G2a in Sardinia and the Sherden

[Alan]

I agree with you about Sardinia/Sherdana naming of the island - the known name bearers can, as in a lot of other places, be carried there by a late enough arrival of dominant people - concerning the other parts of your post, I have no sound opinion : I wrote it was difficult to determine the different waves and their respective weight because even with an autosomal "sketche" of the population, we can't be sure than a componant was not carried there by more than a (cultural/etnic) wave, more than a time, cause a lot of colonizators came from the same geographical areas or from someones very close... by exception, here, the Y and mt Hg deeply studied could give us more accurate answers for different waves of same source.
I'm not sure of nothing here; I bet just that the majority of autosomals of Sardinians came there at Mesolithic times and NOT at Neolithic times (but neolithic could have brought some of them too), but it's just a bet - more than a scenario is possible?

I understand what you mean but we know from Neolithic and Mesolithic samples that only the farmers are predominantly Mediterranean and the Hunter and Gatherers are North European like. So I dont believ we can assume that this component reached Europe during Mesolthic.

The thing is that Europe lacks West Asian like admixture till the Bronze age.

[/all the way, the 'west-asian component' is somehow distinct from the 'gedrosian' one (concerning the I-E link)- I 'll try to study better the different poolings of autosomals proposed for today populations and the FEW pre-bronze people autosomals we have (almost nothing!) in Europe
good afternoon!

Actually Gedrosia is much more similar to the original West Asian component than Caucasus. Dienekes did break up of the components on his blog.

Gedrosia/Caucasus are jsut two parts of the West Asian component. And the Gedrosia component seems more West Asian like than Caucasus.


Here is Dienekes break up of the components.





Gedrosia is 90% West Asian + 10% ANI South Asian
Caucasus is 55% West Asian + 40% Southern and 5% Atlantic Baltic.

http://dienekes.blogspot.de/2012/09/inter-relationships-between-dodecad-k7b.html


have a good night.

 

[ElHorsto]

I understand what you mean but we know from Neolithic and Mesolithic samples that only the farmers are predominantly Mediterranean and the Hunter and Gatherers are North European like.

What you probably mean is not North_European but Atlantic_Baltic. If I remember correctly, the hunter-gatherer from La Brana was only half North_European and half Atlantic_Med (K12b). Only the Ajv samples from Sweden are mostly North_european, but they are not so relevant since they were found very far in the north-east already. But even they already had some Atlantic_med admixture (approx. 10-20 %). I'm not saying that Atlantic_Baltic is false, just that it seems to me less of a pure component than North_european.

So I dont believ we can assume that this component reached Europe during Mesolthic.

Based on the agruments above I'd argue that earlier mesolithic southern/mediterranean admixtures in southern and western hunter-gatherers are not so unlikely.

 

[MOESAN]

What you probably mean is not North_European but Atlantic_Baltic. If I remember correctly, the hunter-gatherer from La Brana was only half North_European and half Atlantic_Med (K12b). Only the Ajv samples from Sweden are mostly North_european, but they are not so relevant since they were found very far in the north-east already. But even they already had some Atlantic_med admixture (approx. 10-20 %). I'm not saying that Atlantic_Baltic is false, just that it seems to me less of a pure component than North_european.

Based on the agruments above I'd argue that earlier mesolithic southern/mediterranean admixtures in southern and western hunter-gatherers are not so unlikely.

I agree with you: everyone has his arguments here:
but as you I noitice some 'mediterranean' (not too well defined) componant among hunters-gatherers of far Sweden, even if light, and heavier among the one of Spain (so, before neolithic!) - I note too what I consider as marks of "archaïc' remnant from Asia and Africa among the swede hunter-gatherers!!! maybe the african part was mixed with a sub-componant of 'mediterranean'? (mesolithic?) -
concerning the problem of lack or 'west-asian' among the swede farmer I think that farming was introduced in Scandinavia as on other European shores (Atlantic for the most) by 'long-barrows-' akin people (atlantic neolithic, inspired by cardial maybe but creating their own culture after , not the continental danubian neolithic where I see Y-G2 associated to Y-E1b-V13, at first - and, leaved apart the problem of boosting elite, I think that these megalithers beared a lot of 'mediterranean', and no 'west-asian' -
all that if the so called H-G-s was not yet crossed with some neolithic females... but the %s seam showing that it would have been old, what I don't believe...

