A Genome-Wide Study of Modern-Day Tuscans: Revisiting Herodotus's Theory on the Origi

Angela

Elite member
Messages
21,823
Reaction score
12,325
Points
113
Ethnic group
Italian
I just gave this paper a quick run through. I may change my mind after a more thorough analysis, but I think they may have gotten it wrong. Again...even with whole genomes. It's not to say that the conclusion is necessarily incorrect. It's just to say that their analysis doesn't prove it.

You would think that by now some of these academics would have read and absorbed Lazaridis et al, and realized that you can't use modern populations to analyze ancient gene flow. Nor can use use IBS segments. IMO, you need ancient DNA, and absent that, some really sophisticated IBD analysis.

Here's the link to the paper.
http://www.plosone.org/article/info:doi/10.1371/journal.pone.0105920
 
You would think that by now some of these academics would have read and absorbed Lazaridis et al, and realized that you can't use modern populations to analyze ancient gene flow. Nor can use use IBS segments. IMO, you need ancient DNA, and absent that, some really sophisticated IBD analysis.

Well hopefully they will soon, so that we won't get a confusing inaccurate and outdated study every month that just gives wrong information and confuses everyone.
 
Here are some gems of generalization from the paper:

"however, almost all the studies agree that there is a proportion of their mtDNA pool that could be traced to somewhere in the Middle East, thus testifying to an ancient connection between both regions."

I have news for these guys: so does the mtDna of 80% of Europeans.

Then they resurrect this old chestnut: both humans and cattle reached Etruria from the Eastern Mediterranean area by sea. Hence the Eastern origin of Etruscans, first claimed by the classic historians Herodotus and Thucydides, receives strong independent support

Yes, plants, cattle, sheep, goats, and the knowledge of what to do with them all came from the eastern Mediterranean. That doesn't prove that a specific breed came in 800 BC!

As to ancient Etruscan mtDna see:
Silvia Ghirotto et al 2013
http://www.plosone.org/article/info:doi/10.1371/journal.pone.0055519
See the discussion at the Dienekes site:
http://dienekes.blogspot.com/2013/02/etruscan-mtdna-origins-ghirotto-et-al.html
As a knowledgeable poster explained, the Etruscan mtDna in this large sample is a mixture of U5 and J. Looks like admixture betwee Neolithic and Mesolithic peoples to me...

Also, these guys should know by now that without resolution of mtDna on a very detailed subclade level, and some agreement on mutation rates, it's impossible to use mtDna to track migration flows precisely.

As for the Armenian connection...

Of course there's more "Caucasus" like ancestry in Tuscans than in northern Italians, but the question is, when did this Caucasus like ancestry arrive in Europe? I'll buy Metal Ages...Oetzi had a little bit already, but this doesn't prove that a specific migration in 800 BC brought it!

Also, there's more in southern Italians yet, and more in Greeks. There's the same amount in Balkan people. Were they all settled in the first millennium BC by people from Anatolia?

Take a look at the "Caucasus" proportion in these groups as per the Dienekes K-12b analysis"
https://docs.google.com/spreadsheet...EY4Y3lTUVBaaFp0bC1zZlBDcTZEYlE&hl=en_US#gid=0
Northern Italians: 22.9
Romanians: 28.4
Tuscans: 30.5
Bulgarians: 30.7
Southern Italians/Sicilians: 36.5
Greeks: 37.4

Do I think it's possible that there was a late movement from Anatolia (first millenium BC) into Central Italy? Yes, I do; it's just that neither mtDNA, ancient or modern, is going to prove it, and especially not at the level of resolution which currently exists. A comparision of the full genomes of modern Tuscans to other modern populations doesn't prove it either. Who says this isn't Neolithic era? We need a high quality Etruscan genome and the genome of prior inhabitants from the same area.

I do think that there's generally something to be said for an additional Bronze Age gene flow (and later) into Italy not only from the north, but also from the south east.

