search this blog

Sunday, November 30, 2014

Short clip: The making of modern Europe

Simple but, I think, very cool animation: ten ancient genomes analyzed with the Eurogenes K15. More elaborate clips are on the way.

And this is basically the same thing, but restricted to samples from Hungary.


jackson_montgomery_devoni said...

Awesome graphic animation David. This actually presents quite a nice visual as to the ancient genetic fluctuations that occurred in Europe over time and their relations to certain present day populations. I really did not expect Motala12 to score so high in the North_Sea cluster. I thought that this sample would have scored higher Baltic but obviously his Baltic and East_Euro scores are also quite high. I think it is very obvious now that the North_Sea cluster, the Baltic cluster and the East_Euro cluster are composed by far mostly of Mesolithic hunter-gatherer origins when seen in modern Europeans. The Atlantic cluster probably also has a lot of Mesolithic alleles in it.

Tesmos said...


Nice video. Will there be any new calculator soon?

Krefter said...

I liked it. Because it doesn't put its faith into EEF-WHG-ANE. It simply shows how K15 results changed over time in Europe, and how modern ones are a mix of ancient ones.

I happen to have been doing something similar yesterday. I gave a simple PP presentation for my family to show how ancient DNA has identified our pre-historic ancestors.

I used ancient K 15 scores and our K15 scores in the end to show how we are a mix of Mesolithic Euros and Middle eastern-type people, but it was way to complicated for them.

I suggest doing something more simple and eye-catching in the future, if you want viewers.

Fancy looking maps, K15 scores in a spreadsheet, short and big-lettered messages, pictures of ancient people like La Brana-1, etc. And also sequels which will keep people watching. That is if you want viewers.

Mateus Smith said...


This is off-topic, but I was reading about the impact of the 30 years war on Central Europe, and this part got me wondering:

"Germany lay desolate. The population had fallen from 21 million to perhaps 13 million... The manly exploits of Spanish, Swedish, Italian, Croat, Flemish, and French soldiers had changed the racial composition of the people."

In your opinion, did these soldiers really leave a considerable genetic impact on the German people? About how much, would you say?

Thank you.

BTW, nice video.

Krefter said...


I'm 25% or more German and Swiss and 23andme confirmed this, but based on admixtures on GEDmatch my German-Swiss fits best as N Italian with NE European influence.

I know Swiss can be similar to N Italy, but I'm only 6.25% Swiss.

BTW, one of my Swiss ancestors invented the shoulder padding people wear in suits today. So, whenever you wear a suit you can thank him.

Krefter said...

Fanty said...


I did not read about the impact of raping in the 30 years war.

What I did read is, that the 30 years war killed about 50% of the German population. And that the population gaps had been filled by massive migration into Germany, from elsewhere.

Pre-genetic-testing was typically claimed: "Balkan" or the region of what is now Hungary would have been a massive source of migration to Germany after the 30 years war.

My opinion is: Far too few I2a (that "Slavic" (Balkan+Ukraine) type of I) in Germany for the Balkan beeing a huge source.

On the other hand, older Oracles that used Germany as only 1 population suggested for 2 Population mix: Germany = Scandinavia + Balkan mixed

But meanwhile we have 3 German pops and that suggest:

Westgermans: Scandinavian+French mixed
Eastgermans: Dutch+Westslavic mixed

The mediteranean element of French and the eastern European element from Westslavs together made it look like Balkan admixture.

Davidski said...


I'm working on a new test now, using the new EEF genomes, like NE1. But I'm not happy with the results so far, so it might take a while.


I've never seen any evidence of substantial foreign gene flow into Germany since the Migration Period.

I'd say that internal migrations, especially across the north of Germany, have had the biggest effect on German genetic structure, by homogenizing much of the country.

Chad Rohlfsen said...

Nice work David! Any chance you'll make one with just the Hungarian genomes?

Krefter said...

"Nice work David! Any chance you'll make one with just the Hungarian genomes?"

He probably just did this for the fun of it. There's not much research usefulness in it. It's more of a presentation, and meant to be attractive.

Krefter said...

Any news on ancient genomes from the Middle east being sampled? I would imagine early Neolithic ones would be the first to be sampled.

Chad Rohlfsen said...

Is this new?

Davidski said...

