for proof that a model will predict all unknown data, we would be looking for proof that our model has correctly predicted known data, which would give evidence toward whether or not such a model would predict unknown data. You're basically arguing that you can't train a predictive algorithm, which is almost always false.
 
sparkey said:
Who dates M423/L178 at 8k years? Nordtvedt dates the TMRCA of all known branches at ~12k and its overall clade age (which is the important one for testing in this case) at ~16k, as I said. See here.



It will even be possible to validate existing dating models once we have enough ancient DNA. As is, any validation gives too large error bars.
Click to expand...

sparkey said:
You're reaching an absurdity because your definition of "validation" is too strict. We wouldn't be looking for proof that a model will predict all unknown data, we would be looking for proof that our model has correctly predicted known data, which would give evidence toward whether or not such a model would predict unknown data. You're basically arguing that you can't train a predictive algorithm, which is almost always false.
Click to expand...
No, what I was saying was that models will always be prone to error and that they cannot be validated. There may be periods that have rapid mutation, aka Ghould's punctuated equilibrium, and mutaion rates in the future will most certainly be in the hands of genetic engineers.

Certain random events are impossible to model.
 
james stock said:
No, what I was saying was that models will always be prone to error and that they cannot be validated. There may be periods that have rapid mutation, aka Ghould's punctuated equilibrium, and mutaion rates in the future will most certainly be in the hands of genetic engineers.

Certain random events are impossible to model.
Click to expand...

The idea that Y-line mutation rates aren't too random to model is an assumption that we'd be validating, though. That's the point. If the model consistently predicts the age of ancient samples across time and place, then tada, we've validated. If it gets some groups wrong and others right, or systematically gets all groups wrong but in different ways, then that assumption is incorrect, and we'll need to adjust or train our model somehow to be able to cope with the apparent randomness. If it still doesn't work for other groups coming in, then we can assume that it is too random to model. But not before then.
 
You must log in or register to reply here.

This thread has been viewed 28117 times.

Back
Top