 

[Knovas]

The Mediterranean clusters are the most remote one's overall according to Dienekes', so concerning West Eurasia they are the best defined among all.

There you are from the latest Globe13 analysis: http://dodecad.blogspot.com.es/2012/10/globe13-participant-results.html

Bra1 already shows 25% of this component. I know the SNPs tested are very low, but this isn't exactly what I would call a minor component. So in my opinion it's likely that the so called Mediterranean component was already present in Southwest Europe during the Mesolithic, and who knows if also along the rest of Southern Europe.

http://dienekes.blogspot.com.es/2012/10/assessment-of-ancient-european-dna-with.html

 

[MOESAN]

The Mediterranean clusters are the most remote one's overall according to Dienekes', so concerning West Eurasia they are the best defined among all.

There you are from the latest Globe13 analysis: http://dodecad.blogspot.com.es/2012/10/globe13-participant-results.html

Bra1 already shows 25% of this component. I know the SNPs tested are very low, but this isn't exactly what I would call a minor component. So in my opinion it's likely that the so called Mediterranean component was already present in Southwest Europe during the Mesolithic, and who knows if also along the rest of Southern Europe.

http://dienekes.blogspot.com.es/2012/10/assessment-of-ancient-european-dna-with.html

I agree concerning elderness 'mediter' in Europe, before neolithic introduction (maybe the "second mesolithic", rather than the "first mesolithic"?: two candidates for the first mesolithic: a paleolithic conitnuity + an early mesolithic introduction from East, yet) -
for the definition of 'mediterranean' you can see, reading the link you send me (thanks) that according to the branching 'systhem', respective positions of 'mediter', 'west-asian', 'south-west-asian' and 'north-european' are not so clear -
I would be glad if I could knew how these branchings are made concerning phylology, aside the geographic aspect (I suppose it is difficult)-

 

[ElHorsto]

A related article:
"Anthropologist suggests Mediterranean islands inhabited much earlier than thought"

http://phys.org/news/2012-11-anthropologist-mediterranean-islands-inhabited-earlier.html

So basically it claims that pre-neolithic humans were able to travel by sea to mediterranean islands.

(found at Dienekes Blog)

 

[ElHorsto]

I meant only the first part of the article, because the second part subjects pre-Homo-Sapiens.

 

[Knovas]

Moesan, it's clear when you check the Fst distances spreadsheet: the Mediterranean cluster is the most distant from both inner Africa and East Asia (including Oceania). That's why in the picture the rest of components seem to deviate from it.

What we can see is:
- Noth European is more African and East Asian shifted compared to Mediterranean.
- Southwest Asian is more African shifted compared to Mediterranean and North European.
- West Asian is more African and East Asian shitfted compared to Mediterranean. At the same time, this component is more African shifted than North European (being similar regarding Asia), and more Asian shifted than Southwest Asian (being similar regarding Africa).

That's a good summary overall I think.

 

[ElHorsto]

- Noth European is more African and East Asian shifted compared to Mediterranean.

I'm not an expert, but I don't see African shift in North European:
North European deviates towards Amerindian (~East Asian), yes, but not African. By aligning a straight line between Mediterranean and Amerindian you will see that North European is placed at this line and even a bit more shifted to the left. But if NE would be African shifted, it would appear right from this line. By euclidean measures it seems closer to African, but that's only because Amerindian has similar Y-position like African, not X-position. If I understood these eigenvectors(?) correctly, then absolute distances alone can be misleading if deviation-vectors are not regarded.