I just posted the clip with the Hungarian samples. See above.

By the way, ChromoPainter/fineSTRUCTURE aren't new. See here...

Mateus Smith said...

Thanks for clearing that up, everyone.

Tesmos said...

I see, thank you for the heads up David. Btw, Motala12's North Sea score is quite high so could he or/and his people have contributed to the modern cluster North Sea?

Davidski said...

Yeah, I think SHG stragglers contributed to the present-day Scandinavian gene pool, but it's not possible to estimate how much from these results.

Shuffle said...

Well, I am 100 % South-German (Bavarian, and all-over-South-West-Germany) and I come out as 50 % North-Italian and 50 % Swedish in K15, and 50 % East-Norwegian and North-Italian in MDLP. In MDLP in the 4-peoples approximation MDLP gives me some slavic, Eurogenes brings me out as Northern-European, West- and East-Mediterranean. I think this fits pretty good the population history of South-Germany, a kind of a mix between Celto-Romanic peoples (North-Italian is probably an admixture of Gallo-celtic and Mediterranean populations) and Germanic (scandinavianlike populations). The repopulation after the 30-years-war has probably more been due to high birthrates, Germany had very high birthrates until 1910! Some migration might have been from the alpine regions, it is well known that the southern part of the Alsace, the Sundgau, has been repopulated by people from Switzerland, it could have been similar in some bavarian regions with Tyrol, but there was certainly not really massive immigration from Hungary, this would have let many more traces e. g. in Church-records after 1648!

Grey said...


"The manly exploits of Spanish, Swedish, Italian, Croat, Flemish, and French soldiers had changed the racial composition of the people."

In regions where kids need high paternal investment to thrive I don't think rape can have a big genetic impact on its own (capturing a wife is a different thing).

In places like the Congo where the women do the farm work then maybe it could.

If you look at the birth rate in the US doing the colonization period it shows how fast people can breed when there is land to expand into.

If that idea was correct then it might display in higher than expected homogeneity (if that is testable?)

Grey said...

very cool videos btw - very dramatic

Davidski said...

This paper might be very informative in this context.

Arch Hades said...

I think Polako/Eurogenes has dethroned Dienekes as the best Genome blogger on the net.

Simon_W said...

@ Krefter
The Swiss are not an ethnic group but a multiethnic state, and the different groups within are the German Swiss, the French Swiss, the Italian Swiss and the Rhaeto-Romance Swiss. Thus, treating the Swiss as a single population in ADMIXTURE runs makes little sense. It's like with Belgians or the British, but probably even more acute. The only Swiss groups which are similar to North Italians are the Italian Swiss and the Rhaeto-Romance ones. The German Swiss certainly are not like North Italians. Ethnically and linguistically, they are closest to Southwestern Germans and Alsatians (also to the westernmost Austrians, the people of Liechtenstein, and some tiny German minorities in northernmost Italy).

Simon_W said...

@ Arch Hades

I tend to agree. At least Dienekes seems to have reduced his blogging activities greatly, so I'm glad there is the Eurogenes blog. But I'm grateful Dienekes didn't give up blogging completely, it's always good to have some additional channels of information.

Simon_W said...

Very interesting animation. Contains nothing new, but it helps in grasping the differences between the samples fully.

Perhaps the biggest change is between CO1 and BR1. The change is from a typical EEF population to a population that was already quite similar to modern Europeans. The admixing population responsible for this change must have had very strong HG ancestry, a decent amount of ANE, but strikingly little Eastern_Euro and no East_Med and West_Asian. It had a lot of shared drift with Lithuanians. Unfortunately we don't know which y-haplogroups were typical for them. It might have been R1b, but I could imagine as well that it was something unexpected like I2. One thing is clear: If the Yamnaya population was like a 50-50 mix of Karelian HGs and modern Armenians, then the population responsible for the change to BR1 wasn't Yamnaya. Don't know what to make of this atm.

Interestingly, the subsequent populations, exemplified by BR2 and IR1, became increasingly more eastern, as the Eastern_Euro increased greatly, as well as the East_Med and later the West_Asian component. That BR2's haplogroup was J2 goes well in line with this. I guess, at least partly this must be related with increasing Yamnaya ancestry, moving from the Black Sea upwards along the Danube.

