Sardinian Y-DNA Phylogeny per Francalacci et al. 2013

[Angela]

This is now the second study in a row that confirms R1b-U152 to be ~10% in Sardinia;

Boattini et al 2013 [82 samples] = 9.7% R1b-U152
http://www.plosone.org/article/info:doi/10.1371/journal.pone.0065441
Francalacci et al 2013 [1200 samples] = 10.7% R1b-U152
(current study)

Needless to say that only the Roman times and the Medieval times can be the source for this
substantial (every tenth Sardinian) amount of R1b-U152 in Sardinia;


I think the main source however; is def. the Medieval times;

David Abulafia - The New Cambridge Medieval History: Vol.V (1999)
As has been seen, Castel di Castro or Cagliari was founded by the Pisans in 1217, with imposing fortifications; and, even though settlement by Sards was prohibited, it rapidly grew to contain a population of several thousand (between 7,000 and 10,000 at the end of the thirteenth century).....In the same judgeship, in the south-west of Sardinia, Count Ugolino founded around 1250 the city of Iglesias, which was to grow within a couple of generations to a population not far below 10,000.....Another town which experienced growth was the reborn centre at Olbia, on the north-east coast of Sardinia; Civita or Terranova was described in a Pisan document as a quasi civitas.....This urbanisation was rendered possible by the emigration from the mainland, especially from Pisa and its contado;

Pisa and Genoa conquered Sardinia in 1016; The Pisans were greatly crippled by the Genoese at the Battle of Meloria (1284) and 40 years later lost control over Sardinia; Thats over 300 years with massive urbanisation (Tuscan migrations) across the 13th cen.

So in terms of R1b-U152
- what the medieval Lombards are for Sicily the medieval Tuscans are for Sardinia;
NW Italy (medieval Lombardy) = 32.2% R1b-U152 [161 samples] - Boattini et al 2013
Tuscany = 37.4% R1b-U152 [123 samples] - Boattini et al 2013

If the Francalacci study gave geographical coordinates for their samples, and a lot of the U-152 was from, say, that southern coastal area around Cagliari, it would tie it up nicely.

It would have been interesting to see, also, if a lot of that U-152 was from the northern coast where they speak the Corsican dialect. (Gallurese) Corsican is, of course, usually considered closest to Tuscan, although since Corsica was ruled for so long by Genova, there is some Ligurian influence as well. One thing that could be done, perhaps, would be to check the snps for U-152 in Corsica versus those in Sardegna.

It would also have been nice to see what the y dna is like in Alghero, where they still speak Catalan, especially since the historical records say that the natives (some mixture of Sardi, Liguri, Toschi) were expelled.
http://en.wikipedia.org/wiki/File:Sardinia_Language_Map.png

 

[Maciamo]

Both Oetzi and Gok 4 had about 7% S.W.Asian, and therefore it's extremely likely that it was indeed part of the genomic structure of the Neolithic farmers who made their way into Europe.
http://dienekes.blogspot.com/2012/04/first-look-at-dna-of-neolithic.html

Yes, you are right. My mistake. I got mixed up with the K=12, K=12a and K=12b.

In the K=12b you cited the admixtures should be as follows:

- Sardinians have 2.6% of Northwest African, 5.8% of Southwest Asian, 20.9% of Caucasian and 70.5% of Atlantic-Med.
- Lebanese have 4.7% of Northwest African, 23.5% of Southwest Asian, 41.3% of Caucasian and 11.8% of Atlantic-Med.
- Ötzi has 5.7% of Northwest African, 7.6% of Southwest Asian, 22.3% of Caucasian and 57.7% of Atlantic-Med.

I will re-calculate the proportion of Phoenician vs Neolithic autosomes in Sardinians later.

 

[Knovas]

It would also have been nice to see what the y dna is like in Alghero, where they still speak Catalan, especially since the historical records say that the natives (some mixture of Sardi, Liguri, Toschi) were expelled.
http://en.wikipedia.org/wiki/File:Sardinia_Language_Map.png

If true, then R1b would be clearly dominant (as it is in Catalonia). Interesting point.

 

[Angela]

Yes, you are right. My mistake. I got mixed up with the K=12, K=12a and K=12b.

