Southern Italian Ethnogenesis (My theory)

No, Sarno defined EEF as "Sardinian-like" and CHG as "Caucasus-like," well differentiated from "Near-Eastern".



For the record, many studies do not even use CHG, but you don't see anyone arguing that CHG should not be used. What you cannot accept is simply that you have too much Natufian/PPN even for the tastes of MLBA Central Anatolians.

SampleTUR_Barcin_NGEO_CHGLevant_PPNBIRN_Ganj_Dareh_NRUS_Samara_HG
TUR_Kaman-Kalehoyuk_MLBA54,817,213,814,20
Italian_Apulia55,91210,16,715,3

SampleTUR_Barcin_NGEO_CHGIRN_Ganj_Dareh_NRUS_Samara_HGLevant_Natufian
TUR_Kaman-Kalehoyuk_MLBA62,316,51506,2
Italian_Apulia60,412,26,814,85,8

Look up post #26: according to the same paper anatolian_N had a very big chunk of the "near eastern" component and were not even mostly Sardinian-like but had also a very big chunk of the Caucasian-like component.
 
Sarno et al. 2017

vfbbrIc.png
 
Anatolian is near eastern do you struggle with geography?

I'm well aware that Kaman-Kalehöyük is Near Eastern, but what I am trying to make you understand from 3 pages of thread is that it is not the only source and our high percentage of Natufian/PPN compared to it is proof of that.

Also CHG is not that much different from Iran_N. But they are dramatically different from Natufian. Also I already told you G25 is meaningless to me especially because the papers already model Southern Italians (accurately).

Iran_N is as different to CHG as an Englishman is to a Syrian. Both are necessary. If you do not trust Global25 it is not my problem, I do not trust papers either, so as far as I am concerned the discussion can end here.

Sarno et al. 2021 with actual aDNA in modeling

So now Sardinians are modeled as 15% Iran_N? Thanks for the laugh. You know, these are supposed to be the ones who model us "accurately"...

C'mon, we've already settled that Sarno 2017 was flawed (since we ought to conclude that Anatolian_neolithic had yemeni-like gene flows as well as north Europeans in the BA), and Sarno 2021's caucasian/near eastern ancestral sources are whatever brought extra CHG in the bronze age, since it then goes on to say that southern Italians are particuarly close to ancient BA Minoans and Myceneans samples (and the leaks show also to ancient Greeks). How you infer from that that you need extra southern (Levantine/southern eastern Anatolian) ancestry is your problem to be honest.
The clade branching seems a bit odd to me, I'll be honest, but it included only very northern Italians (Bergamo) and no Greeks, so I wonder what other results you'd have seen if those were included.

If we take Campanians as a proxy for southern Italians, they are as distant to Mycenaeans as they are to Cypriots, and they are as distant to Minoans as they are to Catalans. "Being close" means everything and nothing. How do I infer that extra-Levantine is needed? From these models with Global25: look at the difference in terms of components between an Italian LBA population and today's local population. But now they're going to tell me that Global25 is amateurish, that Davidski is a Nazi, that it's all a conspiracy against us and the usual things just to avoid accepting reality.

SampleTUR_Barcin_NGEO_CHGIRN_Ganj_Dareh_NLevant_PPNBRUS_Samara_HGITA_VillabrunaMAR_Taforalt
Sicilian_East/West51,4108,31313,82,41,1
TUR_Kaman-Kalehoyuk_MLBA54,817,214,213,8000
ITA_Sicily_LBA836005,45,30,3
GRC_Minoan_Lassithi86,694,40000
 
the discussion can end here.

Thank god.

But before you go away, I would like to know what kind of education level you have where you think a blogger is more reputable than actual experts, the whole field at that.
 
I'm well aware that Kaman-Kalehöyük is Near Eastern, but what I am trying to make you understand from 3 pages of thread is that it is not the only source and our high percentage of Natufian/PPN compared to it is proof of that.
Iran_N is as different to CHG as an Englishman is to a Syrian. Both are necessary. If you do not trust Global25 it is not my problem, I do not trust papers either, so as far as I am concerned the discussion can end here.
So now Sardinians are modeled as 15% Iran_N? Thanks for the laugh. You know, these are supposed to be the ones who model us "accurately"...
If we take Campanians as a proxy for southern Italians, they are as distant to Mycenaeans as they are to Cypriots, and they are as distant to Minoans as they are to Catalans. "Being close" means everything and nothing. How do I infer that extra-Levantine is needed? From these models with Global25: look at the difference in terms of components between an Italian LBA population and today's local population. But now they're going to tell me that Global25 is amateurish, that Davidski is a Nazi, that it's all a conspiracy against us and the usual things just to avoid accepting reality.
SampleTUR_Barcin_NGEO_CHGIRN_Ganj_Dareh_NLevant_PPNBRUS_Samara_HGITA_VillabrunaMAR_Taforalt
Sicilian_East/West51,4108,31313,82,41,1
TUR_Kaman-Kalehoyuk_MLBA54,817,214,213,8000
ITA_Sicily_LBA836005,45,30,3
GRC_Minoan_Lassithi86,694,40000

I have gone through raveanne to check the samples, and they don't match those used on G25; Maybe some Na.... are taken from Parolo 2015, but it would be good if you could for once provide some evidence instead of relegating the onus to others everytime.

My problem with G25 is that there is no transparency regarding where the samples are taken from, and to be honest also how they are treated/converted to G25 format. Add to that the renowned fact that on Anthrogenica there were people that passed fake kits of south Italians, and you can see why the skepticism is warranted.
Furthermore, it has already been pointed out that even north Italians get not trivial amount of Levantine, which is not accounted for in any "explanation" I have seen for south Italy, if we are willing to call those historical BS "explanations".
So here I go over the other problem I see with such theories: they lack any historical data whatsoever, somehow mentions of "east med migrants" during the Empire, or during hellenistic times, or even resettlements of near easterners during the Byzantine period, were taken as explanations despite the fact there's nothing in the whole historical literature that grant these theories any weight, otherwise someone would have suspected important near easterner contribution to south Italy well before archeogenetics on the ground of archeology or historiography alone.

You modelled Campanians on Anthrogenica as 45% BA Syria, which ought to ring as an extremely implausible model. The only thing I can agree with is that G25 samples do need Levantine, even when the Latin outliers were used or even when Anatolia_BA was used, but it is also why they seem fake as hell to me, honestly they look like partially Jewish.
 
You cite Sarno and disagree with other things she says. Don't cherry pick. Be honest with yourself.

I quoted her only because you asked me for a paper to prove that 2+2 makes 4. And I did, but my opinions do not depend on what she says in the papers.

I have gone through raveanne to check the samples, and they don't match those used on G25; Maybe some Na.... are taken from Parolo 2015, but it would be good if you could for once provide some evidence instead of relegating the onus to others everytime.

My problem with G25 is that there is no transparency regarding where the samples are taken from, and to be honest also how they are treated/converted to G25 format. Add to that the renowned fact that on Anthrogenica there were people that passed fake kits of south Italians, and you can see why the skepticism is warranted.

It is not the users who manage the modern samples that are on Global25, at least not in the case of Italy, they are taken from the papers but I do not know which ones they are. There are inaccuracies regarding Italians, see for example Liguria which has only 1 sample which is not even representative, or Piedmont which has samples from the southeast of the region and is more representative of Liguria or even Emilia rather than Piedmont, or Swiss_Italian which has 3 samples including 2 with ancestry from other parts of Italy.

As for Campania, the samples starting with "CMP" are from Naples and were initially labelled as "Italian_Naples" on Global25. I tried averaging Dodecad K12b using the same samples that are on Global25 except the ones with CMP, and the average is almost identical to the existing one.

Distance to: Italian_Campania_G25
0.92589927 Italian_Campania
2.32210615 Italian_Sicily
2.39774153 Italian_Molise
3.01846227 Italian_Basilicata
3.09908284 Italian_Calabria
3.66595001 Italian_Abruzzo
4.09456065 Ashkenazi_Jew
4.65260566 Italian_Apulia
4.97939398 Moldovan_Jewish
5.56348941 Greek_Lemnos
5.90415231 Greek_Fournoi
6.30293003 Greek_Izmir
6.88548034 Greek_Foca
7.05937688 Greek_Crete
7.39172270 Italian_Jew
8.12766814 Greek_Athens
8.28052395 Greek_Central
8.50210647 Greek_Icaria
8.70597794 Greek_Kos
9.21142543 Italian_Marche
9.44439990 Greek_Rhodes
9.50928504 Italian_Lazio
9.87882847 Sephardic_Jew

Furthermore, it has already been pointed out that even north Italians get not trivial amount of Levantine, which is not accounted for in any "explanation" I have seen for south Italy, if we are willing to call those historical BS "explanations".
So here I go over the other problem I see with such theories: they lack any historical data whatsoever, somehow mentions of "east med migrants" during the Empire, or during hellenistic times, or even resettlements of near easterners during the Byzantine period, were taken as explanations despite the fact there's nothing in the whole historical literature that grant these theories any weight, otherwise someone would have suspected important near easterner contribution to south Italy well before archeogenetics on the ground of archeology or historiography alone.

Lombards do get some Iran_N, but the Middle Eastern component (Natufian, PPN) is very low.

Target: Italian_Lombardy
Distance: 0.0378% / 0.03783877
63.0 TUR_Boncuklu_N:ZHAG_BON004___BC_7950___Coverage_67.13%
20.6 UKR_Meso:I1763___BC_8131___Coverage_70.73%
8.9 GEO_CHG:KK1___BC_7728___Coverage_99.87%
4.2 IRN_Ganj_Dareh_N:I1954___BC_8212___Coverage_79.77%
2.8 Levant_PPNC:I1699___BC_6750___Coverage_25.43%
0.5 RUS_Sidelkino_HG:Sidelkino___BC_9371___Coverage_84.07%

Target: Italian_Lombardy
Distance: 0.0380% / 0.03802084
65.9 TUR_Boncuklu_N:ZHAG_BON004___BC_7950___Coverage_67.13%
20.2 UKR_Meso:I1763___BC_8131___Coverage_70.73%
8.2 GEO_CHG:KK1___BC_7728___Coverage_99.87%
5.2 IRN_Ganj_Dareh_N:I1954___BC_8212___Coverage_79.77%
0.5 RUS_Sidelkino_HG:Sidelkino___BC_9371___Coverage_84.07%

You modelled Campanians on Anthrogenica as 45% BA Syria, which ought to ring as an extremely implausible model. The only thing I can agree with is that G25 samples do need Levantine, even when the Latin outliers were used or even when Anatolia_BA was used, but it is also why they seem fake as hell to me, honestly they look like partially Jewish.

The purpose was not to give an exact percentage but to prove that Levant came out in the models. I did several models using only Bronze Age samples and got different percentages, not just 45% which is clearly exaggerated (I will likely have gotten that percentage by setting the populations to be chosen to 2 or 3). Doing models with such mixed recent populations is a mess, it's all bloody intertwined and one population "overshadows" the other, that's why I prefer to do the usual ones with ancestral components (EEF, CHG, Steppe, etc.).

Target: Italian_Campania
Distance: 0.0083% / 0.00832839 | R3P
40.7 TUR_Kaman-Kalehoyuk_MLBA:MA2200___BC_1850___Coverage_80.26%
39.1 CZE_LBA_Knoviz:I13788___BC_1050___Coverage_70.32%
20.2 Levant_Sidon_MBA:ERS1790731___BC_1800___Coverage_83.40%

Target: Italian_Campania
Distance: 0.0067% / 0.00670417 | R4P
37.1 ITA_Sicily_MBA:I4109___BC_1566___Coverage_68.35%
23.1 Levant_Sidon_MBA:ERS1790731___BC_1800___Coverage_83.40%
22.3 ARM_Lchashen_MBA:DA35___BC_1309___Coverage_67.26%
17.5 Bell_Beaker_NLD:I4076___BC_1813___Coverage_37.88%

Target: Italian_Campania
Distance: 0.0051% / 0.00508700 | R5P
25.4 TUR_Kaman-Kalehoyuk_MLBA:MA2203___BC_1850___Coverage_49.75%
25.0 Bell_Beaker_NLD:I4076___BC_1813___Coverage_37.88%
19.7 ITA_Sicily_MBA:I4109___BC_1566___Coverage_68.35%
19.0 TUR_Alalakh_MLBA:ALA029___BC_1793___Coverage_57.87%
10.9 Levant_Baqah_BA:I6569___BC_1350___Coverage_55.78%
 
Yeah, the same paper that models Southern Italians without the need of PPN Levant. Great work at violating yourself.

[FONT=Verdana, Arial, Tahoma, Calibri, Geneva, sans-serif]2.8% Levant in Lombardy is wrong, 20+% in Campania is just asinine[/FONT]
 
Er Monnezza: I have been following this discussion closely and have been sort of getting a read of what is going on before "allegorically cross the Rubicon and enter the fray so to speak" (the Discussion). In full disclosure, I think Jovialis K8 Italian Model is valid and is in-line with the recent research on modern Italians

Regarding the discounting of the academic papers, what then becomes the basis for forming opinions regarding Italian genetics? I see come concerns regarding Sarno et al 2017, point taken. The Sarno et al 2021 paper seems to me to be consistent with other papers in the last 3 years on Italian DNA done by other Italian academic research teams. Still, having criticisms of a paper is fine, but again I would like to hear what those criticisms are? For the record, I like Jovialis, Angela, Leopoldo, and others here tend to lean towards what the academic papers find. I can't speak for what goes on at other sites that perhaps you post at since I am not a member of those sites (Anthrogenica, Apricity). In full disclosure, I have lurked at the Apricity and I think you are a frequent poster over there. From my perspective, the few times I have ventured over there to read the comments, I find the discussion over there lacking as very few over there seem to cite DNA studies but rather seem to focus on Carelton Coon type Anthropological analyses and discussion. Perhaps I am overly generalizing, but that is what I got the few times I went over there.

A calculator result in my opinion, be it Dodecad12B, Euro G25, MDLP16, etc that significantly differs from the extant published literature is questionable to me (maybe samples used to run the models, ancestral source populations used are not appropriate proxies to model modern populations).

There have been numerous papers in the last 6 years that have analyzed Italian samples from most if not all of the 20 political regions. The Sazzini et al 2016 paper (Sarno is a co-author) "Complex interplay between neutral and adaptive evolution shaped differential genomic background and disease susceptibility along the Italian peninsula" analyzed 747 modern Italians from the North, Central, South/Sicily and Sardinia. Figure 2 provides a PCA plot that has been replicated in other papers with other samples, over and over again. Admixture results are presented in Figure 3. The Raveane et al 2019 paper "Population structure of modern-day Italians reveals patterns of ancient and archaic ancestries inSouthern Europe". I will note that this paper has a totally different research team of Italian academics as well as other European academicians as co-authors. 1,616 modern Italians (maybe some of the samples used in prior studies, but still > 747 from Sazzini et al 2016) were analyzed.Figure 2 in that paper, which is well known and cited here among Eupedia Italians and Italians in the diaspora whose ancestors and ancestry is 100% from one of the 20 regions of Modern Italy provides admixture results that are "consistent" with what is reported in Sazzini et al 2016.

None of these papers are hinting at the 10.9% to 23.1 Levant Bronze Age ancestry in modern Southern Italians from your G25 results. Conversely, what has been documented in the papers above and others like Antonio et al 2019 (ancient Rome) and Ferandes et al 2020 (ancient Sicily and Sardinia) that Iran_NEO and/or CHG type ancestry was appearing in Lazio and Sicily in the Bronze age and more recently with Raveane et al 2022 "Assessing temporal and geographic contacts across the Adriatic Sea through the analysis of genome-wide data from Southern Italy" which models all Italian regions in the paper (Sardinia, Lombardy, Tuscany, Sicily, Apulia, Calabria) and all the Greek samples with Iran_NEO.

So my friendly suggestion is this. Take all the PCA plots of all the recent Italian DNA studies (and their respective admixture models) and plot the PCA's and show the ancestral/source population models. I suggest all to try to come up with some best estimate to avoid any concerns one way or the other regarding this paper vs. that paper. Next, run the Dodecad 12B models, G25 Models, MDLP16 and some of the other Euro models (K13/15) and plot the results. Then, lets see which ones converge to the the published papers and which ones diverge from them. For the sake of argument, lets assume the G25 with the Levant component that you are showing (10-23.1% per post 69) is the most divergent relative to the extant Italian DNA studies. Could you then explain such divergence? Could you tie your model to historical events, archeological records, linguistics, etc.?


In my experience here at Eupedia, the Dodecad 12B models and PCA's at least for me, and pretty much every other person of Italian ancestry from Italia Sud (but I would say Centro and Nord Italians as well) is the one that provides the best distances and when plotting on PCA most closely converges with the extant Italian DNA studies.
 
Last edited:
I quoted her only because you asked me for a paper to prove that 2+2 makes 4. And I did, but my opinions do not depend on what she says in the papers.
It is not the users who manage the modern samples that are on Global25, at least not in the case of Italy, they are taken from the papers but I do not know which ones they are. There are inaccuracies regarding Italians, see for example Liguria which has only 1 sample which is not even representative, or Piedmont which has samples from the southeast of the region and is more representative of Liguria or even Emilia rather than Piedmont, or Swiss_Italian which has 3 samples including 2 with ancestry from other parts of Italy.
As for Campania, the samples starting with "CMP" are from Naples and were initially labelled as "Italian_Naples" on Global25. I tried averaging Dodecad K12b using the same samples that are on Global25 except the ones with CMP, and the average is almost identical to the existing one.
Distance to: Italian_Campania_G25
0.92589927 Italian_Campania
2.32210615 Italian_Sicily
2.39774153 Italian_Molise
3.01846227 Italian_Basilicata
3.09908284 Italian_Calabria
3.66595001 Italian_Abruzzo
4.09456065 Ashkenazi_Jew
4.65260566 Italian_Apulia
4.97939398 Moldovan_Jewish
5.56348941 Greek_Lemnos
5.90415231 Greek_Fournoi
6.30293003 Greek_Izmir
6.88548034 Greek_Foca
7.05937688 Greek_Crete
7.39172270 Italian_Jew
8.12766814 Greek_Athens
8.28052395 Greek_Central
8.50210647 Greek_Icaria
8.70597794 Greek_Kos
9.21142543 Italian_Marche
9.44439990 Greek_Rhodes
9.50928504 Italian_Lazio
9.87882847 Sephardic_Jew
Lombards do get some Iran_N, but the Middle Eastern component (Natufian, PPN) is very low.
Target: Italian_Lombardy
Distance: 0.0378% / 0.03783877
63.0 TUR_Boncuklu_N:ZHAG_BON004___BC_7950___Coverage_67.13%
20.6 UKR_Meso:I1763___BC_8131___Coverage_70.73%
8.9 GEO_CHG:KK1___BC_7728___Coverage_99.87%
4.2 IRN_Ganj_Dareh_N:I1954___BC_8212___Coverage_79.77%
2.8 Levant_PPNC:I1699___BC_6750___Coverage_25.43%
0.5 RUS_Sidelkino_HG:Sidelkino___BC_9371___Coverage_84.07%
Target: Italian_Lombardy
Distance: 0.0380% / 0.03802084
65.9 TUR_Boncuklu_N:ZHAG_BON004___BC_7950___Coverage_67.13%
20.2 UKR_Meso:I1763___BC_8131___Coverage_70.73%
8.2 GEO_CHG:KK1___BC_7728___Coverage_99.87%
5.2 IRN_Ganj_Dareh_N:I1954___BC_8212___Coverage_79.77%
0.5 RUS_Sidelkino_HG:Sidelkino___BC_9371___Coverage_84.07%
The purpose was not to give an exact percentage but to prove that Levant came out in the models. I did several models using only Bronze Age samples and got different percentages, not just 45% which is clearly exaggerated (I will likely have gotten that percentage by setting the populations to be chosen to 2 or 3). Doing models with such mixed recent populations is a mess, it's all bloody intertwined and one population "overshadows" the other, that's why I prefer to do the usual ones with ancestral components (EEF, CHG, Steppe, etc.).
Target: Italian_Campania
Distance: 0.0083% / 0.00832839 | R3P
40.7 TUR_Kaman-Kalehoyuk_MLBA:MA2200___BC_1850___Coverage_80.26%
39.1 CZE_LBA_Knoviz:I13788___BC_1050___Coverage_70.32%
20.2 Levant_Sidon_MBA:ERS1790731___BC_1800___Coverage_83.40%
Target: Italian_Campania
Distance: 0.0067% / 0.00670417 | R4P
37.1 ITA_Sicily_MBA:I4109___BC_1566___Coverage_68.35%
23.1 Levant_Sidon_MBA:ERS1790731___BC_1800___Coverage_83.40%
22.3 ARM_Lchashen_MBA:DA35___BC_1309___Coverage_67.26%
17.5 Bell_Beaker_NLD:I4076___BC_1813___Coverage_37.88%
Target: Italian_Campania
Distance: 0.0051% / 0.00508700 | R5P
25.4 TUR_Kaman-Kalehoyuk_MLBA:MA2203___BC_1850___Coverage_49.75%
25.0 Bell_Beaker_NLD:I4076___BC_1813___Coverage_37.88%
19.7 ITA_Sicily_MBA:I4109___BC_1566___Coverage_68.35%
19.0 TUR_Alalakh_MLBA:ALA029___BC_1793___Coverage_57.87%
10.9 Levant_Baqah_BA:I6569___BC_1350___Coverage_55.78%


I had seen this kind of discussions on anthrogenica: in the end your only "evidence" is G25, and when asked about what warrants G25 over the scientific literature you only blabber about how "no paper is perfect, how the experts get it wrong sometimes", and the like. Which I agree with, to an extent, but to spot some mistakes or unconvincing conclusions here and there is utterly different from dismissing wholesomely the entirety of the scientific literature as you do.

It wouldn't be fair to dismiss G25 either without good reasons, but I have already spellt them out and I shall again to lay them down:

1) It isn't clear where the samples are from, and it was one reason why Davidsky himself took down the official dataset to go over it 1 or 2 years ago, but the Italians' averages haven't changed much, and there are still some issues, as you yourself have conceded (although for north Italians); you've claimed the Campanian samples are from official studies but have failed to specify them, it doesn't mean they necessarily aren't and you may simply not be able to find them but ask somebody then, or Davidsky himself. Some tried to ask him but he didn't answer.
2) On top of that, there have been numerous individuals that passed around fake kits of alleged south Italians.
3) There are individuals who pretend to be 100% south Italian but also admit that they have other ancestral roots but write it off as "not important", and indeed in the anglosphere there are a lot of mixed individuals who just claim Italian roots(I recall the discussions here with Principe azzurro, who honestly doesn't look Italian at all and brought up his own results to argue south Italians have Levantine because he does despite he is partially Jewish according to his own admission)
4)Davidsky himself has an odd grudge against south Italians, in 2016 he claimed that there was no evidence of CHG ancestry in Italians and Greece but there was a "clear cline from north Italy to the Levant", and he went as far to say that south Italians ought to be kicked out of Europe, so sorry, when the trustworthiness not only of the selection of the samples but also of their treatment/conversion to G25 format depends on such an individual or close collaborators, one is more than justified in being extremely skeptical.

The most important thing to keep in mind is that these four points are important BECAUSE G25 RESULTS ARE EXTREMELY DIVERGENT COMPARED TO ANYTHING ELSE for south Italians, so it doesn't cut when you just present G25 results as if they were gospel and dismiss the entirety of the genetic literature. When some papers are brought up to back up G25, either they are misread or they've flaws which bring to evidently false conclusions (as BA north Europeans having direct levantine, yemeni-like ancestry), and which are addressed.

If Campanians or south Italians in general had 20-25% Levantine admixture, why no genetist has ever found it out? I said otherwise that I could see why some single digit of Levantine admixture could not have been detected, but we are talking about a quarter of ancestry stemming from the Levant (along with another 20-25 from Anatolia and almost zero Greek), which happens also to be "balanced out" by 15-20% northern European-like admixture: I'd rather not going over why such a scenario is utterly impossible given historical evidence, and also the uniparental data does not support it, and not even the social dynamics present in the classical age and antiquity.
 
Yeah, the same paper that models Southern Italians without the need of PPN Levant. Great work at violating yourself.

Even without the "need" for CHG, but this does not imply that they do not have it.

2.8% Levant in Lombardy is wrong, 20+% in Campania is just asinine

It is asinine to overlook that the first percentage was obtained with samples from the Mesolithic and Early Neolithic, whereas the second with samples from the 2nd millennium BC.

I had seen this kind of discussions on anthrogenica: in the end your only "evidence" is G25, and when asked about what warrants G25 over the scientific literature you only blabber about how "no paper is perfect, how the experts get it wrong sometimes", and the like. Which I agree with, to an extent, but to spot some mistakes or unconvincing conclusions here and there is utterly different from dismissing wholesomely the entirety of the scientific literature as you do.

Different geneticists can each model a population differently, and scientific literature is a big word; the models they make are no more valuable than the ones I can make with Global25, or the ones Jovialis makes with Dodecad K12b, or the ones a more experienced user might make with qpAdm.

It wouldn't be fair to dismiss G25 either without good reasons, but I have already spellt them out and I shall again to lay them down:

1) It isn't clear where the samples are from, and it was one reason why Davidsky himself took down the official dataset to go over it 1 or 2 years ago, but the Italians' averages haven't changed much, and there are still some issues, as you yourself have conceded (although for north Italians); you've claimed the Campanian samples are from official studies but have failed to specify them, it doesn't mean they necessarily aren't and you may simply not be able to find them but ask somebody then, or Davidsky himself. Some tried to ask him but he didn't answer.

I honestly do not understand the point of this controversy. Do you think the results would change significantly if the samples were provided directly by you and me instead? They would not change much, and the trend would be the same anyway. Also you were saying that you thought they were fakes and some even partially Jewish, which is not true, you can see on a PCA that they are quite distinct.

Link on imgur: dhXVIJH.png

2) On top of that, there have been numerous individuals that passed around fake kits of alleged south Italians.

There may have been cases, but numerous individuals is an exaggeration. Moreover, this has absolutely nothing to do with Global25.

3) There are individuals who pretend to be 100% south Italian but also admit that they have other ancestral roots but write it off as "not important", and indeed in the anglosphere there are a lot of mixed individuals who just claim Italian roots(I recall the discussions here with Principe azzurro, who honestly doesn't look Italian at all and brought up his own results to argue south Italians have Levantine because he does despite he is partially Jewish according to his own admission)

I remember Principe, his results were divergent. He had more EEF and Taforalt than average but he even had less of the Middle Eastern component than average, which might suggest that he has no Jewish ancestry at all. But that's his personal business and I don't want to look into it.

4)Davidsky himself has an odd grudge against south Italians, in 2016 he claimed that there was no evidence of CHG ancestry in Italians and Greece but there was a "clear cline from north Italy to the Levant", and he went as far to say that south Italians ought to be kicked out of Europe, so sorry, when the trustworthiness not only of the selection of the samples but also of their treatment/conversion to G25 format depends on such an individual or close collaborators, one is more than justified in being extremely skeptical.

A lot has changed from 2016 to now. Even more has changed from 2010 or so when Davidski allegedly made those racist comments (which I have no proof of anyway). We are not that important to him, he cares more about the Steppe or at any rate Eastern Europe which is his home area, not about Italians.

The most important thing to keep in mind is that these four points are important BECAUSE G25 RESULTS ARE EXTREMELY DIVERGENT COMPARED TO ANYTHING ELSE for south Italians, so it doesn't cut when you just present G25 results as if they were gospel and dismiss the entirety of the genetic literature. When some papers are brought up to back up G25, either they are misread or they've flaws which bring to evidently false conclusions (as BA north Europeans having direct levantine, yemeni-like ancestry), and which are addressed.

If Campanians or south Italians in general had 20-25% Levantine admixture, why no genetist has ever found it out? I said otherwise that I could see why some single digit of Levantine admixture could not have been detected, but we are talking about a quarter of ancestry stemming from the Levant (along with another 20-25 from Anatolia and almost zero Greek), which happens also to be "balanced out" by 15-20% northern European-like admixture: I'd rather not going over why such a scenario is utterly impossible given historical evidence, and also the uniparental data does not support it, and not even the social dynamics present in the classical age and antiquity.

Geneticists don't usually make models with Bronze Age samples, and I don't like making them either. The reason I specified before, one component eats the other and it's all too intertwined.
 
Ah, yes, it's like the good old dna forums days. "Er Monezza", like Sickelliot of yore, appropriates the insights I provide for his own, i.e. the Piedmont sample more likely being mountain Liguria. If he were writing a paper, that would be plagiarism.

Odd he forgets to mention that the Corsican samples might have some French in them perhaps. I mean, in for a penny, in for a pound, as they say.

The reason that my "predictions" have been infinitely better than those of Davidski is that, most important, I am honest to a fault, and second, I stick with the data, genetic and archaeological, and third, I apply common sense and the most rigorous logical reasoning I can muster.

The data presented in the academic papers do not support these extremely high numbers for actual "Levantine" ancestry, although it wouldn't bother me in the slightest were they accurate. I see no reason to believe that Italy was different from every other place in the ancient world in that every single body found was of a permanent immigrant, and "the tail into the Levant" disappeared, as makes sense from a historical perspective because as time passed the seat of commerce and power left Rome as it declined, and moved to Constantinople.

That said, I find the abstract of the upcoming paper from the Reich Lab extremely interesting, as are the results of the "Roman Era" Greek. I do not find it at all impossible that as time passed more Levantine ancestry made its way into the people of Anatolia, the Greek Islands, Greece itself, and on to Italy. We'll see what the details are when we get a look at the whole paper. However, my respect for the Reich Lab notwithstanding, if the argument were for some reason to turn out to be that all of the remains from port cities are to be counted as potential ancestors of modern Italians, then they're wrong.

On another level, this is to some extent a tempest in a teapot. As the charts provided by Jovialis show, Anatolian farmers can be modeled with both CHG/Iran Neo and "Levantine" ancestry. It went to Europe with them, and so has spread to all Europeans, south to north, east to west, since the Neolithic.

One might ask, what's the big deal then? The big deal is the virulent anti-Semitism of people like Davidski and his Slavophile Nazi followers on Eurogenes. His hatred of the Jews spreads to all Levantines, and he is desperate to wall off "his" part of Europe from the contagion, and limit it to Italy, perhaps Greece and other parts of Southern Europe.

He is aided, bizarrely enough, by a Jewish (or part Jewish) poster, or so he claims, on anthrogenica who wants Italians to be part of the "tribe", or "wanna be" Jews, (a group not uncommon in the U.S., where Jews have a very high status for understandable reasons) followers of the sadly misinformed "Calabrian rabbi", or a sad young man who hates Jews because of his obsession with the Palestinian cause, and has targeted them at university, and for family reasons I will not describe, because even he is deserving of some consideration, even though he extends none to others, wants to find Jewish ancestry in Southern Italians.

It's a pretty pathetic bunch. They'd make absolutely terrible "expert" witnesses whose conclusions would not and should not be given any weight. Even papers with some flaws are infinitely more to be trusted, especially given that there is absolutely no transparency, neither as to the data nor as to the methodology. The credulity of some people in the amateur population genetics community is astounding. Everyone should demand that the source of the samples be produced, as well as an explanation of the methodology so that everyone can see whether the results can be replicated. If that isn't done the results should be ignored.
 
Beyond the greater or lesser proportion of Levantine origin that southern Italians have... I will never understand why that can be a cause for shame. Levant, Mesopotamia, the Nile Valley...are at the origin of western civilization, while at the same time, other peoples who would acquire prestige millennia later, lived in a primitive situation...
 
Beyond the greater or lesser proportion of Levantine origin that southern Italians have... I will never understand why that can be a cause for shame. Levant, Mesopotamia, the Nile Valley...are at the origin of western civilization, while at the same time, other peoples who would acquire prestige millennia later, lived in a primitive situation...

Well said, italouruguayan

Surely there is at least some, the model shows. Whether it is trace amounts as my model shows, as well as academic studies show, or it is 20% as the crypto-anti-Semites show.
 
To the imbecile I just banned, this is what an Natufian looks like? Probably an infamous example, but a good example of the phenotypic diversity you see in Puglia nonetheless. Raffaele Sollecito, from the region of Bari:

rr8dKkzl.png
 
Different geneticists can each model a population differently, and scientific literature is a big word; the models they make are no more valuable than the ones I can make with Global25, or the ones Jovialis makes with Dodecad K12b, or the ones a more experienced user might make with qpAdm.

This means nothing since most genetists still make roughly similar models to say the least, that is I do not see the sort of "relativism" you seem to imply, and honestly quite consistently south Italians get modelled as similar to ancient SE european samples as a combination of CHG/Iran_N+EEF+WSH (yamna) in similar proportions; furthermore I deduce your lack of understanding of how archeogenetics works: I agree that potentially the models they or others can make are no more valuable than what we can do, but archeogenetics papers have much more than ADMIXTURE models, and trend to have many tests whose results ought to converge in order to draw plausible conclusions.
Furthermore, the last paper on southern Italians did try to check whether modern south Italians have affinities to middle Easterners or north Africans due to admixture and the results have been negative.



I honestly do not understand the point of this controversy. Do you think the results would change significantly if the samples were provided directly by you and me instead? They would not change much, and the trend would be the same anyway. Also you were saying that you thought they were fakes and some even partially Jewish, which is not true, you can see on a PCA that they are quite distinct.

Link on imgur: dhXVIJH.png



There may have been cases, but numerous individuals is an exaggeration. Moreover, this has absolutely nothing to do with Global25.



I remember Principe, his results were divergent. He had more EEF and Taforalt than average but he even had less of the Middle Eastern component than average, which might suggest that he has no Jewish ancestry at all. But that's his personal business and I don't want to look into it.



A lot has changed from 2016 to now. Even more has changed from 2010 or so when Davidski allegedly made those racist comments (which I have no proof of anyway). We are not that important to him, he cares more about the Steppe or at any rate Eastern Europe which is his home area, not about Italians.

In short you are saying that it doesn't really matter whether the samples or their conversion to the G25 format (that is they haven't been tinkered with) are trustworthy or not, which is nonsense. In normal circumstances, I'd agree that it would be conspiratorial to believe a priori that any sample not provided by researchers must be fake, but I've listed the reasons why it is warranted in this case, since 1) there are many potential sources for such fake kits, sources in contact with Davidsky/the amateur genetist world, and 2) the curator of the selection and conversion of the samples has manifested racist remarks towards south Italians. Now, I am not the type that cares more about the ethical standards of the person that utters a proposition than the proposition's truth, but since the only warrant of the proposition's truth (that the samples and their conversion are legit) depends on the trustworthiness of the person, my remarks aren't an ad hominem but legitimate concerns. It isn't enough to write them off by saying that the aforementioned individuals have nothing to do with G25 (they might have been removed, but Sizzi's samples were incorporated in G25), and that Davidsky has changed: once they've been caught lying, it is fair not to trust them.

Geneticists don't usually make models with Bronze Age samples, and I don't like making them either. The reason I specified before, one component eats the other and it's all too intertwined.

Another answer that does not answer anything: you've said that BA/IA "east meds" (SE europeans) had not the levantine component modern ones have according to you, so it must have come later, thus after the IA and during the hellenistic period or the imperial age, so as a proxy Levant_BA does represent the amount of gene flow that happened during that era; the actual component that isn't very "historical correct" is the Anatolia_BA one, since Anatolia_IA seems to have been only half Anatolia_BA and half Balkan_IA, so you'd have 50% Anatolian IA and 25% Levantine_BA/IA admixture in modern southern Italians followed by some 10-15% northern European admixture, and this is the reason why I've said that the G25 south Italian samples behave like ashkenazis: they do not only require a more southern component, but also a northern one, so they would be a tripartite mixture of ancient southern Europeans, northern Europeans and Levantines/Near Easterners. The history of the diaspora of Ashkenazi jews explain quite easily how they got their genetic profile, but when I ask what historical events caused southern Italians to happen to have a similar genetic profile, the only answers I get are pseudohistorical ramblings or cricket noise.
 

This thread has been viewed 68288 times.

Back
Top