So it looks as though the biggest change in ANE levels was achieved quite early on by a HG derived population who wasn't obviously Yamnaya-related, and later eastern groups continued to push westwards.

Another thing that struck me is the South_Asian ancestry in Hinxton4. Can it be possible that this is related with the inflated Gedrosia and MDLP Indo-Iranian ancestry on the British Isles?? It's dubious that this can be explained away as noise; I've checked the K15 spreadsheet, and the South_Asian component is still more than 1% in Orcadians. Could it be possible that R1b in western Europe was associated with slight South_Asian ancestry!? This would mean western European R1b isn't derived from the Carpathian Basin, since neither BR1 nor BR2 had any South_Asian.

Simon_W said...

@ Fanty

The Dodecad K12b oracle suggested that Germans are a mix of Orcadians and Hungarians. Even if this was accurate I'd rather think of Bronze Age population movements than of effects of the 30 years war. Both the Tumulus and the Urnfield burial customs may have originated near Hungary. Also with respect to the French-Scandinavian mix in western Germany, the pre-Germanic Celtic substrate probably had a more important effect than the 30 years war, and in eastern Germany the Slavic substrate.

@ Shuffle

Interesting what you said about the Sundgau. Actually I had noticed before that the people from the Sundgau sound more „Swiss“ than the traditional dialect of the city of Basel does!

Shuffle said...

Yes, the dialect in the Sundgau differs from the other alemannic dialects in the Alsace because of the swiss settlers. If you can read German you can read here a bit

The other way round Hungary had considerable German influence. After the turkwars the Austrians repopulated large abandoned areas in Southern-Hungary with German Settlers (so called Swabians), today there live still about 200000 people of German descent in Hungary, and probably many more are of German descent, because from 1866 on the Hungarians magyarized actively the other nationalities in Hungary and they were rather successful in the case of the swabians, so many took Hungarian names and began to speak Hungarian, Hungarian also was the language of higher education etc. so many Germans got Magyarized, so the demographic impact of Germans on Hungary was considerable!

Simon_W said...

I now have read the paper linked by David, by G. Kulcsar and V. Szeverenyi, it's exciting! I've drawn four main conclusions from it, and an additional fifth insight:

1. The source for the EHG admixture in the Mako culture must be the Yamnaya east of the Tisza - there really is no other explanation, and it seems possible, if this was from the early western Yamnaya periphery where the Caucasus admixture may still have been minimal.

2. Although these Yamnaya people had an important effect on the Mako gene pool, culturally they were not really the dominant source for the Mako culture. Vucedol seems to have had the bigger formative influence. Which means that we cannot conclude that the Yamnaya people imposed their language on the Carpathian Basin. Mere biological importance via genes isn't as good an indicator for linguistic dominance as cultural dominance is. And Vucedol was quite a dominant culture. It was considered a "Kurganised culture" (i.e. patriarchalic, warlike etc) by Gimbutas and they had fortified hilltops etc, really hard to overthrow.

3. Yet Vucedol probably wasn't EHG admixed! The early Vucedol in the south and the west of the Carpathian Basin was contemporaneous with the late Baden in the north of the Carpathian Basin. The latter were typical EEFs, and I really don't believe that people to the west of them were more EHG admixed than them.

4. The idea that R1b in German Bell Beakers is in any way Yamnaya-derived can be thrown into the rubbish bin. Obviously there was never any Yamnaya in Germany. The idea has been put forward that Vucedol influence may have carried Yamnaya admixture westwards along the Danube. But as I have just argued above, Vucedol probably wasn't Yamnaya admixed. They were culturally „Kurganized“ (in the widest possible sense), but not Yamnaya derived.

5. It's possible that the original R1b carriers in western Europe didn't have EHG admixture, but picked it up from the GAC/Corded substrate in Germany. After all, we now know that the dominant haplogroup of mondern Scandinavians, I1, was from EEF people. Hence, probably neither IE language nor strong HG ancestry was associated with the haplogroup that is now dominant in northern Germanics. With R1b it may have been similar. Or partly so.

Simon_W said...

And in case someone thinks the late Vucedol, which was contemporaneous with the early Mako, might have taken R1b to the German Bell Beakers - this doesn't work out either, because late Vucedol was a Bronze Age culture, while German Bell Beakers still were on a Copper Age level. Forget it.