In the K=12b you cited the admixtures should be as follows:

- Sardinians have 2.6% of Northwest African, 5.8% of Southwest Asian, 20.9% of Caucasian and 70.5% of Atlantic-Med.
- Lebanese have 4.7% of Northwest African, 23.5% of Southwest Asian, 41.3% of Caucasian and 11.8% of Atlantic-Med.
- Ötzi has 5.7% of Northwest African, 7.6% of Southwest Asian, 22.3% of Caucasian and 57.7% of Atlantic-Med.

I will re-calculate the proportion of Phoenician vs Neolithic autosomes in Sardinians later.

These kinds of percentages highlight why I don't think modern Levantine populations are a very good proxy for the Neolithic migrants from West Asia.

Ed. Sorry, that was off-topic.

 

[Angela]

The Ligures are a very very interesting point;
That truly deserves a thread of their own (still waiting on Taranis); - In full detail and explanation;

In whatever scenario the Ligures are a source for R1b-U152 (and i do believe they can be considered one)
It can not apply to Sardinia - since Sardinia was never settled by the Ligures;

Sardinia was mostly settled (pre-Roman 238BC) by Iberians, Phoenicians, Libyans and of course the local Mountaineer tribes - Parati / Sossinati / Balari / Aconites
who Strabo claims to be of a mythical backround - Iolaus (sons of Herakles);
and Pausanias mentions Trojans in Sardinia;

In this respect it is interesting to note that Bronze-age Sardinians clustered closest to Minoans and Bronze-age Iberians

Hughey et al 2013
http://www.nature.com/ncomms/journal/v4/n5/full/ncomms2871.html
http://www.nature.com/ncomms/journal/v4/n5/fig_tab/ncomms2871_T1.html

Whereas modern-day Sardinians are isolated and cluster the closest to neolithic corpses;


The Ligures as a source for R1b-U152 can only apply to Corsica and East Sicily (Siculi);

The U-152 in northern Sardegna could still be connected to Ligures as a hypothesis; it would just mean that the Corsicans were the intermediaries. The U-152, if any, found in Cagliari in the south could more directly be linked to Toscana (Pisa area to be precise.)

Just as an aside, Liguria was influenced by more than one wave of people up to late in the first centuries B.C., when, genetically, things in most of Italy, with the exception perhaps of the south, seem to have frozen. Liguria is home, of course, to famous Paleolithic sites, Neolithic sites, (Cardial on the coast, but of a different variety at the mountainous margins) and then you have the mysterious Ligures, who, in addition to whatever they were carrying in terms of y dna and autosomally, absorbed the prior inhabitants. The final major movement of which I'm aware was from Gallic tribes. If I were to speculate, I would say that is the explanation for some of the more "western" clades of R1b, like DF27 and L-21.

As for Sicily, the same thing applies as applies to Sardegna. The U-152, could, as some have hypothesized, be of Ligure origin, but just mediated by the "Lombard" settlers of medieval Sicily. "Lombards" in that context, were not only people from Lombardia, but also Liguria and Toscana.

If I'm not mistaken, the clustering with Minoans and Iberians for the Sardinians which you mention above only has to do with uniparental markers. (Crete was settled multiple times from the Levant and West Asia, from the Neolithic to the Bronze Age.)

We don't have autsomal analysis for either of these groups. The only autosomal analyses for European Neolithic farmers that we have are from Oetzi and Gok 4. Oetzi is indeed closest to Sardinians. Gok 4 not quite as much. If I recall correctly, Dienekes did an IBD analysis showing Gok 4's similarity to Greeks. I believe Dienekes also posted about Gok 4's similarity to northern Italians, (based on the overall ancestry calculators) but I didn't think his Gok 4 calculator was very successful.

 

[Nobody1]

If the Francalacci study gave geographical coordinates for their samples, and a lot of the U-152 was from, say, that southern coastal area around Cagliari, it would tie it up nicely.

It would have been interesting to see, also, if a lot of that U-152 was from the northern coast where they speak the Corsican dialect. (Gallurese) Corsican is, of course, usually considered closest to Tuscan, although since Corsica was ruled for so long by Genova, there is some Ligurian influence as well. One thing that could be done, perhaps, would be to check the snps for U-152 in Corsica versus those in Sardegna.

