Dodecad Caucasian DNA likely related to Danubian Farmers

[BillT]

The distribution of Caucasian DNA in Europe is frankly pretty unbelievable and perplexing. There is 10% in Belgians and only 4% in Dutch. It is damn near absent in Scandinavians, and over 10% in Polish people. And yet it is 3% in Russians. It is 20% in Sardinians, close to 10% in much of Spain AND 0% IN THE BASQUE COUNTRY. There is a wild amount of variability to this admixture and there is no straightforward cline, or vector we can point to for this admixture.
I would however argue that there is a likely vector for this admixture: the Early European Farmers who took the Danubian route into Europe. We know that one group took the Med, and one group went inland.
It's either that, or it maps to a particular type of Paleolithic/Mesolithic Hunter Gatherer

[Maciamo]

I understand your confusion. The reason is that this admixture was spread in several steps, as I explain here.

1) The Caucasian admixture was first brought to Europe by Neolithic farmers, but got progressively diluted (37% in Neolithic Greece, 31% in the Balkans, 28% in Italy, 12% in Scotland, 7% in Globular Amphora Poland).

2) A new wave of Near Eastern Chalcolithic farmers and the advance of the Kura-Araxes culture from the South Caucasus spread Caucasian admixture across the Near East and Southeast Europe. The descendants of Kura-Araxes spread to Greece during the Minoan period. Minoans had 38% to 45% of Caucasian admixture, higher than any Neolithic culture in Europe or even Anatolia (Boncuklu Aceramic Neolithic had only 30 to 33% of Caucasian).

3) Ancient Greeks later colonised Italy and Iron Age samples of Greeks from Italy show as much as 40% of Caucasian admixture (against 13% to 21% for Iron Age Latins).

4) By the time of Emperor Augustus, Italians of Italic/Latin, Etruscan and Greek origin had been intermarrying for centuries and the average "Roman" from the Italian peninsula had levels of Caucasian admixture around 20% to 30%. Migrations from Italy during the Roman period spread this admixture around Europe, notably in regions with lots of Roman colonies like Southeast France, southern and eastern Iberia and Belgium + Rhineland (lots of legions stationed in the region and lots of Roman villas in Wallonia).


The Basque lack the Caucasian admixture because they were hardly Romanised at all - so little in fact that they managed to hold to their language, unlike anyone else in western Europe.

[BillT]

I appreciate all the work you are doing, but you have to stop speaking like an authority on these matters when I don't believe you are. Academic study, after academic study, can find close to 0 genetic impacts of the Romans throughout many of the places they colonized. Not to mention the Romans could not have had any impact east of the Rhine in Poland etc., Assuming this data isn't pure noise, then what is implied here is two waves of Early European Farmers. There is already strong evidence for two separate routes of colonization.

[Jovialis]

It should also be noted that the steppe-related CHG/IN in Europe is subsumed in other components in this calculator. Steppe ancestry is 40% CHG - 60% EHG, which technically makes Norwegians the most Caucasian-like in Northern Europe.



Modified by me:

[Maciamo]

I appreciate all the work you are doing, but you have to stop speaking like an authority on these matters when I don't believe you are. Academic study, after academic study, can find close to 0 genetic impacts of the Romans throughout many of the places they colonized. Not to mention the Romans could not have had any impact east of the Rhine in Poland etc., Assuming this data isn't pure noise, then what is implied here is two waves of Early European Farmers. There is already strong evidence for two separate routes of colonization.

I never said that the Romans spread this admixture to Poland or Eastern Europe. But migrations within Europe didn't stop after the fall of the Roman Empire. Here are a few examples (among many):

• In the Middle Ages and Renaissance, West Germans moved eastward to Brandenburg, Prussia, Silesia, Romania, and even the Lower Volga. Many of these Germans had some Roman ancestry from the numerous legions stationed along the Rhine for 400 years and big Roman cities like Cologne or Trier.
• Germans and other Europeans (French Huguenots, Walloons) moved to Scandinavia from the 16th century onwards.
• European royals and aristocrats have intermarried with one another across the continent since the Middle Ages. Since they had more children than average, many of their junior lines and illegitimate children eventually contributed to the gene pool of the general population in each country, helping spread more evenly various admixtures.
• Gene diffusion take place simply by ordinary people finding a spouse from a few villages away, or moving to a different city or region. This diffusion process is slow, but inevitable and its effect is cumulative over centuries and millennia.