I saw a speculation on Anthrogenica this morning about the proto-Indo-Europeans possibly carrying a lot of "Anatolian" like or "Near Eastern" like ancestry. (The mtDna of the Yamnaya people looks totally "Neolithic" from the evidence of that thesis. Perhaps the full genomes will show something different, but then that would mean that they left all their own women at home, or they didn't like them very much.)

It would certainly be interesting if Dienekes was onto something back in the day when he speculated that the proto-Indo-Europeans had their origin in the Armenian Highlands. Perhaps they remained more "Anatolian" like or Caucasus like in the southern regions, even if in the more northern areas they were different.
 
I just gave this paper a quick run through. I may change my mind after a more thorough analysis, but I think they may have gotten it wrong. Again...even with whole genomes. It's not to say that the conclusion is necessarily incorrect. It's just to say that their analysis doesn't prove it.

You would think that by now some of these academics would have read and absorbed Lazaridis et al, and realized that you can't use modern populations to analyze ancient gene flow. Nor can use use IBS segments. IMO, you need ancient DNA, and absent that, some really sophisticated IBD analysis.

Here's the link to the paper.

The paper predicts about 25% ME admixture among Tuscans which isn't really any different from results 3 years ago when Dienekes first opened up the Dodecad project using Admixture. (Just add West Asian with SW Asian)

I'm anticipating that the "first farmers" brought some ME admixture to Europe, but then subsequent waves of eastern people brought more ME admixture. YDNA J2 comes to mind as the strongest candidate for the later waves and the real differentiation between north and southern Europe.
 
Here are some gems of generalization from the paper:

"however, almost all the studies agree that there is a proportion of their mtDNA pool that could be traced to somewhere in the Middle East, thus testifying to an ancient connection between both regions."

I have news for these guys: so does the mtDna of 80% of Europeans.

Then they resurrect this old chestnut: both humans and cattle reached Etruria from the Eastern Mediterranean area by sea. Hence the Eastern origin of Etruscans, first claimed by the classic historians Herodotus and Thucydides, receives strong independent support

Yes, plants, cattle, sheep, goats, and the knowledge of what to do with them all came from the eastern Mediterranean. That doesn't prove that a specific breed came in 800 BC!

As to ancient Etruscan mtDna see:
Silvia Ghirotto et al 2013
http://www.plosone.org/article/info:doi/10.1371/journal.pone.0055519
See the discussion at the Dienekes site:
http://dienekes.blogspot.com/2013/02/etruscan-mtdna-origins-ghirotto-et-al.html
As a knowledgeable poster explained, the Etruscan mtDna in this large sample is a mixture of U5 and J. Looks like admixture betwee Neolithic and Mesolithic peoples to me...

Also, these guys should know by now that without resolution of mtDna on a very detailed subclade level, and some agreement on mutation rates, it's impossible to use mtDna to track migration flows precisely.

As for the Armenian connection...

Of course there's more "Caucasus" like ancestry in Tuscans than in northern Italians, but the question is, when did this Caucasus like ancestry arrive in Europe? I'll buy Metal Ages...Oetzi had a little bit already, but this doesn't prove that a specific migration in 800 BC brought it!

Also, there's more in southern Italians yet, and more in Greeks. There's the same amount in Balkan people. Were they all settled in the first millennium BC by people from Anatolia?

Take a look at the "Caucasus" proportion in these groups as per the Dienekes K-12b analysis"
https://docs.google.com/spreadsheet...EY4Y3lTUVBaaFp0bC1zZlBDcTZEYlE&hl=en_US#gid=0
Northern Italians: 22.9
Romanians: 28.4
Tuscans: 30.5
Bulgarians: 30.7
Southern Italians/Sicilians: 36.5
Greeks: 37.4

Do I think it's possible that there was a late movement from Anatolia (first millenium BC) into Central Italy? Yes, I do; it's just that neither mtDNA, ancient or modern, is going to prove it, and especially not at the level of resolution which currently exists. A comparision of the full genomes of modern Tuscans to other modern populations doesn't prove it either. Who says this isn't Neolithic era? We need a high quality Etruscan genome and the genome of prior inhabitants from the same area.

I do think that there's generally something to be said for an additional Bronze Age gene flow (and later) into Italy not only from the north, but also from the south east.