It would also have been nice to see what the y dna is like in Alghero, where they still speak Catalan, especially since the historical records say that the natives (some mixture of Sardi, Liguri, Toschi) were expelled.
http://en.wikipedia.org/wiki/File:Sardinia_Language_Map.png

100% agree;
If the study would have chopped the samples acc. to geographic region it would have been a much more clearer insight - Historically;

The 82 Sardinian samples from Boattini et al 2013 came from the provinces Olbia, Nuoro and Oristano;
Olbia/Nuoro [40 samples] = 12.5% R1b-U152
Oristano [42 samples] = 7.1% R1b-U152

Looks more frequent on the Eastern Coast than Western Coast;
unfortunately no samples from the Southern Coast or Northern Coast

Boattini et al 2013 - TableS2 (Supporting Information)
http://www.plosone.org/article/info:doi/10.1371/journal.pone.0065441


Alghero would be very interesting also for admixture; maybe they are the only "Sardinians" (Aragonese/Catalans) that cluster with others (+levels of inter-mixture with the local Sardinians over the centuries);
Y-DNA i would expect a lot (or decent amount) of R-SRY2627;

 

[Maciamo]

100% agree;
If the study would have chopped the samples acc. to geographic region it would have been a much more clearer insight - Historically;

The 82 Sardinian samples from Boattini et al 2013 came from the provinces Olbia, Nuoro and Oristano;
Olbia/Nuoro [40 samples] = 12.5% R1b-U152
Oristano [42 samples] = 7.1% R1b-U152

Looks more frequent on the Eastern Coast than Western Coast;
unfortunately no samples from the Southern Coast or Northern Coast

Boattini et al 2013 - TableS2 (Supporting Information)
http://www.plosone.org/article/info:doi/10.1371/journal.pone.0065441


Alghero would be very interesting also for admixture; maybe they are the only "Sardinians" (Aragonese/Catalans) that cluster with others (+levels of inter-mixture with the local Sardinians over the centuries);
Y-DNA i would expect a lot (or decent amount) of R-SRY2627;

Yes, that's really a shame Francalacci didn't at least mention which places were sampled for the study. Anyway there wasn't a single R1b-SRY2627 out of the 1200 samples tested, so it is possible that there were no sample from Catalan-speaking Alghero.

 

[Mikewww]

.. In whatever scenario the Ligures are a source for R1b-U152 (and i do believe they can be considered one) It can not apply to Sardinia - since Sardinia was never settled by the Ligures;

I wouldn't be quite so certain (as Angela noted earlier) of that as U152 is not the dominant haplogroup in Sardinia so there would be no requirement to have a large Ligurian presence and the flow could have come from Corsica. There are theories that the Ligures are connected with the Iberians. If you have people in NW Italy, SE France and Iberia connected, it's hard to believe none would have touched Sardinia.

However, let us set aside the Ligures for a moment. My memory was wrong on what I thought Rocca had said or at least I can't find what he posted. More recently, this May, he wrote,

"The earliest branches of U152 probably expanded somewhere in coastal SE France or NW Italy during the Late Copper Age. From there, there was a secondary expansion that occurred during the Bell Beaker "reflux" period. This expansion probably gave rise to Z36 somewhere in the central Alpine passes between Italy and Switzerland and L2 somewhere in the Eastern Bell Beaker Province (S. Germany, E. Switzerland, Bohemia, Hungary) and so it seems to have a 'Germanic' look to it. As part of the reflux period, NE Italy was heavily influenced by that province too (see Begleitkeramik) and formed the launch point for the Polada Culture. Many of the areas where L2 makes up a large percentage of U152 also shows U106 in important numbers.

Z56 is a little tougher to figure out. Based on some off-modal values, it seems to have occurred after Z36 and L2. Its most important frequency probably lies somewhere between Tuscany and Modena. Z56 could have expanded with the Terramare Culture.