 

[Knovas]

The picture shows a general trend, but to see the exact distances you have to use this: https://docs.google.com/spreadsheet/ccc?key=0ArAJcY18g2GadF9CLUJnTUdSbkVJaDR2UkRtUE9kaUE#gid=3

That's what I referred, and it's clear that Mediterranean is less African, even if not significantly. Dienekes' blogged about this before, I think I posted the comment in another thread, and he admitted that the Mediterranean related clusters were the most remote ones, even concerning African ancestry.

 

[Yetos]

El horsto in this forum I read somewhere that in North Europe exist NWest African mtDNA which pass from an Atlantic road. I don't think that these female population went there recently.

 

[ElHorsto]

El horsto in this forum I read somewhere that in North Europe exist NWest African mtDNA which pass from an Atlantic road. I don't think that these female population went there recently.

Yes, I know, and it should affect both, North-European and Mediterranean components. After all, both components are almost similarly distant from african components. Still, it is too early to conclude that these components (e.g. NE) are based on African admixture. African admixture has been shown as extra component in both Bra1 and Ajv. As far as I know, Fst-distances do not distinguish between similarity by admixtures and similarity by drift. However, thanks Knovas for clarifying by providing the Fst-table.

 

[Knovas]

Fst-distances do not distinguish between similarity by admixtures and similarity by drift.

They do. If you check the Southwest Asian component compared to Mediterranean, you'll see the former is more distant than the later concerrning East Asian groups, even when in the picture Southwest Asian could look closer. Imagine that we add more "Asian" to the so called Southwest Asian, and the component would cluster West Asian in the picture. So the distances reflect things how really are.


However, there you are what Dienekes blogged, which is very ilustrative and expalins why some admixtures did not affected this kind of components (or did it in a lesser degree):

As for the African/Sub-Saharan components, they tend to be closer to the Southwest Asian/Red Sea components, not the Mediterranean/Atlantic_Med one.

The Mediterranean components appear to be the most remote ones overall (also evidenced by the fact that Basques and Sardinians nearly always form the peak in the West/East Eurasian/African triangle), which makes sense since the region where the Mediterranean/Atlantic_Med component is modal is most remote from both Africa and Asia along the land migration routes.

 

[Knovas]

By the way, the North African MtDNA mentioned above refers to U6 subclades. It's true that U6 in North Africa is likely Mesolithic and even Paleolithic in age there, but this in my opinion reflects the ancient fingerprints of Cro-Magnons, so it had nothing to do with African ancestry at first. We don't know if U6 came from North Africa being already mixed with some sort of African, or if U6 in Europe represents minor Paleolithic remnants among other maternal lines, such as U5 for instance. All possibilities are open.

 

[ElHorsto]

By the way, the North African MtDNA mentioned above refers to U6 subclades. It's true that U6 in North Africa is likely Mesolithic and even Paleolithic in age there, but this in my opinion reflects the ancient fingerprints of Cro-Magnons, so it had nothing to do with African ancestry at first. We don't know if U6 came from North Africa being already mixed with some sort of African, or if U6 in Europe represents minor Paleolithic remnants among other maternal lines, such as U5 for instance. All possibilities are open.

I agree very much. It is also noteworthy that African admixture in Bra1 turned out to be east african, not west african as one might expect.

 

[ElHorsto]

They do. If you check the Southwest Asian component compared to Mediterranean, you'll see the former is more distant than the later concerrning East Asian groups, even when in the picture Southwest Asian could look closer. Imagine that we add more "Asian" to the so called Southwest Asian, and the component would cluster West Asian in the picture. So the distances reflect things how really are.

Makes sense.

However, there you are what Dienekes blogged, which is very ilustrative and expalins why some admixtures did not affected this kind of components (or did it in a lesser degree):

As for the African/Sub-Saharan components, they tend to be closer to the Southwest Asian/Red Sea components, not the Mediterranean/Atlantic_Med one.

The Mediterranean components appear to be the most remote ones overall (also evidenced by the fact that Basques and Sardinians nearly always form the peak in the West/East Eurasian/African triangle), which makes sense since the region where the Mediterranean/Atlantic_Med component is modal is most remote from both Africa and Asia along the land migration routes.

That's what I expected too. I just was surprised that the proximity of North-European to Africa is relatively low. After all, NE and Med appear to have almost the same distance from Africa, while it is not surprising that NE is closer to Asian/Amerindian than Med. is.

 

[Alan]

I agree with you: everyone has his arguments here:
but as you I noitice some 'mediterranean' (not too well defined) componant among hunters-gatherers of far Sweden, even if light, and heavier among the one of Spain (so, before neolithic!) - I note too what I consider as marks of "archaïc' remnant from Asia and Africa among the swede hunter-gatherers!!! maybe the african part was mixed with a sub-componant of 'mediterranean'? (mesolithic?) -
concerning the problem of lack or 'west-asian' among the swede farmer I think that farming was introduced in Scandinavia as on other European shores (Atlantic for the most) by 'long-barrows-' akin people (atlantic neolithic, inspired by cardial maybe but creating their own culture after , not the continental danubian neolithic where I see Y-G2 associated to Y-E1b-V13, at first - and, leaved apart the problem of boosting elite, I think that these megalithers beared a lot of 'mediterranean', and no 'west-asian' -
all that if the so called H-G-s was not yet crossed with some neolithic females... but the %s seam showing that it would have been old, what I don't believe...

I think farming must have been introduced from the east since the Mediterranean component is linked almost exclusively to Haplogroup G2a which is very low in Western Europe and farming on it self originated from the east. We know that non farmer population of Europe are highly North European like, they probably absorbed some Mediterranean during Neolithic from the Farmers. Also note that during Neolithic Bulgaria was also populated by Sardinian like population. It looks much more plausible that the original Mediterranean Farmer component was replaced by a new wave of migration. This is why the traces of farming dna is very weak in East Europe but high in more isolated areas of West and Southwest Europe. Simply because they could withstand or preserve themselves from a new invading group (Indo-Europeans?)

 

[Knovas]

I agree very much. It is also noteworthy that African admixture in Bra1 turned out to be east african, not west african as one might expect.

The East African it's pretty insignificant considering the low number of SNPs available, and I think Dienekes' does not consider this real. On the other hand, East African is linked to West Eurasia, don't know if this could help on altering the results in this case.

But yours Horsto was a good point, mentioning that if U6 was responsible of the African shift in the Northern European cluster, then it should have also affected the so called Mediterranean in a similar degree. Who knows, the possibility that U6 was in Europe even before that in North Africa it's still alive. So doesn't matter which possibility is the correct one, since Alan probably gives part of the answer: another migration affecting Northern Europeans but not Southern Europeans made the difference, and that's why the Northern European cluster appears closer. It worth mentioning that if Mediterranean was in Europe during the Mesolithic, then the classical Neolithic farming theory only matches the increase of this component, not the introduction. Whatever it was the reason of the African shift in the Northern European cluster compared to Mediterranean, this occurred long before the Indo-European invasions, and I don't have a more specific explanation than the one given by Dienekes'. Obviously something happened, but don't know what...I don't see now U6 was determinant, but maybe I'm missing something.

Actually, Northern Europeans as whole are Asian shifted, while Southern Europeans tend to be African shifted (Basques and Sardinians are peculiar as we all know and do not match this). But we should also consider other admixtures regarding this.

 

[Wilhelm]

I also do believe some kind of mediterranean was already in Europe in Mesolithic, the one found in La Brana and the Swedish Hunter-Gatherers...whereas the west-asian and SW-Asian components only appear in the Neolithic remains, so this latter is definately a neolithic marker but not Mediterranean (well, East-Med kind of med might be neolithic, but it appears that a great part of the Atlanto-mediterranean or Western-Mediterranean type of med was already there in pre-neolithic).

 

[Kardu]

I think it's curious to mention that there is a striking similarity between Sardnian(and Corsican) and Georgian traditional polyphonic singing...

Similarity of lullabies, with a refrain Nana/Nanina is most peculiar. The refrain must be an echo of a prayer to the Mother Goddess Nana/Inana. So the connection probably dates from the period when Nana was venerated in Asia Minor.
(Nana appears in Greek lullabies as well)