I saw a speculation on Anthrogenica this morning about the proto-Indo-Europeans possibly carrying a lot of "Anatolian" like or "Near Eastern" like ancestry. (The mtDna of the Yamnaya people looks totally "Neolithic" from the evidence of that thesis. Perhaps the full genomes will show something different, but then that would mean that they left all their own women at home, or they didn't like them very much.)

It would certainly be interesting if Dienekes was onto something back in the day when he speculated that the proto-Indo-Europeans had their origin in the Armenian Highlands. Perhaps they remained more "Anatolian" like or Caucasus like in the southern regions, even if in the more northern areas they were different.


Angela, I agree with the most of your observations: mt DNA needs very deep analysis before sending some basis to theory - Near-Eatsern and Anatolia-Caucasus has been the center of diffusion of a lot of cultures and populations, northwards, westwards and eastwards, at more than a time
 
That said, the homophony of several names as Iberians or Sardes from East to West could very well be more than the fruit of hazard...
 
The paper predicts about 25% ME admixture among Tuscans which isn't really any different from results 3 years ago when Dienekes first opened up the Dodecad project using Admixture. (Just add West Asian with SW Asian)

I'm anticipating that the "first farmers" brought some ME admixture to Europe, but then subsequent waves of eastern people brought more ME admixture. YDNA J2 comes to mind as the strongest candidate for the later waves and the real differentiation between north and southern Europe.

The point, surely, is that these authors have not proved their claim? Not that it matters to me personally in the long run whether they are "native" or from Anatolia so long as I match them a bit when one of their genomes is finally sequenced! I just dislike bad science.

As for admixture analyses, I am no longer enamored of them, despite the fact that I used one upthread. :) Even before Lazaridis et al, it was clear that they obscure ancient genetic flows as much or perhaps more than they elucidate them. They are just geographical "poolings" to steal a term that I very much like. Or perhaps they are like artifacts of the algorithm in a way.

In terms of the genetics of Tuscans, they are, according to Lazaridis et all, approximately 75% EEF, 14% WHG, and 11% ANE.
View attachment 6657


Also according to the Lazaridis analysis, Tuscans can be modeled as an admixture of Sardinians and ANE. That makes total sense to me.

Furthermore, as we know, Sardinians are the closest we have to EEF. See this comment from Figure 2 PCA Analysis in the paper finally published a few days ago.
'Stuttgart clusters with other NeolithicEuropeans and present-day Sardinians"

I definitely have said, and I believe, that J2 in Italy is late Neolithic at the earliest, and probably Bronze Age or later. Other than the proposed direct flow from Anatolia for the Etruscans, it would have been mediated through the Greeks, imo, or at least I don't know of any other possible source in terms of Tuscany, other than perhaps by way of the Balkans.

We won't know what the proportions were in the various post Stuttgart/Oetzi migrations to the area until ancient samples from those migration areas are analyzed. In that context, I'm very interested to see the results for the Samara population, especially at the point when they are proposed to be moving toward Europe, and also, of course, for a full analysis of the Yamnaya population. It would be nice to finally get one from the Sea Peoples, perhapsm or people from the Balkans at various stages.

I don't think it's very helpful to talk about the" Middle East". The Middle East from which ancient man first went to Europe? The Middle East at the time of the early Neolithic? The Middle East of today, by which you perhaps mean the Levant? The Near East of today? I don't think so as far as modern Middle Easterners or Near Easterners are concerned, no disrespect to them.

If we've learned anything, surely we've learned that the people who came to Europe bringing farming were not the same as the people in the Near or Middle East today, largely, in terms of the Near East, because of their large modern day proportions of ANE, (almost 30% in the Caucasus and at high levels everywhere) and in case of the Levant, of SSA gene flow. That's what I was alluding to when I suggested that perhaps these authors hadn't fully absorbed the implications of the Lazaridis et al analysis.

Perhaps it's better, as Moesan alluded to, to think of it as a question of repeated gene flows from the area of the Caucasus.

Ed. I think it's also interesting to consider the case of the Portalon Iberian farmer. His remains date to 2,000 BC and yet he clusters very close to modern Tuscans.

See:
http://dienekes.blogspot.com/2013/11/iberian-neolithic-farmer-dna.html

Other bloggers have opined on the subject.
 
Here are some gems of generalization from the paper:

"however, almost all the studies agree that there is a proportion of their mtDNA pool that could be traced to somewhere in the Middle East, thus testifying to an ancient connection between both regions."

I have news for these guys: so does the mtDna of 80% of Europeans.

LOL, you're talking about haplogroup H right?



Do I think it's possible that there was a late movement from Anatolia (first millenium BC) into Central Italy? Yes, I do; it's just that neither mtDNA, ancient or modern, is going to prove it, and especially not at the level of resolution which currently exists. A comparision of the full genomes of modern Tuscans to other modern populations doesn't prove it either. Who says this isn't Neolithic era? We need a high quality Etruscan genome and the genome of prior inhabitants from the same area.

So what you're saying is that only a full genome comparison of modern Tuscans and Etruscans + previous ancient peoples may be able to prove that or indeed anything concerning recent origins (also in connection to all populations including AJs)?
 
I don't think it's very helpful to talk about the" Middle East". The Middle East from which ancient man first went to Europe? The Middle East at the time of the early Neolithic? The Middle East of today, by which you perhaps mean the Levant? The Near East of today? I don't think so as far as modern Middle Easterners or Near Easterners are concerned, no disrespect to them.

If we've learned anything, surely we've learned that the people who came to Europe bringing farming were not the same as the people in the Near or Middle East today, largely, in terms of the Near East, because of their large modern day proportions of ANE, (almost 30% in the Caucasus and at high levels everywhere) and in case of the Levant, of SSA gene flow. That's what I was alluding to when I suggested that perhaps these authors hadn't fully absorbed the implications of the Lazaridis et al analysis.

Bingo, and, maybe, just maybe, that's why according to Lazaridis AJs don't plot in the modern near east, but in the gap between Europe and the near east, just a little closer to Europe than Stuttgart, alongside Maltese and Sicilians, and, not major admixture with a European population (although the high IBD sharing with modern Greeks, the gene flow from them and the apparent genetic continuity of Greeks still doesn't make me put that off the table, so I guess I'll have to patiently wait for a full genome comparison with ancient populations).

Link to supposed genetic continuity among Greeks:
http://dienekes.blogspot.co.il/2014/08/craniofacial-morphology-of-greeks.html
 
I think that subject must expand to even north of Italy, to val Camunnico,
personally I believe from toponyms, that they were from the same material,
 
IMO there must have been some immigration, but not massive, of some people with some better knowledge, probably of metal smelting than the local Italic and Celtic tribes at the time , they founded a new society of which they themself where the elite
the immigration was so small, it is hardly detectable in the DNA
you can ask the same question about copper age southern Iberia, some 5200 years ago : who founded Los Millares and introduced copper smelting ?
and southern Iberia 4000 years ago : who founded La Bastida and later created El Argar cities and started bronze smelting ?

these are things we'll never know in detail, it takes just a small elite tribe with a specific knowledge and the right organizational skills to have a very big local impact
 
IMO there must have been some immigration, but not massive, of some people with some better knowledge, probably of metal smelting than the local Italic and Celtic tribes at the time , they founded a new society of which they themself where the elite
the immigration was so small, it is hardly detectable in the DNA
you can ask the same question about copper age southern Iberia, some 5200 years ago : who founded Los Millares and introduced copper smelting ?
and southern Iberia 4000 years ago : who founded La Bastida and later created El Argar cities and started bronze smelting ?

these are things we'll never know in detail, it takes just a small elite tribe with a specific knowledge and the right organizational skills to have a very big local impact

what yeasr are you talking about by celts in italy...........they only arrived around 500BC ......................which asks the question , if R-U152 is celtic, did it only arrive in Italy around 500BC
 
LOL, you're talking about haplogroup H right?

Not only mtDNA H...last time I checked, U4 and U5 only account for at most 20% of modern European lineages, and most of it can be found in the far north, both east and west, where you also find the highest WHG proportions.

So what you're saying is that only a full genome comparison of modern Tuscans and Etruscans + previous ancient peoples may be able to prove that or indeed anything concerning recent origins (also in connection to all populations including AJs)?

Yes, I think that only a full genome comparison of "Etruscans" and "Villanovans" say, or at least people from the closest prior central Italian culture which didn't cremate their dead would answer the question satisfactorily. I think the situation is the same for the Ashkenazim.

One final thing occurs to me in connection to the Tuscans. The published version of Lazardis et al does state that in Finestructure linked mode, Stuttgart clusters not with Sardinians but with North Italians and Tuscans. "This might indicate that despite the fact that Sardinians have more EEF-related ancestry than any other population, the EEF ancestry in some continental southern European populations may be more closely related to the early farmers of central Europe."


I have to think about that some more, and go over the tables again if I get a chance. I don’t quite understand it. I'm sure that they've written elsewhere that EEF is Stuttgart. Although in other places they've said that EEF is Stuttgart, Oetzi, and Gok. Maybe all they mean is that Sardinians are closest to Oetzi, but North Italians and Tuscans are closer to Stuttgart? Certainly in the PCA, Stuttgart is closer to Sardiniansthan even to Tuscans, although that's a projection, and PCAs in general and that one in particular are not the optimum way of looking at genetics, in my opinion.

However, it seems that this result might support the position of those who hold that any input from the Near East post Neolithic was not terribly significant. Even if J2 was involved, I think we’ve seen numerous examples where the ydna doesn’t affect the autosomal results all that much.
 
Yes, I think that only a full genome comparison of "Etruscans" and "Villanovans" say, or at least people from the closest prior central Italian culture which didn't cremate their dead would answer the question satisfactorily. I think the situation is the same for the Ashkenazim.

One final thing occurs to me in connection to the Tuscans. The published version of Lazardis et al does state that in Finestructure linked mode, Stuttgart clusters not with Sardinians but with North Italians and Tuscans. "This might indicate that despite the fact that Sardinians have more EEF-related ancestry than any other population, the EEF ancestry in some continental southern European populations may be more closely related to the early farmers of central Europe."


I have to think about that some more, and go over the tables again if I get a chance. I don’t quite understand it. I'm sure that they've written elsewhere that EEF is Stuttgart. Although in other places they've said that EEF is Stuttgart, Oetzi, and Gok. Maybe all they mean is that Sardinians are closest to Oetzi, but North Italians and Tuscans are closer to Stuttgart? Certainly in the PCA, Stuttgart is closer to Sardiniansthan even to Tuscans, although that's a projection, and PCAs in general and that one in particular are not the optimum way of looking at genetics, in my opinion.

However, it seems that this result might support the position of those who hold that any input from the Near East post Neolithic was not terribly significant. Even if J2 was involved, I think we’ve seen numerous examples where the ydna doesn’t affect the autosomal results all that much.

Alright, thanks for the answer and the detailed information. I'm pretty sure Semitic Duwa already stated that several pre exile Jews have been sampled and analyzed, so far... Apparently maternal lineage W was common among them (which makes sense since its origin is in west Asia and appears in the region of origin including south Asia and Europe).

P.S Apparently, at least today, the highest concentration of haplogroup W is in northern Pakistan, but it appears as a minority subclade in all the regions I mentioned.
Also, W* is found among the Svan people of the Caucasus (Georgia) at a frequency of 8.3%.
 
what yeasr are you talking about by celts in italy...........they only arrived around 500BC ......................which asks the question , if R-U152 is celtic, did it only arrive in Italy around 500BC

Italic tribes arrived in the Po Valley 1500 BC
1300 BC other tibes arrived from north of the Alps who cremated their deads - I call them bronze age Celts - and Italic people moved further south
Villanova II started 900 BC, first Etruscan cities being founded, this was an iron age culture
soon after Golaseccans arrived, first iron age celts, they were connected to Halstatt celts
400 BC celts invading Italy were La Tene iron age celts

definition of who are celts and who are not is not clear
 
IMO there must have been some immigration, but not massive, of some people with some better knowledge, probably of metal smelting than the local Italic and Celtic tribes at the time , they founded a new society of which they themself where the elite
the immigration was so small, it is hardly detectable in the DNA
you can ask the same question about copper age southern Iberia, some 5200 years ago : who founded Los Millares and introduced copper smelting ?
and southern Iberia 4000 years ago : who founded La Bastida and later created El Argar cities and started bronze smelting ?

these are things we'll never know in detail, it takes just a small elite tribe with a specific knowledge and the right organizational skills to have a very big local impact


I agree with that, and my hunch is that these tribes came from the eastern Mediterranean by sea, perhaps from the area of the Aegean, at least in terms of the Etruscans, as that would tie in rather nicely with evidence of an "Etruscan language" in that area.

This is all just speculation, of course. The answer will come with the ancient dna, if we ever get it.

definition of who are celts and who are not is not clear

Whatever is the case in other parts of Europe, I would agree that this is the case for Italy...when and where did the Italo-Celts separate into two different groups...where do the Ligures fit in...how do the Boi fit in? More importantly, for this discussion, how different were these people from one another genetically? Since we don't have ancient dna for each tribe, I think it's impossible to say. For my purposes, they're just northern migrations post Neolithic.
 
@Angela

last days I am looking toponyms around and inside Val Cammunnico,
do you have any genetic statistics? especially from Laas/Lasha?
 
@Angela

last days I am looking toponyms around and inside Val Cammunnico,
do you have any genetic statistics? especially from Laas/Lasha?

@Angela

last days I am looking toponyms around and inside Val Cammunnico,
do you have any genetic statistics? especially from Laas/Lasha?

Laas? This is the Val Camunica:
http://siti.voli.bs.it/gian/images/foto/cartina.jpg

I'm not familiar with any samples specifically from that area.

Given its geographic location, the results from Bergamo or Brescia would probably be closest.
Bergamo is used in all academic and hobbyist analyses as the Northern Italian reference sample, so autosomally I would look at those results.

Despite small sample sizes, Boattini et al is still, imo, the best analysis of uniparental markers in Italy, both because of their higher level of resolution and because they do their sampling based on surname origin, which is essential in a country like Italy.

See Boattini et al: http://www.plosone.org/article/info:doi/10.1371/journal.pone.0065441#pone.0065441.s001

Table 1 lists the frequencies of ydna haplogroups by region. Both Region III and IV would be applicable.

Boattini is extensively analyzed on this Board. You can find the threads through the search engin

All of Maciamo's maps of uniparental markers are useful as well.

The Capocasa et al study does look at isolates, but not specifically from the Val Camunica. Plus, due to budgetary constraints, no doubt, they tried to do the analyses based on old studies which were not very resolved, so the final analysis uses only 7 STRs. Very disappointing.

Ed. A study from the Estonian Institute did examine some Italian isolates. I'll try to dig it out later tonight to see if any of them were near this region.
 
I refreshed my recollection of the papers from the Estonian Institute and the isolates are in northeastern Italy and Val Borbera in basso Piemonte.

FWI, the creator of the MDLP calculator obviously has access to the raw date from the Estonian Institute papers. That's why so many northern Italians are now getting Piemonte as one of their matches on that run.

The samples from this institute are not, in my opinion, the best choice for autosomal analysis. Their sampling locations are always these very unrepresentative genetic isolates, chosen perhaps because they're studying recessive or otherwise genetic dieseases. In the case of the Val Borbera, in particular, I don't think it's very representative of general northwestern Italian populations. It has a very unique history. The same holds true for their Apulia sample.

I do understand the need for population geneticists not to sample in large cities where there has been a lot of mixing from disparate areas over the centuries. In such cases, a middle ground is, I think, to take a Boattini approach if the concern is not to mix southerners with northerners or Venetians with the Piemontese; use surnames. That's what Cavalli Sforza did for a lot of his studies.
 

This thread has been viewed 162544 times.

Back
Top