Of course all subsequent expansions from those areas would have scattered the big three subclades (Urnfield, Hallstatt, La Tene, Romans). L2's main subclade (Z367 and its subclade L20) probably expanded with the RSFO Urnfield Culture in France and made its way into the isles with La Tene." ​

http://www.anthrogenica.com/showthr...ly-branches-expand&p=6230&viewfull=1#post6230

Of course the bulk of the R1b-U152 in Sardinia is Z192+ which is L2- Z56- Z36-. Other than Sardinia, we have found Z192 in Tuscany, Mexico and Ireland. Yes, it could be Roman related, but as Richard cited above, U152 could easily have been in several major cultures and these ancient cultures ringed the Ligurian Sea. U152 could easily have reached Sardinia several times in prehistory and leading up to the Roman Empire era.

I make these points in response to your post,

... Needless to say that only the Roman times and the Medieval times can be the source for this substantial (every tenth Sardinian) amount of R1b-U152 in Sardinia...

I just don't see how you can demonstrate the Roman and Medieval times as the only periods of U152 immigration into Sardinia.

 

[Nobody1]

@ Mikewww

Would be good to know whether Mr. Rocca considers U152 to be an Indo-European or a pre-Indo-European lineage;
By the looks of it he has no problem assigning U152 to every single culture zone from the Chalcolithic to the Iron-age; regardless of historical backround;


I consider U152 to be Indo-European and strongly associated with the Umbrians [Terremare and Urnfield (Villanova/Golasecca)] and with the proto-Kelts [Tumulus and Urnfield (Hallstatt A-B)]
With U152 being a substantial element among the proto-Kelts/Kelts and the most dominant element amongst the Umbrians (ITALICS);

A common link/root that the Keltics and Italics share within the Indo-European realm is also given in Linguistics;

David Rankin - Celts and the Classical World (1986)
As in Celtic, so also in Italic, /p/—/k w/ becomes / kw/—/k w/ and /p/—/p/ according to dialect, as in 'quinque' and 'Pompeius'. The 'b' future, which is found in Latin and other Italic dialects, occurs also in Celtic: both Celtic and Italic retain the /samo/ suffix for the superlative degree of adjectives. These are some of the obvious points of resemblance. There are considerably more. In the range of correspondences, Latin seems to have points markedly in common with q-Celtic, whereas Oscan and Sabellian seem to be closer to p-Celtic. These and other factors could suggest an earlier geographical continuity between the ancestors of Italic and Celtic speakers.


As for Sardinia;
Based on the Historic events on Sardinia; i personally do not see any other source that could justify 10.7% U152 in moder-day Sardinia other than the vast Tuscan migrations of the medieval times (13th cen) -[post #36]

 

[Sile]

Both Oetzi and Gok 4 had about 7% S.W.Asian, and therefore it's extremely likely that it was indeed part of the genomic structure of the Neolithic farmers who made their way into Europe.
http://dienekes.blogspot.com/2012/04/first-look-at-dna-of-neolithic.html

G2a3a (M406) forms up to 50% of the G2a in the eastern Mediterranean. It is found very commonly in southern Italy. Various people have tried to tie its presence there to Greek and Balkan expansion, but it could be Neolithic, of course. Interesting that it's so low in Sardegna.

But Maciamo is stating G-L91 is north-African and Otzi is confirmed G-L91. How is a north-african have 7% SW Asian?

 

[Maciamo]

But Maciamo is stating G-L91 is north-African and Otzi is confirmed G-L91. How is a north-african have 7% SW Asian?

I believe that G-L91 could have been spread by the Neolithic farmers who hopped from Greece to Italy to the Maghreb and to Mediterranean Spain and France. Just have a look at my map of the diffusion of agriculture.

The Southwest Asian and Caucasian admixtures would have been part of the original genome of Levantine farmers. But it seems that they also carried genes that were mistakenly reported as Northwest African admixture, perhaps because they have survived better in Northwest Africa today.

Another possibility would be that those are the genes linked to E-M81 and some other E1b1b lineages that spread from the Levant to southern Europe and North Africa. Another migration, probably in the Late Palaeolithic, would have brought E-M81 both to Northwest Africa and to the Levant, explaining how this admixture.

 

[Nobody1]

Bekada et al 2013
http://www.plosone.org/article/info:doi/10.1371/journal.pone.0056775#close

G = [M201,M285,P15,P16,M406]

Sahara/Mauritania = 0% G --- [189 samples]

Morocco = 0.7% G --- [760 samples]

Algeria = 0% G --- [156 samples]

Tunisia = 0.2% G --- [601 samples]

Libya = 0% G --- [83 samples]

Egypt = 5.7% G --- [370 samples]


It looks like (in a Neolithic farming context) the most highest amount G in N Africa is in fertile Egypt

Since this study encompasses M201,M285,P15,P16,M406 as G
- im not sure how much of it is L91 an how much isnt;


Haplogroup G

 

[Maciamo]

Bekada et al 2013
http://www.plosone.org/article/info:doi/10.1371/journal.pone.0056775#close

G = [M201,M285,P15,P16,M406]

Sahara/Mauritania = 0% G --- [189 samples]

Morocco = 0.7% G --- [760 samples]

Algeria = 0% G --- [156 samples]

Tunisia = 0.2% G --- [601 samples]

Libya = 0% G --- [83 samples]

Egypt = 5.7% G --- [370 samples]


It looks like (in a Neolithic farming context) the most highest amount G in N Africa is in fertile Egypt

Since this study encompasses M201,M285,P15,P16,M406 as G
- im not sure how much of it is L91 an how much isnt;


Haplogroup G

Egypt is the closest to the source of the Neolithic (the Levant) and indeed the most fertile. But haplogroup G is almost certainly not the only one spread by Neolithic farmers. There was also E1b1b and perhaps also some J and T.

What's interesting in the case of Sardinia is that 60 out of 63 J1 samples belong to the J1-P58 subclade, and among them 58 belong to the Semitic L147.1 subclade, the main Arabic cluster and the Jewish Cohanim haplotype. That is why I think it can only be Phoenician in origin. The potentially Neolithic J1 found in central or northern Europe typically belongs to the other, rarer subclades (Z2223, M365.1, Z1828), which are also found in Anatolia and the Caucasus.

 

[Nobody1]

I also view the Phoenician/Punic input in Sardinia to be very noteworthy;

Martin Goodman - The Roman World 44 BC - AD 180 (1997)
Latin rights were gradually extended to the inhabitants of the existing cities during the first century AD, but still in the second century a neo-Punic inscription set up in one town described local magistrates as suffetes, and cultic inscriptions reveal the continued worship of Punic divinities only thinly disguised behind Roman names.

The Phoenician colonies in Sardinia remained strongly Phoenician/Punic
all the way into Roman times (2nd cen. AD)

Phoenician colonies - foundations began ~800 BC

Certain Levantine lineages such as the J1 sub-clade you describe can be attributed to the sole Phoenician context & the North African lineage E-M81 within the total Punic (Carthage) context - ancient-Berbers (Libyans)


The constant link with Sardinia and the East Mediterranean Sea-Peoples (especially the Sherden) also needs a closer look;
The Phoenician Stele of Nora (9th cen BC) describes the island as Shardan (while the Greeks called it Ichnusa)

I. E. S. Edwards - The Cambridge Ancient History (1975)
In the earliest Phoenician inscription found in Sardinia, that from Nora, probably of the ninth century B.C., although it is incomplete, the name of the island appears as Shardan (be-shardan), and thus the identification of Sardinia with the Sherden seems much strengthened.

Stele of Nora (9th cen BC) - Phoenician SHRDN [Shardan]

Pausanias mention of Trojans and Strabo mention of Iolaus and the Children of Herakles also point to the East Mediterranean;
What part exactly the mysterious Sea-Peoples played on Sardinia can not (prob. never) be fully reconstructed;

 

[Sile]

Egypt is the closest to the source of the Neolithic (the Levant) and indeed the most fertile. But haplogroup G is almost certainly not the only one spread by Neolithic farmers. There was also E1b1b and perhaps also some J and T.

What's interesting in the case of Sardinia is that 60 out of 63 J1 samples belong to the J1-P58 subclade, and among them 58 belong to the Semitic L147.1 subclade, the main Arabic cluster and the Jewish Cohanim haplotype. That is why I think it can only be Phoenician in origin. The potentially Neolithic J1 found in central or northern Europe typically belongs to the other, rarer subclades (Z2223, M365.1, Z1828), which are also found in Anatolia and the Caucasus.

Professor Jobling of England states. on T haplgroup
"The haplogroup has probably been present for centuries in the 'indigenous' population of western Europe," says Professor Jobling (of University of Leicester), "and is not exclusive to the Middle East and Africa." [10] According to limited data from commercial testing, of the European nations, men in Italy may have the highest frequency of haplogroup T, with as many as 3.9% of Italian males belonging to this haplogroup.
Appears to follow a Megalithic and civilized pattern, could be Dinarics/Atlanto-Mediterraneans.
The highest concentration of T is in Italy and according to professor Jobling its likely of Upper Paleolithic European derivation and he pushes for a European(which includes the central and Mediterranean(including Iberia). So its either that or a signature of the Italo-Romans as Wales was an important area.

with this link
https://docs.google.com/viewer?a=v&...&sig=AHIEtbTCXSjQab5Zs3_614593zeCnkndFQ&pli=1

It must have come earlier
The Upper Paleolithic (or Upper Palaeolithic, Late Stone Age) is the third and last subdivision of the Paleolithic or Old Stone Age as it is understood in Europe, Africa and Asia. Very broadly, it dates to between 50,000 and 10,000 years ago,

 

[Robert6]

Among the 15 R1a1a individuals were 11x Z282 (among which 5x Z280, 1x M458>L1029 and 5x probably M458>L260) and 4x Z93 (including 3x Z94>L342.2>Z2123). The Z2123 is Middle Eastern or Central Asian and could have been brought either by the Phoenicians or the Alans.

The R1a-Z2123 possibly is there from Aghlabids
https://commons.wikimedia.org/wiki/File:Aghlabids_Dynasty_800_-_909_(AD)_-_Arabic.svg?uselang=ru

The Aghlabids are from Al-Tamimi(Banu Tamimi)

ALTamimi cluster within Z2123

9. ...>Z93>Z94>Z2124>Z2125>Z2123-D-x Arab cluster (another Big Y needed)
M6982 need earliest known paternal ancestor info Kuwait R-M417 13 24 16 11 11-11 12 12 10 13 11 29
343784 Qatar R-M512 13 24 16 11 11-11-12-14 12 12 10 13 11 29
178906 A Yousif ALTamimi Saudi Arabia R-M512 13 25 16 10 11-12 12 12 10 13 11 29
M7066 Kuwait R-M512 13 25 16 10 11-12 12 12 10 13 11 29
162855 A Ali ALTamimi Saudi Arabia R-M512 13 25 16 10 11-14 12 12 10 13 11 29
178907 ALAmrawi ALTamimi Saudi Arabia R-M512 13 25 16 10 11-14 12 12 10 13 11 29
178905 ALAmrawi ALTamimi Saudi Arabia R-M512 13 25 16 10 11-14 12 12 10 13 11 29
160271 Aal Qatar R-M417 13 25 16 10 11-14 12 12 10 13 11 29
157103 BniAmr ALTamimi Saudi Arabia R-M198 13 25 16 10 11-14 12 12 10 13 11 29
M6895 Al-Tamimi Saudi Arabia R-M512 13 25 16 10 11-14 12 12 10 13 11 29
157619 BniAmr ALTamimi Saudi Arabia R-M512 13 25 16 10 11-14 12 12 10 13 11 29
157621 BniAmr ALTamimi Saudi Arabia R-M512 13 25 16 10 11-14 12 12 10 13 11 29
M7052 United Arab Emirates R-M512 13 25 16 11 11-11 12 12 10 13 11 29
M7067 United Arab Emirates R-M512 13 25 16 11 11-11 12 12 10 13 11 29
M7059 Kuwait R-PAGES00007 13 25 16 11 11-11 12 12 10 13 11 29
M6458 need earliest known ancestor info Kuwait R-Z94 13 25 16 11 11-14 12 12 10 13 11 29
M6183 F. Bo alhossain Kuwait R-M512 13 25 16 11 11-14 12 12 10 13 11 29
M6679 Kuwait R-M512 13 25 16 11 11-14 12 12 10 13 11 29
M6285 Ahmed Mohamed Kabeer Qatar R-M512 13 25 16 11 11-14 12 12 11 13 11 29
M6459 United Arab Emirates R-M512 13 25 16 13 11-14 12 12 10 13 11 29
M7013 Kuwait R-M512 13 26 16 10 11-11 12 12 10 13 11 29

 

[Hauteville]

Someone can explain me why there are some A1b and R2a?