Just look at the Hinxton Anglo-Saxon samples. They had 0.5% of Caucasian. Iron Age Britons had about 1%. Yet modern English have about 3% of Caucasian. Obviously it must have come after the Anglo-Saxon migrations, and not from the Vikings either.

[Maciamo]

It should also be noted that the steppe-related CHG/IN in Europe is subsumed in other components in this calculator. Steppe ancestry is 40% CHG - 60% EHG, which technically makes Norwegians the most Caucasian-like in Northern Europe.



Modified by me:

The Caucasian admixture from Dodecad K12b is not the same as CHG. For example the Yamna samples from Ukraine had only 2 to 7% of Caucasian admixture, but 59% to 64% of North European admixture (found in both WHG and EHG).

[Jovialis]

The Caucasian admixture from Dodecad K12b is not the same as CHG. For example the Yamna samples from Ukraine had only 2 to 7% of Caucasian admixture, but 59% to 64% of North European admixture (found in both WHG and EHG).

I wish there was a calculator that had academic components based on ancient source populations, rather than having to rely on outmoded constructs based on modern populations.

[Jovialis]

Modern Caucasian populations can be modeled with a fair amount of Anatolian_N, and Steppe ancestry within and of itself:

[Jovialis]

I never said that the Romans spread this admixture to Poland or Eastern Europe. But migrations within Europe didn't stop after the fall of the Roman Empire. Here are a few examples (among many):

• In the Middle Ages and Renaissance, West Germans moved eastward to Brandenburg, Prussia, Silesia, Romania, and even the Lower Volga. Many of these Germans had some Roman ancestry from the numerous legions stationed along the Rhine for 400 years and big Roman cities like Cologne or Trier.
• Germans and other Europeans (French Huguenots, Walloons) moved to Scandinavia from the 16th century onwards.
• European royals and aristocrats have intermarried with one another across the continent since the Middle Ages. Since they had more children than average, many of their junior lines and illegitimate children eventually contributed to the gene pool of the general population in each country, helping spread more evenly various admixtures.
• Gene diffusion take place simply by ordinary people finding a spouse from a few villages away, or moving to a different city or region. This diffusion process is slow, but inevitable and its effect is cumulative over centuries and millennia.

Just look at the Hinxton Anglo-Saxon samples. They had 0.5% of Caucasian. Iron Age Britons had about 1%. Yet modern English have about 3% of Caucasian. Obviously it must have come after the Anglo-Saxon migrations, and not from the Vikings either.

You see a noticeable rise of Caucasian and Anatolian admixture from British and Nordic IA to their modern populations:

[Jovialis]

I understand your confusion. The reason is that this admixture was spread in several steps, as I explain here.

1) The Caucasian admixture was first brought to Europe by Neolithic farmers, but got progressively diluted (37% in Neolithic Greece, 31% in the Balkans, 28% in Italy, 12% in Scotland, 7% in Globular Amphora Poland).

2) A new wave of Near Eastern Chalcolithic farmers and the advance of the Kura-Araxes culture from the South Caucasus spread Caucasian admixture across the Near East and Southeast Europe. The descendants of Kura-Araxes spread to Greece during the Minoan period. Minoans had 38% to 45% of Caucasian admixture, higher than any Neolithic culture in Europe or even Anatolia (Boncuklu Aceramic Neolithic had only 30 to 33% of Caucasian).

3) Ancient Greeks later colonised Italy and Iron Age samples of Greeks from Italy show as much as 40% of Caucasian admixture (against 13% to 21% for Iron Age Latins).

4) By the time of Emperor Augustus, Italians of Italic/Latin, Etruscan and Greek origin had been intermarrying for centuries and the average "Roman" from the Italian peninsula had levels of Caucasian admixture around 20% to 30%. Migrations from Italy during the Roman period spread this admixture around Europe, notably in regions with lots of Roman colonies like Southeast France, southern and eastern Iberia and Belgium + Rhineland (lots of legions stationed in the region and lots of Roman villas in Wallonia).


The Basque lack the Caucasian admixture because they were hardly Romanised at all - so little in fact that they managed to hold to their language, unlike anyone else in western Europe.

I generally believe this is the case, in regards to Ancient Greeks mixing with Italics.

But also take a look at R437.

[Jovialis]

The Caucasian admixture from Dodecad K12b is not the same as CHG. For example the Yamna samples from Ukraine had only 2 to 7% of Caucasian admixture, but 59% to 64% of North European admixture (found in both WHG and EHG).

Another issue with using modern populations is that it is not reliable in determining percentages for these populations. Because the calculator measures admixture within the frame of modern population; WHG and EHG fall outside of the modern population continuum. Thus, I find Dodecad is better for looking at more recent samples, who are already mixed. Frankly, the components should be looked at with extreme caution, and not be taken seriously. EHG, WHG, Ukranian_N, etc are wrongly clustered very close to one another in PCAs, forming a funnel.



They should look more like this, if there were proper ancestral components for determining admixture rates:

[Jovialis]

^^FYI this is actually with G13, but that is the more advanced version of the dodecad calculator. Based on what I have seen, Dodecad 12b is even worse for these HGs.

[BillT]

Maciamo, the reality is that many studies show the great German migration and other events like the Roman Empire to have had little impact on the DNA of other peoples. Many studies show that local genetic structure has been in place for a very long time at least until recently.
And finally, there is a massive, MASSIVE amount of population structure related to that caucasian admixture. It is not on any straight forward easy to analyze cline. It is fairly abundant in Sardinia and completely absent in the Basque country, two places that we imagine to be EEF refuges. Sorry but there is no straightforward way to interpret this result. Honestly, your genetic diffusion hypothesis implies this admixture would have diffused into Northern Germany and Scandinavia. It did not. And also there are so few ancient samples and their DNA is low coverage/low quality so we have to be careful what conclusions we draw comparing them to an admixture derived from a modern pop.

[BillT]

You bring up some really great points Jovialis

[Jovialis]

We do not need Dodecad to understand how much non-steppe related CHG/Iran_N is in places like Italy for example, the more recent study, Sarno et al. 2021, on the region shows the maximum was only 29% at the southern most tip of the peninsula:



The Dodecad component Caucasian is a modern population, already mixed with overlapping source populations as I have demonstrated up thread. It is an outmoded, and confounding construct, that must not be taken seriously.

Finally, to better characterize the ancestral composition of Southern Italian populations, we inferred their mixture proportions with respect to a four-population model of admixture including all the above-mentioned WHG, Neolithic, CHG/Iran_N and Steppe-related main sources, using qpAdm. All Italian populations were successfully modeled as characterized by a relatively high amount of Anatolian Neolithic ancestry, with the major contribution observed in Sardinians (Fig. 4, Suppl. Table S8). The remaining ancestries were assigned to a lower WHG contribution and to differential influences of Steppe_EMBA and CHG/Iran_N in the profiles of Northern and Southern Italians, respectively (Fig. 4, Suppl. Table S8). In fact, while Steppe ancestry is greater in North Italy (~ 27%), the Iran_N/CHG-related source is more present in South Italy with the highest values (~ 29%) observed in the populations from the Aspromonte area.

https://www.nature.com/articles/s41598-021-82591-9

[Jovialis]

Sarno et al. 2021 Supplement 2

Population	Ancestral Components
	Iran_N	WHG	Steppe_EMBA	Anatolia_N
Sardinian	0.145	0.122	0.075	0.658
North_Italy	0.157	0.079	0.272	0.492
Benevento	0.240	0.039	0.168	0.553
Castrovillari	0.251	0.037	0.161	0.551
Catanzaro	0.278	0.025	0.131	0.566
Aspromonte	0.292	0.024	0.113	0.571

Non-Steppe Related CHG/Iran_N Percentages:

• Sardinia: 14.5%
• North Italy: 15.7%
• Benevento: 24%
• Castrovillari: 25.1%
• Catanzaro: 27.8%
• Aspromonte: 29.2%

[Jovialis]

Admixture rates according to CC Wang et al. 2019: