How much Celto-Germanic are Iberians and North Italians?

[Token]

The fact that extra Yamnaya admixture to the exclusion of the local Beaker samples of Southwestern Europe can be detected in modern-day Iberians and Italians might represent later steppe-rich Celto-Germanic historical introgessions. Something similar can be observed in the Balkans, where additional Yamnaya in modern-day Greeks to the exclusion of Mycenaeans is suggestive of Slavic input.

In order to detect later admixture events, such as Hallstatt movements spreading Celtic languages to Iberia and Northern Italy and later Völkerwanderung, i've isolated the pre Iron Age genepool of modern-day Southwestern Europeans, represented by the local Bronze Age averages.

As references to post IA introgressions, i've used Anglo-Saxon as a proxy for generic Germanic admixture taking into account that all Migration Period Germanic populations sampled until now cluster quite close to each other. The Bylany Hallstatt outgroup serves as a proxy for Celtic admixture for obvious reasons. Everything in conform with historical accounts.

[1] "distance%=1.4002"

Portuguese

Beaker_Iberia,49.4
Anatolia_BA,15.6
England_Anglo-Saxon,13.6
Mozabite,11.2
Hallstatt_Bylany,10.2


[1] "distance%=1.7954"

Spanish_Castilla_Y_Leon

Beaker_Iberia,47
England_Anglo-Saxon,19.8
Anatolia_BA,13.6
Mozabite,11.2
Hallstatt_Bylany,8.4

[1] "distance%=1.3976"

Italian_Bergamo

Beaker_Northern_Italy,45.4
Anatolia_BA,22.2
Hallstatt_Bylany,17.8
England_Anglo-Saxon,13.6
Mozabite,1

[Not a Cop]

Is this K25? Are you shure it's sensitive enough to defferinate between proto-Celts and proto-Germanics? Can you show fits with only Celts and only Germanics?

[Token]

Is this K25? Are you shure it's sensitive enough to defferinate between proto-Celts and proto-Germanics? Can you show fits with only Celts and only Germanics?

Yes, to some extent. Using only Hallstatt or only Germanic worsen the fit significantly for all of them.
Here are Basques for comparison, comparatively much less Germanic and no Celtic at all. Pretty accurate, as there were no Celts in Vasconia.

[1] "distance%=1.9562"

Basque_Spanish

Beaker_Iberia,91.8
England_Anglo-Saxon,5.8
Anatolia_BA,2.4

Using the Celto-Germanic Roman era gladiator as proxy works equally well.

[1] "distance%=1.6884"

Spanish_Castilla_Y_Leon

Beaker_Iberia,47.6
England_Roman,28.6
Anatolia_BA,12.6
Mozabite,11.2

[1] "distance%=1.4448"

Italian_Bergamo

Beaker_Northern_Italy,52
England_Roman,27
Anatolia_BA,19.4
Mozabite,1.6

[Damião de Góis]

Ideally you shoud use Ostrogoths for North Italy and Visigoths and Suebi for Iberia. Not sure about the celts.
What does Galicia (Suebi) and Asturias (supposed Visigoth refuge) score?

[Token]

Ideally you shoud use Ostrogoths for North Italy and Visigoths and Suebi for Iberia. Not sure about the celts.
What does Galicia (Suebi) and Asturias (supposed Visigoth refuge) score?

There are no Ostrogothic or Suebic samples avaiable for now.

[FilhoV]

The fact that extra Yamnaya admixture to the exclusion of the local Beaker samples of Southwestern Europe can be detected in modern-day Iberians and Italians might represent later steppe-rich Celto-Germanic historical introgessions. Something similar can be observed in the Balkans, where additional Yamnaya in modern-day Greeks to the exclusion of Mycenaeans is suggestive of Slavic input.

In order to detect later admixture events, such as Hallstatt movements spreading Celtic languages to Iberia and Northern Italy and later Völkerwanderung, i've isolated the pre Iron Age genepool of modern-day Southwestern Europeans, represented by the local Bronze Age averages.

As references to post IA introgressions, i've used Anglo-Saxon as a proxy for generic Germanic admixture taking into account that all Migration Period Germanic populations sampled until now cluster quite close to each other. The Bylany Hallstatt outgroup serves as a proxy for Celtic admixture for obvious reasons. Everything in conform with historical accounts.

[1] "distance%=1.4002"

Portuguese

Beaker_Iberia,49.4
Anatolia_BA,15.6
England_Anglo-Saxon,13.6
Mozabite,11.2
Hallstatt_Bylany,10.2


[1] "distance%=1.7954"

Spanish_Castilla_Y_Leon

Beaker_Iberia,47
England_Anglo-Saxon,19.8
Anatolia_BA,13.6
Mozabite,11.2
Hallstatt_Bylany,8.4

[1] "distance%=1.3976"

Italian_Bergamo

Beaker_Northern_Italy,45.4
Anatolia_BA,22.2
Hallstatt_Bylany,17.8
England_Anglo-Saxon,13.6
Mozabite,1


Run my coordinates on this run

I sent it via messages a few weeks ago

[Jana]

Fantastic thread , looks pretty realistic to me :thumb001:

[Ibericus]

..Beakers themselves also contain the same ancestry as Celto-Germanics (Bronze-Age Central EUrope) so it's much higher than that.

[Coolguy1]

[1] "distance%=2.1329 / distance=0.021329"

Spanish_Andalusia (N=5)

Basque (N=9) 57.1
Mycenaean (N=4) 18.8
Corded_Ware_Germany_M348611 15.8
Moroccan (N=4) 8.3

This is just a hypothetical fit, who knows whether it has any historical accuracy.

[Grace O'Malley]

This is what I come out using those pops. I know it's not a good fit but I just find it interesting to compare. Everything gets soaked up into the Anglo-Saxon because no Beaker_Britain or Central European Beaker.

Distance 1.8454

Anatolia_EBA 0
Beaker_Iberia 6.67
England_Anglo Saxon 86.67
Hallstatt_Bylany 6.67
Mozabite 0

Irish

Fit 1.2395

Anatolia_EBA 0.83
Beaker_Iberia 7.5
England_Anglo Saxon 85.83
Hallstatt_Bylany 5.83
Mozabite 0

Replacing with England_Roman

Fit 2.0903 (Myself)

Anatolia EBA 0
Beaker_Iberia 0
England_Roman 95.83
Hallstatt Bylany 4.17
Mozabite 0

Fit 1.2903 (Irish)

Anatolia_EBA 0
Beaker_Iberia 0.83
England_Roman 98.33
Hallstatt_Bylany 0.83
Mozabite 0

[Token]

..Beakers themselves also contain the same ancestry as Celto-Germanics (Bronze-Age Central EUrope) so it's much higher than that.

Beaker most certainly didn't spoke neither Celtic nor Germanic. They probably spoke something related but much more archaic and closer to PIE. Celtic and Germanic couldn't have been fully differentiated earlier than the Late Bronze Age, and Celtic package spread mostly by elite dominance over Bronze Age natives by Hallstatt chiefs. Proto-Germanic developed in Scandinavia and didn't reached Central Europe earlier than the Iron Age.

[Peterski]

Northern Iberia also has substantial Insular Celtic and Breton (from French Brittany) admixture, so you should add some Iron Age or Roman Era Britons to your model, not just Hallstatt.

Meanwhile Anglo-Saxons are probably not the best proxy for Suebians, Vandals and Visigoths.

[Rædwald]

How many Spaniards on average are I1?

[Token]

Northern Iberia also has substantial Insular Celtic and Breton (from French Brittany) admixture, so you should add some Iron Age or Roman Era Britons to your model, not just Hallstatt.
They settled there but how can you prove that? Galicia does draw some Breton, but G25 has some problems in differentiating Northern European admixture as people living there are and were always very similar in terms of deep ancestry. For something like that you'd need to search for patterns of rare allele sharing.

[1] "distance%=1.4746"

Spanish_Galicia

Beaker_Iberia,45.8
Anatolia_BA,13.8
Mozabite,12.2
England_IA,11.6
England_Anglo-Saxon,10
Hallstatt_Bylany,6.6

[Peterski]

They settled there but how can you prove that?

Modern Galicians and Asturians have strong affinity with Brittany, Cornwall, Wales, Ireland.


[1] "distance%=1.4746"

Spanish_Galicia

Beaker_Iberia,45.8
Anatolia_BA,13.8
Mozabite,12.2
England_IA,11.6
England_Anglo-Saxon,10
Hallstatt_Bylany,6.6

Try also with England Roman because those who migrated were Romano-Britons.

[Dick]

How many Spaniards on average are I1?

I did the Big Y test at ftdna and oddly enough I have a Spanish match with the exact same Snps. Time to most recent common ancestor; 2700 years before present. It's not some close match but interesting since our lineages are from the opposite sides of Europe.

[Token]

Modern Galicians and Asturians have strong affinity with Brittany, Cornwall, Wales, Ireland.
I don't think so, show me data.



Try also with England Roman because those who migrated were Romano-Britons.
England Roman was most likely not a native.

[StonyArabia]

I did the Big Y test at ftdna and oddly enough I have a Spanish match with the exact same Snps. Time to most recent common ancestor; 2700 years before present. It's not some close match but interesting since our lineages are from the opposite sides of Europe.

Some Slavic people were present in Moorish Spain as mercenaries, could be from them, quite interesting.

[Peterski]

BTW what is the historical explanation of high North African admix in Galicia?

It has unusually high North African for a region in the north of Iberia.


England Roman was most likely not a native.

Which one? There were several samples.


Some Slavic people were present in Moorish Spain as mercenaries, could be from them, quite interesting.

They even started their own kingdom but it was around modern Valencia/Murcia borderland and in Baleares:

https://en.m.wikipedia.org/wiki/Taifa_of_Dénia

https://en.m.wikipedia.org/wiki/Mujāhid_al-ʿĀmirī

[Peterski]

I did the Big Y test at ftdna and oddly enough I have a Spanish match with the exact same Snps. Time to most recent common ancestor; 2700 years before present. It's not some close match but interesting since our lineages are from the opposite sides of Europe.

I also did it, soon you will see a Polish flag somewhere here, next to Spanish and British flags:

https://www.yfull.com/tree/R-L617/

https://i.imgur.com/wlztMpF.png

[Damião de Góis]

If I1 peaks in Galicia / North Portugal then the most likely explanation is the Suebi Kingdom. But even that is speculation.

[Dick]

If I1 peaks in Galicia / North Portugal then the most likely explanation is the Suebi Kingdom. But even that is speculation.

My match is from Galicia. Admins said probably from Suebi.


Suebi were assimilated by Slavs too since they came from eastern Europe


https://upload.wikimedia.org/wikipedia/commons/2/2b/Suebic_migrations.jpg

[Dick]

I also did it, soon you will see a Polish flag somewhere here, next to Spanish and British flags:


your subclade can always change though with new matches. Mine has changed twice so far.

[Peterski]

your subclade can always change though with new matches. Mine has changed twice so far.

I still don't have full results yet. I only got STR results, but I can't download my BAM file yet.

[Dick]

I still don't have full results yet. I only got STR results, but I can't download my BAM file yet.

You could've inputted them to ysearch but apparently it's illegal now in the EU

http://ysearch.org/

[Peterski]

http://ysearch.org/

It says:

Service Unavailable
HTTP Error 503. The service is unavailable.

Maybe I will send you in a PM and you will input them?

[Grace O'Malley]

There are Galicians in the G25 database so they can be modelled if someone wants to do it. You should pick up differences with them if they are different than other Spanish.

[Dick]

It says:

Service Unavailable
HTTP Error 503. The service is unavailable.

Maybe I will send you in a PM and you will input them?

I can't access it anymore either. They were a website from the EU so I guess they shut them down totally.


https://i.imgur.com/KzB7ZRI.jpg

[Rædwald]

I can't access it anymore either. They were a website from the EU so I guess they shut them down totally.


https://i.imgur.com/KzB7ZRI.jpg

Fucking EU.

[Aren]

There’s 2000-3000 year gap between the beakers and the Celtic/Germanic migrations. I think your numbers are way too inflated, and we’ll probably see this when more relevant roman era samples from southern Europe start to show up.

[Peterski]

There’s 2000-3000 year gap between the beakers and the Celtic/Germanic migrations. I think your numbers are way too inflated, and we’ll probably see this when more relevant roman era samples from southern Europe start to show up.

I wonder if these people were R1b like modern Basques:

https://en.m.wikipedia.org/wiki/List_of_the_Pre-Roman_peoples_of_the_Iberian_Peninsula#Non-Indo-European_speakers

They occupied a very large part of the peninsula, not just some enclaves.

[Token]

There’s 2000-3000 year gap between the beakers and the Celtic/Germanic migrations. I think your numbers are way too inflated, and we’ll probably see this when more relevant roman era samples from southern Europe start to show up.

You are exaggerating, the time gap between Iberian Beaker and Hallstatt is not more than 1500 years. Anyway, there were no relevant migrations into Iberia during these ~1500 years gap, just like there were no migrations into the British Isles until the Iron Age Celtic expansions. Late Iron Age Iberians were Basque like, so not much different from the local Beaker samples. There is a ~4500 years gap from Bell Beaker to modern-day Basques, and they are still broadly similar.

[1] "distance%=2.013"

Basque_Spanish

Beaker_Iberia,94.2
Hallstatt_Bylany,3.8
Anatolia_BA,2

Indo-European speaking Iberians clearly have additional Celto-Germanic, Roman and North African admixture. My numbers makes a lot of sense.

[1] "distance%=1.4562"

Spanish_Castilla_Y_Leon

Basque_Spanish,48.6
Hallstatt_Bylany,14.8
England_Anglo-Saxon,13.4
Anatolia_BA,11.8
Mozabite,11.4

[Peterski]

There are Galicians in the G25 database so they can be modelled if someone wants to do it.

I don't have time for G25 now, unless someone sends me a spreadsheet with only ancient populations included*. But if you use G25 then don't pre-select just 5 or 6 populations. Do runs with 50 or 100 populations. I think it is silly when people use models designed specifically to confirm their preconceived ideas (like Token did - he specifically designed his model in such a way to show what he wanted to prove).

As my post above shows, Galicians have ca. 10% of Cornwall/Bretagne ancestry - and it is confirmed by historical records.

*I won't use models with just 5-6 pre-selected ancient populations like Token did, but if anything I will run it with all populations from a given period (for example: all Late Bronze Age populations; all Iron Age populations, etc.) and let the algorithm choose freely. But as Token said, I'm not sure if G25 is good for such "free models", because it has problems with differentiating between some groups.

=====

Token, why not use RISE174 (South Sweden IA, 427-611 AD) instead of Anglo-Saxons, who were likely mixed with Celts.

According to Jordanes Goths originated from Scandinavia, not from England. Suebians as well never lived in England.

You should as well include Insular Celts in your model, and include a non-nonsense Germanic population - some which makes more sense (Swedish RISE174, or even unmixed Lombards, Bajuwarii or whatever - anything would be better than Anglo-Saxons).

If you want an unbiased model you should include many populations or at least populations that make sense.

BTW, there is no hard evidence that Hallstatt was ancestral to all Celts including Celtiberians. This is just one of theories.

[Grace O'Malley]

Run my coordinates on this run

I sent it via messages a few weeks ago

This is what I get for you.

Fit 1.5101(Scaled)

Anatolia_EBA 17.5
Beaker_Iberia 33.33
England_Anglo-Saxon 2.5
Hallstatt_Bylany 35
Mozabite 11.65

[Peterski]

Maybe Jutland Iron Age or/and Wielbark will be good proxy for Goths:

http://www.actabp.pl/#Archiwum?./supl/2_2018.html

https://i.postimg.cc/VN9xkBV3/Screenshot_20180929-073310_Drive.jpg

[Grace O'Malley]

I don't have time for G25 now, unless someone sends me a spreadsheet with only ancient populations included*. But if you use G25 then don't pre-select just 5 or 6 populations. Do runs with 50 or 100 populations. I think it is silly when people use models designed specifically to confirm their preconceived ideas (like Token did - he specifically designed his model in such a way to show what he wanted to prove).

As my post above shows, Galicians have ca. 10% of Cornwall/Bretagne ancestry - and it is confirmed by historical records.

*I won't use models with just 5-6 pre-selected ancient populations like Token did, but if anything I will run it with all populations from a given period (for example: all Late Bronze Age populations; all Iron Age populations, etc.) and let the algorithm choose freely. But as Token said, I'm not sure if G25 is good for such "free models", because it has problems with differentiating between some groups.

=====

Token, why not use RISE174 (South Sweden IA, 427-611 AD) instead of Anglo-Saxons, who were likely mixed with Celts.

According to Jordanes Goths originated from Scandinavia, not from England. Suebians as well never lived in England.

You should as well include Insular Celts in your model, and include a non-nonsense Germanic population - some which makes more sense (Swedish RISE174, or even unmixed Lombards, Bajuwarii or whatever - anything would be better than Anglo-Saxons).

If you want an unbiased model you should include many populations or at least populations that make sense.

BTW, there is no hard evidence that Hallstatt was ancestral to all Celts including Celtiberians. This is just one of theories.

They match Hallstatt better than the Irish do.

[Peterski]

They match Hallstatt better than the Irish do.

But according to more recent theories, Celts entered Iberia already in the Late Bronze Age.

And Hallstatt is an Iron Age culture, which was probably not ancestral to all Celtic groups.

[Grace O'Malley]

But according to more recent theories, Celts entered Iberia already in the Late Bronze Age.

And Hallstatt is an Iron Age culture, which was probably not ancestral to all Celtic groups.

Yes I agree there. I don't think Celts were uniform anyway.

[Peterski]

I don't think Celts were uniform anyway.

TBH, I think that every "Proto-People" were uniform at the beginning.

Those Indo-European groups were extremely patriarchal and patrilineal but quite inclusive for foreign women, I suppose.

So as men took wifes, their offspring became autosomally diversified.

[Kelmendasi]

I wonder if these people were R1b like modern Basques:

https://en.m.wikipedia.org/wiki/List_of_the_Pre-Roman_peoples_of_the_Iberian_Peninsula#Non-Indo-European_speakers

They occupied a very large part of the peninsula, not just some enclaves.
Imo they probably had a good amount of I2a1a-M26. The R1b in Basques is from IE input. I guess the ancestral population to the Basque could also have had large amounts of R1b from contact with IE speaking neighbors.

[Peterski]

The R1b in Basques is from IE input.

Basques are over 90% R1b though - more than Indo-European speakers in Iberia.

How did they become R1b while retaining their Non-Indo-European language?

And why are all other Iberians less R1b than Basques?

====

Before the Roman conquest, large parts of Iberia were Non-Indo-European:

https://en.m.wikipedia.org/wiki/List_of_the_Pre-Roman_peoples_of_the_Iberian_Peninsula#Non-Indo-European_speakers

The same applies to Italy before Romanization, there were many Non-IE groups.

[Grace O'Malley]

TBH, I think that every "Proto-People" were uniform at the beginning.

Those Indo-European groups were extremely patriarchal and patrilineal but quite inclusive for foreign women, I suppose.

So as men took wifes, their offspring became autosomally diversified.

I don't know what's up with those Anglo-Saxons.

Fit 1.6459 (Me)

Beaker_Britain 31.67
England_Anglo-Saxon 63.33
Hallstatt_Bylany 1.67
Ireland_MN 1.33

Fit 1.2064 (Irish:Average)

Beaker_Britain 20.83
England_Anglo-Saxon 70.83
Hallstatt_Bylany 3.33
Ireland_MN 5

[Peterski]

Fit 1.2064 (Irish:Average)

Beaker_Britain 20.83
England_Anglo-Saxon 70.83
Hallstatt_Bylany 3.33
Ireland_MN 5

Try this model: remove Anglo-Saxons, add Iron Age Britain and add RISE174 (Iron Age Sweden).

[Pribislav]

Basques are over 90% R1b though - more than Indo-European speakers in Iberia.

How did they become R1b while retaining their Non-Indo-European language?

And why are all other Iberians less R1b than Basques?

R1b men raped native Iberian women.
Women which were raped by R1b men born children and learned them own language. It's well known than children learn language from mother.

[Kelmendasi]

Basques are over 90% R1b though - more than Indo-European speakers in Iberia.

How did they become R1b while retaining their Non-Indo-European language?

And why are all other Iberians less R1b than Basques?
This is just because of R1b males having more male children than their non-R1b counterparts. Ydna frequency doesn't mean much in a lot of cases as it can fluctuate very easily. I'm guessing they retained it from women teaching their children their native language or it could just be from rapes or something like that where the male isn't present in the family. The clade of R1b that Basques carry is definitely IE in origin. Imo, I2a1a-M26 and I2a2a were common Pre-IE haplogroups in Iberia.

[Peterski]

R1b men raped native Iberian women.

Idiotic theory.

[Peterski]

This is just because of R1b males having more male children than their non-R1b counterparts.

There has to be a reason why they had more children. Serological conflict? Immunity to plague?

BTW, I most likely share the same R1b subclade with Basque HG01606 (from 1000 Genomes):

https://i.imgur.com/wlztMpF.png

[Kelmendasi]

There has to be a reason why they had more children. Serological conflict? Immunity to plague?

BTW, I most likely share the same R1b subclade with Basque HG01606 (from 1000 Genomes):

[ig]https://i.imgur.com/wlztMpF.png[/img]
Not sure on why they would have had more children but it could be because of things like better living conditions or just luck in some cases I guess. Interesting that one of the clades seems to be exclusively British and with a low TMRCA, the ancestral clades seem to be linked to Iberia though.

[Peterski]

Ydna frequency doesn't mean much in a lot of cases

Actually it does mean a lot, and "everything is random drift" explanations are bullshit in most cases.

So far all ancient DNA studies show that Y-DNA frequencies correlate very well with ethnic groups.

[Pribislav]

Idiotic theory.

Spaniard Conquistadors in Mexico raped native women. Because of that modern Mexicans are predominantly R1b, and autosomally they are predominantly native (Amerindians).

World would be much peaceful place without R1b people. From the time when they arrived to Europe from Central Asian 5000 years ago they constantly made genocided. R1b made genocid against I and G people in Europe 5000 years ago. Genocide against native population of America, Australia, parts of Africa and Asia was by R1b people.

[Peterski]

the ancestral clades seem to be linked to Iberia though.

Not necessarily. These 3 samples are listed as R1b-L617* in that tree:

YF14855 is Iberian, YF04446 is English and YF04846 is Lithuanian with a Polish surname (but removed his flag for some reason).

My sample is not there but will be added within few months probably.

[Kelmendasi]

Actually it does mean a lot, and "everything is random drift" explanations are bullshit in most cases.

So far all ancient DNA studies show that Y-DNA frequencies correlate very well with ethnic groups.
Depends. Things like diversity matter more and are more accurate when determining origin. Subclades are things that matter as well. Frequency can fluctuate, E-V13 for example is likely to have been less common in the Balkans and haplogroups like R1b probably were far more common. Sure subclades can be linked to ethnicity

[Grace O'Malley]

Try this model: remove Anglo-Saxons, add Iron Age Britain and add RISE174 (Iron Age Sweden).

I couldn't find any Iron Age from Sweden so I added Swedish Vikings. Iron Age Britain is the majority.

Fit 1.624 (Myself)

Beaker_Britain 35
England_IA 57.5
Ireland_MN 0
Sweden_Viking_Age 7.5

Fit 1.372 (Irish:Average)

Beaker_Britain 41.67
England_IA 53.33
Ireland_MN 2.5
Sweden_Viking_Age 2.5

[Kelmendasi]

Not necessarily. These 3 samples are listed as R1b-L617* in that tree:

YF14855 is Iberian, YF04446 is English and YF04846 is Lithuanian with a Polish surname (but removed his flag for some reason).

My sample is not there but will be added within few months probably.
DF27 basal samples seem to be of Iberian origin or ancestry https://www.yfull.com/tree/R-DF27/. Though there are various subclades that are found among other groups such as Sardinians, Brits and even some Belarusians

[Peterski]

I couldn't find any Iron Age from Sweden so I added Swedish Vikings. Iron Age Britain is the majority.

Fit 1.624 (Myself)

Beaker_Britain 35
England_IA 57.5
Ireland_MN 0
Sweden_Viking_Age 7.5

Fit 1.372 (Irish:Average)

Beaker_Britain 41.67
England_IA 53.33
Ireland_MN 2.5
Sweden_Viking_Age 2.5

Well now you can see that obviously the Anglo-Saxons from Global25 are already mixed with Insular Celts...

Iron Age Sweden is RISE174, and you can see that this sample plots far away from some of Anglo-Saxons:

"Germanic (Sweden) 427-611 calCE" is Late Iron Age Sweden (RISE174):

https://www.mupload.nl/img/1yu03d28fa278.png

[Peterski]

DF27 basal samples seem to be of Iberian origin or ancestry

Isn't the oldest known DF27 from Quedlinburg?:

I0806 Bell Beaker, Quedlinburg (2431-2150 BC)
Y-DNA Haplogroup: R1b1a1a2a1a2a-DF27

[Peterski]

Isn't the oldest known DF27 from Quedlinburg?:

I0806 Bell Beaker, Quedlinburg (2431-2150 BC)
Y-DNA Haplogroup: R1b1a1a2a1a2a-DF27

This guy is the oldest known DF27 to date, AFAIK: https://www.oagr.org.au/source/I0806/

http://3.bp.blogspot.com/-bEd_WjzHlpw/VNu4DQTj_0I/AAAAAAAAE-I/zj3c3UucJP8/s1600/Capture.JPG

In the paper he was P312+, but on Anthrogenica they confirmed him to be DF27, AFAIK.

[Kelmendasi]

Isn't the oldest known DF27 from Quedlinburg?:

I0806 Bell Beaker, Quedlinburg (2431-2150 BC)
Y-DNA Haplogroup: R1b1a1a2a1a2a-DF27
Probably is. There was another old DF27 sample found in southern Spain during the late Bronze Age, apparently the site is linked to the Cogotas culture and is believed to have had Bell Beaker influence. So a Bell Beaker spread of DF27 is likely I assume

[Grace O'Malley]

There has to be a reason why they had more children. Serological conflict? Immunity to plague?

BTW, I most likely share the same R1b subclade with Basque HG01606 (from 1000 Genomes):

https://i.imgur.com/wlztMpF.png

They could have had more children if the men became high status and monopolised the women. M222 spread through Northwestern Ireland/Northern Ireland and parts of Scotland because of the Ui Neill dynasty.

[Peterski]

DF27 is in fact quite common in Britain (and my particular subclade has more people from Cornwall than from any other region).

But maybe it is because British people buy DNA tests more often than Iberians or Poles, etc.

[Grace O'Malley]

DF27 is in fact quite common in Britain (and my particular subclade has more people from Cornwall than from any other region).

But maybe it is because British people buy DNA tests more often than Iberians or Poles, etc.

Yes they do. FTDNA is very lopsided with Irish and British.

[Peterski]

They could have had more children if the men became high status and monopolised the women. M222 spread through Northwestern Ireland/Northern Ireland and parts of Scotland because of the Ui Neill dynasty.

In patriarchal societies such as Indo-Europeans children learn the language of the father, not of the women (who were basically treated like commodities by Indo-Europeans); so your explanaton would make sense only if the Basques (and their predecessors - the Aquitani) were IE speakers:

https://www.youtube.com/watch?v=ErXa5PyHj4I#t=4m23s

[Peterski]

Yes they do. FTDNA is very lopsided with Irish and British.

Still DF27 is about as common in Britain as U152, but proportions differ by region:

https://i.pinimg.com/originals/ff/2a/a5/ff2aa5c04d9c90b579a0b0255cc35134.png

[Peterski]

Spaniard Conquistadors in Mexico raped native women. Because of that modern Mexicans are predominantly R1b, and autosomally they are predominantly native (Amerindians).

That's not what happened. Spanish men married some native women. Their children acquired genetic immunity to European diseases from their fathers, and therefore survived recurrent epidemics during the 1500s and 1600s better than pure-blooded Natives, who were dying like flies.

Furthermore modern Mexicans are on average autosomally 50/50 (they are not mostly Amerindian), and in Southern Mexico - where Native populations survived better than in the North - R1b is a minority haplogroup and Native Y-DNA (Q and C) predominate there.

Also modern Mexicans speak Spanish - they adopted the language of R1b men.

So all the factors are very different than in case of Basques.

[Aren]

You are exaggerating, the time gap between Iberian Beaker and Hallstatt is not more than 1500 years. Anyway, there were no relevant migrations into Iberia during these ~1500 years gap, just like there were no migrations into the British Isles until the Iron Age Celtic expansions. Late Iron Age Iberians were Basque like, so not much different from the local Beaker samples. There is a ~4500 years gap from Bell Beaker to modern-day Basques, and they are still broadly similar.

[1] "distance%=2.013"

Basque_Spanish

Beaker_Iberia,94.2
Hallstatt_Bylany,3.8
Anatolia_BA,2

Indo-European speaking Iberians clearly have additional Celto-Germanic, Roman and North African admixture. My numbers makes a lot of sense.

[1] "distance%=1.4562"

Spanish_Castilla_Y_Leon

Basque_Spanish,48.6
Hallstatt_Bylany,14.8
England_Anglo-Saxon,13.4
Anatolia_BA,11.8
Mozabite,11.4

You can't dismiss thousands of years of history just because we aren't aware of any great migrations today. You runs are indicative of all those admixtures(especially North African input), but not set in stone. We would need Roman era samples and pre-Celtic ones aswell. I def think there's some genuine Celtic and Germanic input in Iberia but probably in very low numbers.

Also your run is flawed. You are not using Copper Age/EBA farmer populations of that era in the run. In addition to that, there was hardly any pure Anatolia_BA migrations to Spain, all that "Eastern" BA input must've made it there mostly with the Romans, and since we don't have Roman samples yet we can't rule out that addtional Northern Euro/Beaker like input must've arrived with them aswell, not just Anatolia_BA admix.

Look at this run, better fit and with a relevant sample added(Iberia_CA)

"distance%=1.6875"

Basque_Spanish

Iberia_BA,59.6
England_Anglo-Saxon,23.4
Iberia_Central_CA,10.4
Hallstatt_Bylany,4.8
Anatolia_BA,1.8

Obviously Basques don't have 23% Germanic input, but that the Beaker samples we have now from Iberia are somewhat southern shifted. New samples are on the way with more Yamnaya input.

[Peterski]

M222 spread through Northwestern Ireland/Northern Ireland and parts of Scotland because of the Ui Neill dynasty.

M222 as a whole is ca. 2000 years old (TMRCA):

https://www.yfull.com/tree/R-M222/

I'm sure that dynasty spread only some part of it.

[Peterski]

I guess Polish equivalent of M222 is YP254 (around 10% of all Polish men have this subclade):

https://www.yfull.com/tree/R-YP254/

It also has a high "Poland Concentration Index". About 64% to 79% of all YP254 in the world are Polish.

While YP254 as a whole has TMRCA 2000 years ago, many of its sub-branches have TMRCA around 1050 years ago coinciding with the expansion of the Early Polish Kingdom.

[Pribislav]

That's not what happened. Spanish men married some native women. Their children acquired genetic immunity to European diseases from their fathers, and therefore survived recurrent epidemics during the 1500s and 1600s better than pure-blooded Natives, who were dying like flies.

Furthermore modern Mexicans are on average autosomally 50/50 (they are not mostly Amerindian), and in Southern Mexico - where Native populations survived better than in the North - R1b is a minority haplogroup and Native Y-DNA (Q and C) predominate there.

Also modern Mexicans speak Spanish - they adopted the language of R1b men.

So all the factors are very different than in case of Basques.

Explain than by Basques are most IE people on the world in y dna (90% R1b), and they speak non-IE language?

[FilhoV]

A lot of the Mexicans I see online are Haplogroup J I’ve noticed

[Peterski]

Explain than by Basques are most IE people on the world in y dna (90% R1b), and they speak non-IE language?

Maybe Proto-Basque speakers also expanded from the Steppe during the Bronze Age, together with PIE speakers.

[Aren]

Explain than by Basques are most IE people on the world in y dna (90% R1b), and they speak non-IE language?

Basques are not 90% R1b, weird how this number keeps increasing and increasing on anthrofora xD
Either way modern day distribution don't mean much when we have ancient DNA confirming that the R1b found in all modern European men spread with IE speakers.

[Peterski]

Basques are not 90% R1b

Actually 92% based on this 2015 study, see Supplementary Table S1:

https://www.nature.com/articles/ejhg2015114#supplementary-information

https://i.imgur.com/nh3kse0.png

I'm not counting ethnic Non-Basques living in Basque County of course.


R1b found in all modern European men spread with IE speakers.

Or with Proto-Basque speakers from the Steppe?:


Maybe Proto-Basque speakers also expanded from the Steppe during the Bronze Age, together with PIE speakers.

Did you interview them? Bones cannot speak.

[Aren]

Actually 92% based on this 2015 study, see Supplementary Table S1:

https://www.nature.com/articles/ejhg2015114#supplementary-information

https://i.imgur.com/nh3kse0.png

I'm not counting ethnic Non-Basques living in Basque County of course.



Or with Proto-Basque speakers from the Steppe?:



Did you interview them? Bones cannot speak.

The consensus today is that R1b-L51 spread with Steppe auDNA. I don't have the time or will to discuss some fringe pet theories tbh. Maybe e-mail David Reich instead and discuss with him why you think the R1b found in modern day European men is proto-Basque.

[Peterski]

The consensus today is that R1b-L51 spread with Steppe auDNA.

That's the majority theory.

However not a single L51 has been found in Early Steppe so far (all Yamnaya were Z2103 - which is Eastern R1b). BTW, I'm not saying that Basque R1b is not from the Steppe - I'm just saying that that there could be Proto-Basque speakers in Northern Caucasus (linguists do not exclude the possibility that Proto-Basque language originated from R1b people in Northern Caucasus).

[Peterski]

If you think that Celtic and Germanic expanded only during the Iron Age - not with Bell Beakers in the Bronze Age - then I'm sorry but we don't have any Indo-European language spoken in Western Europe today that can be dated further back than the Iron Age.

So what evidence you have that Early Bell Beakers in Iberia and Britain spoke some Indo-European language?

Which Indo-European language did they speak? We don't even know its name.

Maybe they indeed spoke Vasconic: https://upload.wikimedia.org/wikipedia/commons/thumb/9/92/Proposed_area_of_Vasconic_languages.png/600px-Proposed_area_of_Vasconic_languages.png

https://upload.wikimedia.org/wikipedia/commons/thumb/9/92/Proposed_area_of_Vasconic_languages.png/600px-Proposed_area_of_Vasconic_languages.png


Maybe e-mail David Reich instead and discuss with him why you think the R1b found in modern day European men is proto-Basque.

IIRC, David Reich recently supports origin of R1b-M269 in the Middle East (Armenian Plateau, Caucasus or Iran). Basque and other Non-IE languages spoken in Western Europe until recently (such as Etruscan, Pelasgian, Rhaetian, etc.) could originate there atfer the Neolithic period. Areas like Gascony, Catalonia, Tuscany, Alpine part of Lombardy, South Tyrol etc. were Non-IE speaking until historical times.

[Peterski]

The consensus today is that R1b-L51 spread with Steppe auDNA.

Tell me which Indo-European languages did Early Bell Beakers in Britain and Iberia speak? Go on. You personally claimed that Celtic emerged in Hallstatt, Germanic in Jastorf - so what do we know about languages spoken by Early Bronze Age Beakers?

Steppe auDNA isn't a name of a language, as far as I know.

Also a lot of people are claiming that R1b-M269 actually originated in Iran or south of the Caucasus.

How do you explain high % of R1b in Assyrians, whose language is closer to Etruscan than to Celtic.

[Token]

You can't dismiss thousands of years of history just because we aren't aware of any great migrations today. You runs are indicative of all those admixtures(especially North African input), but not set in stone. We would need Roman era samples and pre-Celtic ones aswell. I def think there's some genuine Celtic and Germanic input in Iberia but probably in very low numbers.
Suppositions. Even though some population movement might have occured, i doubt they had any significant influence on the local genepool, most of the times archeology and linguistics corresponds pretty well to genomics. We know that more than half of Iberia was Celtic speaking before Romans made it there, so you should expect a lot of Celtic admixture, not much less than in the British Isles, even though it is obvious that Celtic material culture and language spread mostly by acculturation of natives.


In addition to that, there was hardly any pure Anatolia_BA migrations to Spain, all that "Eastern" BA input must've made it there mostly with the Romans, and since we don't have Roman samples yet we can't rule out that addtional Northern Euro/Beaker like input must've arrived with them aswell, not just Anatolia_BA admix.
Yes, who said otherwise? Northern input via Romans would only make sense if Romans were more Northern European-like than native Iberians, and that is counterintuitive. If Romans were as much Beaker-like as Iberians or close to that, the deep ancestry (ENF-WHG-Steppe ratio) of Roman era samples would not be much different from the earlier populations, and only the eastern input to the exclusion of previous populations would be suggestive of Roman admixture (talking strictly about ADMIXTURE).


Look at this run, better fit and with a relevant sample added(Iberia_CA)

"distance%=1.6875"

Basque_Spanish

Iberia_BA,59.6
England_Anglo-Saxon,23.4
Iberia_Central_CA,10.4
Hallstatt_Bylany,4.8
Anatolia_BA,1.8

Obviously Basques don't have 23% Germanic input, but that the Beaker samples we have now from Iberia are somewhat southern shifted. New samples are on the way with more Yamnaya input.
The algorithm is compensating the excess of Iberia_CA with more 'Anglo-Saxon' input, Basques are still very close to both Iberia_BA and Beaker_Iberia. Focusing on better fits isn't always ideal when using averages from a so limited data set, a lot of weird percentages can pop up due to the algorithm trying to find a better fit, like the Irish coming up as ~80% Anglo-Saxon in Grace's model to the complete exclusion of Beaker_Britain. Sometimes you should sacrifice statistical distance for a more intuitive run. I doubt new Beaker samples will be significantly more Yamnaya shifted, maybe the earliest ones would be closer to Central European Beakers due to still not having mixed with the native farmers.

[Mens-Sarda]

If you think that Celtic and Germanic expanded only during the Iron Age - not with Bell Beakers in the Bronze Age - then I'm sorry but we don't have any Indo-European language spoken in Western Europe today that can be dated further back than the Iron Age.

So what evidence you have that Early Bell Beakers in Iberia and Britain spoke some Indo-European language?

Which Indo-European language did they speak? We don't even know its name.

Maybe they indeed spoke Vasconic: https://upload.wikimedia.org/wikipedia/commons/thumb/9/92/Proposed_area_of_Vasconic_languages.png/600px-Proposed_area_of_Vasconic_languages.png

https://upload.wikimedia.org/wikipedia/commons/thumb/9/92/Proposed_area_of_Vasconic_languages.png/600px-Proposed_area_of_Vasconic_languages.png



IIRC, David Reich recently supports origin of R1b-M269 in the Middle East (Armenian Plateau, Caucasus or Iran). Basque and other Non-IE languages spoken in Western Europe until recently (such as Etruscan, Pelasgian, Rhaetian, etc.) could originate there atfer the Neolithic period. Areas like Gascony, Catalonia, Tuscany, Alpine part of Lombardy, South Tyrol etc. were Non-IE speaking until historical times.

Traces of some Basque-like substratum are present also in Sardinia and Corsica. In Sardinia there are tons of toponyms, almost 40% in the inner areas of the island that are not in Latin or in actual Sardinian, but they are non-IE toponyms that make sense if translated using Basque or Proto-Basque.

P.S.
About the non-IE languages spoken in Great Britain, from the inscriptions found in Scotland has been proved that Picts spoke a non-IE language, half of the inscriptions are in an unintelligible language, and the other half in a Celtic language that they probably acquired when the Celts arrived on the island, the Picts eventually started to use Celt as second language to communicate with Celts; until their ancestral language disappeared and Celtic remained the only language spoken in the area.

[Aren]

Tell me which Indo-European languages did Early Bell Beakers in Britain and Iberia speak? Go on. You personally claimed that Celtic emerged in Hallstatt, Germanic in Jastorf - so what do we know about languages spoken by Early Bronze Age Beakers?

Steppe auDNA isn't a name of a language, as far as I know.

Also a lot of people are claiming that R1b-M269 actually originated in Iran or south of the Caucasus.

How do you explain high % of R1b in Assyrians, whose language is closer to Etruscan than to Celtic.

I used to think that before knowing of the second more northern shifted Hallstatt sample also being P312. I'm more in line with a pre-proto-Italo-Celtic being the language of all BB now, with some peripheral areas in southern Europe not as much affected by the Steppe-BB to be non-IE speaking but autosomally similar to modern day Basques. I think the new paper on Iberia is gonna show this. Also Jastorf culture was influenced by Hallstatt wasn't it? And U106 seems to have somewhat of a seperate and so far unknown origin to it's sisterclade P312. I'm not sure of were it originated. I still favor Western CW(like the oldest sample we have now from Sweden) or possibly Danish-Norwegian BB. It would make sense considering it's TMRCA of 4700 ybp and U106* being found so far today in a Swede and a Scot.

Earliest R1b-L51 if I'm not misstaken is from Saxony-Anhalt, not from the non-Steppe Iberian Bell Beakers. And the British BB are dated younger than that and clearly atusomally much more related to the BB from Central Europe, even carrying more Steppe auDNA.
I'm just going by the general consensus scholars have atm that Yamnaya was proto-IE(whether pre-proto IE was located south or north of the Caucasus, Yamnaya is still regarded as proto-IE). And as we all know Yamnaya remains today, all of them from the Eastern part of the European Steppe, are R1b-M269 > Z103 so not ancestral to modern Western European L51, but a sisterclade of it. I can't imagine L51 being non-IE with this in mid. Assimilated into the proto-Basque/Vasconic societies sure, but not ancestral to them.

I don't think M269 originated in Iran, but it's not 100% clear. Still the M269 Copper Age sample dated to 6000BC from Hajji Firuz(NW Iran) hasn't been carbon dated yet. But I suspect it will more accurately be dated to the Bronze Age period as some of the other remains found in that cemetery. Either way Z103 was from assimilated IE-speakers in our case and other Near Eastern populations. We already have Steppe-admixed remains from the EBA in Armenia and NW Iran, and we haven't even gotten a single aDNA from Mesopotamia yet. I think we'll be getting a lot of surprises once we get more aDNA from Western Asia.

[Aren]

Suppositions. Even though some population movement might have occured, i doubt they had any significant influence on the local genepool, most of the times archeology and linguistics corresponds pretty well to genomics. We know that more than half of Iberia was Celtic speaking before Romans made it there, so you should expect a lot of Celtic admixture, not much less than in the British Isles, even though it is obvious that Celtic material culture and language spread mostly by acculturation of natives.
What is Celtic admixture though? We see a very clear continuation in Britain from the BA to the Roman era, with no pull towards the two Hallstatt samples we have so far. So was Britain unanimously Celticized without any signs of auDNA but not the same for Iberia? Correct me if I'm wrong but Hallstatt/La Tene materials are found in both Britain and Iberia so material culture doesn't always correlate with autosomal admixture and big population movements. I think Celts were either arleady present in Iberia and Britain at that time or it was spread without any great migrations to both Britain and Iberia.


Yes, who said otherwise? Northern input via Romans would only make sense if Romans were more Northern European-like than native Iberians, and that is counterintuitive. If Romans were as much Beaker-like as Iberians or close to that, the deep ancestry (ENF-WHG-Steppe ratio) of Roman era samples would not be much different from the earlier populations, and only the eastern input to the exclusion of previous populations would be suggestive of Roman admixture (talking strictly about ADMIXTURE).

The new paper coming out soon regarding prehistoric Iberia mentioned a 40% Yamnaya input in the new BB samples they have from Spain. I'm not surprised if that would be true considering that we have two Beall Beaker samples from Southern France that are 45+% Steppe.


The algorithm is compensating the excess of Iberia_CA with more 'Anglo-Saxon' input, Basques are still very close to both Iberia_BA and Beaker_Iberia. Focusing on better fits isn't always ideal when using averages from a so limited data set, a lot of weird percentages can pop up due to the algorithm trying to find a better fit, like the Irish coming up as ~80% Anglo-Saxon in Grace's model to the complete exclusion of Beaker_Britain. Sometimes you should sacrifice statistical distance for a more intuitive run. I doubt new Beaker samples will be significantly more Yamnaya shifted, maybe the earliest ones would be closer to Central European Beakers due to still not having mixed with the native farmers.
I agree a lower fit is not always better, but your runs are doing the same. You are using Anatolia_BA, a kind of southern admix that in no way could've reached Iberia unmixed. In that sense nMonte is giving more Anglo-Saxon and Hallstatt to compensate for it. If we keep getting only 25-30% Steppe Beakers and other BA samples from Iberia then yes, I would agree that there's a northern admix in modern Iberians not explained by the Beakers. I don't think it's Germanic though.
There's no sign of that high amount of Germanic admix in Iberia when looking at uniparental markers. From the new Gothic samples we have from Poland it seems they were mainly I1 which is below 3% almost everywhere in Spain. Another marker for other medieval Germanic genomes we have so far is U106, again around the same amount of % as I1. There's no arguing that these markers have decreased in time because of bottlenecking or whatnot considering Iberia has been one of the most populated region in Europe since that time.

[Token]

What is Celtic admixture though? We see a very clear continuation in Britain from the BA to the Roman era, with no pull towards the two Hallstatt samples we have so far. So was Britain unanimously Celticized without any signs of auDNA but not the same for Iberia? Correct me if I'm wrong but Hallstatt/La Tene materials are found in both Britain and Iberia so material culture doesn't always correlate with autosomal admixture and big population movements. I think Celts were either arleady present in Iberia and Britain at that time or it was spread without any great migrations to both Britain and Iberia.
There is a strong pull towards Hallstatt in Iron Age England. This one is a qpAdm analysis ran by Davidski.

England_IA
Britain_&_Ireland_BA 0.555±0.172
Hallstatt 0.445±0.172
chisq 18.513
tail prob 0.100973


The new paper coming out soon regarding prehistoric Iberia mentioned a 40% Yamnaya input in the new BB samples they have from Spain. I'm not surprised if that would be true considering that we have two Beall Beaker samples from Southern France that are 45+% Steppe.
You interpreted it wrongly, the new paper states that Bronze Age populations derived 40% of their ancestry to the incoming Beakers, not directly from Yamnaya - Yamnayans never made it to Spain. This is consistent with our avaiable Bronze Age Iberian being around ~20% Yamnaya. According to this new abstract, i'm being generous in using Beaker_Iberia as a proxy for pre IA Iberia, they might have been even less steppe than that.

'... the arrival of individuals with steppe-related ancestry had a rapid and widespread genetic impact, with Bronze Age populations deriving ~40% of their autosomal ancestry and 100% of their Y-chromosomes from these migrants.'


I agree a lower fit is not always better, but your runs are doing the same. You are using Anatolia_BA, a kind of southern admix that in no way could've reached Iberia unmixed. In that sense nMonte is giving more Anglo-Saxon and Hallstatt to compensate for it. If we keep getting only 25-30% Steppe Beakers and other BA samples from Iberia then yes, I would agree that there's a northern admix in modern Iberians not explained by the Beakers. I don't think it's Germanic though.
There's no sign of that high amount of Germanic admix in Iberia when looking at uniparental markers. From the new Gothic samples we have from Poland it seems they were mainly I1 which is below 3% almost everywhere in Spain. Another marker for other medieval Germanic genomes we have so far is U106, again around the same amount of % as I1. There's no arguing that these markers have decreased in time because of bottlenecking or whatnot considering Iberia has been one of the most populated region in Europe since that time.
Germanic admixture might not be that substantial outside Northwestern Spain, where there is a comparatively higher percentage of U106 and I1, but haplogroup frequencies doesn't always tells the entire story. By looking at frequencies in Ireland for example, you'd think that they didn't received much Germanic admixture over time, but some parts of it have as high as 20% Scandinavian admixture, not counting the Anglo-Saxon contributions.

[Aren]

There is a strong pull towards Hallstatt in Iron Age England. This one is a qpAdm analysis ran by Davidski.

England_IA
Britain_&_Ireland_BA 0.555±0.172
Hallstatt 0.445±0.172
chisq 18.513
tail prob 0.100973
That's not a very good fit though. Either way his recent genetic drift sensitive PCA shows modern Irish and Welsh well within the variation of the earlier BA samples from Britain and Ireland. They have no particular shift in comparison with these 4000 year old genomes.
https://2.bp.blogspot.com/-optx810q0Ww/W6wo_3TtzTI/AAAAAAAAHP0/SBpyoC044r8yEr7xR_E7XidiMxUhIRABACLcBGAs/s1600/Prehistoric_Britain_%2526_Ireland_PCA.png


You interpreted it wrongly, the new paper states that Bronze Age populations derived 40% of their ancestry to the incoming Beakers, not directly from Yamnaya - Yamnayans never made it to Spain. This is consistent with our avaiable Bronze Age Iberian being around ~20% Yamnaya. According to this new abstract, i'm being generous in using Beaker_Iberia as a proxy for pre IA Iberia, they might have been even less steppe than that.

First of the Iberian Beakers are quite diverse. We get this low 20% number cause of some of the Beakers were not Steppe derived and differed not only autosomally but also Y-DNA(ofc) and by the material they were buried in. I think it's more accurate to distinct these culturally only Beaker types from the Steppe derived ones probably originating in East-Central Europe. One of the six Steppe Iberian Beakers is almost 35% Yamnaya, higher than modern Iberians. Add to that in Southern France of the three Beakers, one is 40% Steppe the other is 45% and the third basically no Steppe.

"distance%=2.8103"

Beaker_Southern_France:I3875

Globular_Amphora,47.4
Yamnaya_Samara,45.6
France_MLN,7

"distance%=4.0694"

Beaker_Iberia:I6623

Globular_Amphora,62.8
Yamnaya_Samara,34
Iberia_Central_CA,3.2

Individuals with Steppe related ancestry doesn't necesarrily mean the Beakers we think of. Either way some of the Beakers from Northern Europe were over 60% Steppe autosomally speaking. I think I saw a vid of some of the authors of this new paper describing the new Iberian Beakers as 40% Yamnaya admixed but maybe they meant Steppe_MLBA and not Yamnaya.

Germanic admixture might not be that substantial outside Northwestern Spain, where there is a comparatively higher percentage of U106 and I1, but haplogroup frequencies doesn't always tells the entire story. By looking at frequencies in Ireland for example, you'd think that they didn't received much Germanic admixture over time, but some parts of it have as high as 20% Scandinavian admixture, not counting the Anglo-Saxon contributions.
Only very slightly higher, but any kind of specific regional genetic profile would be almost gone by know due to internal migrations and mixing in Iberia. I disagree with the notion that there's substantial amount of Germanic admixture anywhere in Iberia.

And regarding Irish, they have very little Scandinavian or Germanic admixture, as shown by the PCA above they are not drifting towards the Germanic samples.

[Nurzat]

my result, though probably not relevant without a proxy for the Slavs.

distance 3.1828

59.17% - Beaker Central Europe
29.17% - Hallstatt Bylany
11.67% - Anatolia BA
0% - England Anglo-Saxon


then modeling myself only with Hallstatt contemporary cultures (Iron Age) - Balkans Iron Age took almost half of the Hallstatt score:

distance 2.6525

61.67% Scythian Ukraine
20% Balkans_IA
16.67% Hallstatt Bylany
1.67% England_IA

[Peterski]

I couldn't find any Iron Age from Sweden

I think it is called "Nordic IA" in Global25 spreadsheet.

[Jana]

Aren writing bullshit again, he is quick to minimalize Celto-Germanic impact in Iberia and especially Scandinavian in Ireland, but quick to inflate North African and Roman, or to claim English are ''very Germanic'' genetically. :picard2:
Biased user with heavy agenda, despite he is quite knowledgable. What a shame.

[Grace O'Malley]

Just using your model Eskimo because I don't know what to use. It doesn't look a great model in that combo for Irish. Anthrogenica has been down for a while so I can't go on there to look at more realistic models.

Fit 1.5138 (Myself)

Anatolia_BA 0
Beaker_Central_Europe 42.5
England_Anglo-Saxon 57.5
Hallstatt Bylany 0

Anatolia BA 0
Beaker_Central_Europe 28.33
England_Anglo-Saxon 70.83
Hallstatt_Bylany 0.83

[Peterski]

my result, though probably not relevant without a proxy for the Slavs.

distance 3.1828

59.17% - Beaker Central Europe
29.17% - Hallstatt Bylany
11.67% - Anatolia BA
0% - England Anglo-Saxon


then modeling myself only with Hallstatt contemporary cultures (Iron Age) - Balkans Iron Age took almost half of the Hallstatt score:

distance 2.6525

61.67% Scythian Ukraine
20% Balkans_IA
16.67% Hallstatt Bylany
1.67% England_IA

Bad fits, large distances.

[Grace O'Malley]

Aren writing bullshit again, he is quick to minimalize Celto-Germanic impact in Iberia and especially Scandinavian in Ireland, but quick to inflate North African and Roman, or to claim English are ''very Germanic'' genetically. :picard2:
Biased user with heavy agenda, despite he is quite knowledgable. What a shame.

Aren's a nice guy and I don't want to criticise him but the Celtic / Germanic map has a limited use which Davidski has said himself. He said it is particularly sensitive to Irish and British drift and that the Global 25 is actually more informative. It has a limited use for a specific purpose. Discussion is interesting but I will still add more weight to what Geneticists that are qualified and working in the field say and who write peer reviewed papers.

I agree Aren is very knowledgeable but I don't agree with everything he says. :) I do appreciate him being on this forum though.

[Grace O'Malley]

Where can I download the most recent version of Global25 spreadsheet?

I'd presume he just keeps updating it here if not I'm sure some of the guys that use it will let you know.

http://eurogenes.blogspot.com/2018/02/unleash-power-global-25-test-drive.html

[Peterski]

Here is what I get for the English, run with 40 ancient populations (Poprad is a Germanic chieftain from Slovakia):

[1] "distance%=0.907 / distance=0.00907"

English

England_Roman 48.0
England_IA 17.2
Wales_CA_EBA 15.7
Nordic_IA 12.7
Iberia_N 4.9
Balkans_IA 1.4
Ireland_EBA 0.1
Anatolia_IA 0.0
Balkans_BA 0.0
Baltic_BA 0.0
Baltic_IA 0.0
Beaker_Britain 0.0
England_CA_EBA 0.0
England_LBA 0.0
England_MBA 0.0
England_N 0.0
Germany_Roman 0.0
Hallstatt_Bylany 0.0
Iberia_BA 0.0
Iberia_Central_CA 0.0
Iberia_ChL 0.0
Iberia_MN 0.0
Iberia_Southwest_CA 0.0
Ireland_MN 0.0
Ostrogothic_Crimea_ACD 0.0
Poland_BA 0.0
Poprad 0.0
Portugal_LNCA 0.0
Portugal_MBA 0.0
Portugal_MN 0.0
Remedello_BA 0.0
Scotland_CA_EBA 0.0
Scotland_LBA 0.0
Scotland_MBA 0.0
Scotland_N 0.0
Scythian_Hungary 0.0
Slavic_Bohemia 0.0
Spain_LNCA 0.0
Unetice 0.0
Wales_N 0.0

[1] "distance%=1.0148 / distance=0.010148"

English_Cornwall

England_Roman 52.80
Wales_CA_EBA 17.30
Nordic_IA 10.60
Poprad 10.15
Iberia_N 5.65
Iberia_BA 2.45
Ireland_EBA 1.05
Anatolia_IA 0.00
Balkans_BA 0.00
Balkans_IA 0.00
Baltic_BA 0.00
Baltic_IA 0.00
Beaker_Britain 0.00
England_CA_EBA 0.00
England_IA 0.00
England_LBA 0.00
England_MBA 0.00
England_N 0.00
Germany_Roman 0.00
Hallstatt_Bylany 0.00
Iberia_Central_CA 0.00
Iberia_ChL 0.00
Iberia_MN 0.00
Iberia_Southwest_CA 0.00
Ireland_MN 0.00
Ostrogothic_Crimea_ACD 0.00
Poland_BA 0.00
Portugal_LNCA 0.00
Portugal_MBA 0.00
Portugal_MN 0.00
Remedello_BA 0.00
Scotland_CA_EBA 0.00
Scotland_LBA 0.00
Scotland_MBA 0.00
Scotland_N 0.00
Scythian_Hungary 0.00
Slavic_Bohemia 0.00
Spain_LNCA 0.00
Unetice 0.00
Wales_N 0.00

[XYZ.2018]

Here is what I get for the English, run with 40 ancient populations (Poprad is a Germanic chieftain from Slovakia):

[1] "distance%=0.907 / distance=0.00907"

English

England_Roman 48.0
England_IA 17.2
Wales_CA_EBA 15.7
Nordic_IA 12.7
Iberia_N 4.9
Balkans_IA 1.4
Ireland_EBA 0.1
Anatolia_IA 0.0
Balkans_BA 0.0
Baltic_BA 0.0
Baltic_IA 0.0
Beaker_Britain 0.0
England_CA_EBA 0.0
England_LBA 0.0
England_MBA 0.0
England_N 0.0
Germany_Roman 0.0
Hallstatt_Bylany 0.0
Iberia_BA 0.0
Iberia_Central_CA 0.0
Iberia_ChL 0.0
Iberia_MN 0.0
Iberia_Southwest_CA 0.0
Ireland_MN 0.0
Ostrogothic_Crimea_ACD 0.0
Poland_BA 0.0
Poprad 0.0
Portugal_LNCA 0.0
Portugal_MBA 0.0
Portugal_MN 0.0
Remedello_BA 0.0
Scotland_CA_EBA 0.0
Scotland_LBA 0.0
Scotland_MBA 0.0
Scotland_N 0.0
Scythian_Hungary 0.0
Slavic_Bohemia 0.0
Spain_LNCA 0.0
Unetice 0.0
Wales_N 0.0

[1] "distance%=1.0148 / distance=0.010148"

English_Cornwall

England_Roman 52.80
Wales_CA_EBA 17.30
Nordic_IA 10.60
Poprad 10.15
Iberia_N 5.65
Iberia_BA 2.45
Ireland_EBA 1.05
Anatolia_IA 0.00
Balkans_BA 0.00
Balkans_IA 0.00
Baltic_BA 0.00
Baltic_IA 0.00
Beaker_Britain 0.00
England_CA_EBA 0.00
England_IA 0.00
England_LBA 0.00
England_MBA 0.00
England_N 0.00
Germany_Roman 0.00
Hallstatt_Bylany 0.00
Iberia_Central_CA 0.00
Iberia_ChL 0.00
Iberia_MN 0.00
Iberia_Southwest_CA 0.00
Ireland_MN 0.00
Ostrogothic_Crimea_ACD 0.00
Poland_BA 0.00
Portugal_LNCA 0.00
Portugal_MBA 0.00
Portugal_MN 0.00
Remedello_BA 0.00
Scotland_CA_EBA 0.00
Scotland_LBA 0.00
Scotland_MBA 0.00
Scotland_N 0.00
Scythian_Hungary 0.00
Slavic_Bohemia 0.00
Spain_LNCA 0.00
Unetice 0.00
Wales_N 0.00

Why those and why Neolithic mixed with Medieval?

[Peterski]

Why those and why Neolithic mixed with Medieval?

If you want an unbiased run, include many populations and let the algorithm choose.

Neolithic - I knew that the algorithm wasn't going to pick it up anyway, so why not?

It picked up Iberian Neolithic. Now I will remove it and see if other Iberian replaces it.

I will also remove England_Roman.

[Grace O'Malley]

If you want an unbiased run, include many populations and let the algorithm choose.

Neolithic - I knew that the algorithm wasn't going to pick it up anyway, so why not?

It picked up Iberian Neolithic. Now I will remove it and see if other Iberian replaces it.

I will also remove England_Roman.

Can you do Ireland to compare? Also Spain?

Thanks

[Peterski]

I wasn't using the most recent version of the spreadsheet for that run.

Now I downloaded the newest version and will take ca. 50 populations.

[Peterski]

Here is a new model with more Germanic populations included:

Modern ENGLISH:

[1] "1. CLOSEST SINGLE ITEM DISTANCES"
Germany_Medieval Scotland_LBA England_MBA
0.02063324 0.02086716 0.02458515
Poprad_Medieval England_IA Italy_Medieval_Collegno
0.02658966 0.02699060 0.02760391
Hungary_Medieval_Szolad Scotland_MBA
0.02880014 0.03032941

[1] "distance%=0.943 / distance=0.00943"

English

Italy_Medieval_Collegno 30.3
Nordic_IA 16.4
England_IA 15.4
Scotland_LBA 12.8
Wales_CA_EBA 10.6
Ireland_EBA 6.2
Iberia_BA 5.1
England_MBA 2.2
Iberia_Central_CA 0.8
Altai_IA 0.0
Anatolia_BA 0.0
Anatolia_EBA 0.0
Anatolia_IA 0.0
Anatolia_MLBA 0.0
Anatolia_Ottoman 0.0
Armenia_EBA 0.0
Armenia_MLBA 0.0
Avar_Hungary_Szolad 0.0
Balkans_BA 0.0
Balkans_IA 0.0
Baltic_BA 0.0
Baltic_IA 0.0
Beaker_Britain 0.0
Cimmerian_Moldova 0.0
England_CA_EBA 0.0
England_LBA 0.0
England_N 0.0
Finn_ancient 0.0
Gepid_Serbia_ACD 0.0
Germany_Medieval 0.0
Germany_Roman 0.0
Golden_Horde_Asian 0.0
Golden_Horde_European 0.0
Hallstatt_Bylany 0.0
Hun-Sarmatian 0.0
Hun_Kazakh_Steppe 0.0
Hun_Tian_Shan 0.0
Hungary_IA 0.0
Hungary_Medieval 0.0
Hungary_Medieval_Szolad 0.0
Iberia_ChL 0.0
Iberia_Southwest_CA 0.0
Iran_recent 0.0
Ireland_MN 0.0
Italian_Tuscan 0.0
Minoan_Lasithi 0.0
Mycenaean 0.0
Netherlands_BA 0.0
Nomad_IA 0.0
Nomad_Medieval 0.0
Ostrogothic_Crimea_ACD 0.0
Poland_BA 0.0
Poprad_Medieval 0.0
Remedello_BA 0.0
Saami_ancient 0.0
Sarmatian 0.0
Sarmatian_Pokrovka 0.0
Sarmatian_Urals 0.0
Sarmatian_West 0.0
Scotland_CA_EBA 0.0
Scotland_MBA 0.0
Scotland_N 0.0
Scythian_Hungary 0.0
Scythian_Moldova 0.0
Scythian_Samara 0.0
Scythian_Ukraine 0.0
Slavic_Bohemia 0.0
Spain_LNCA 0.0
Sweden_Viking_Age 0.0
Turk_Medieval 0.0
Unetice 0.0
Wales_N 0.0

Modern CORNISH:

[1] "1. CLOSEST SINGLE ITEM DISTANCES"
Germany_Medieval Scotland_LBA Italy_Medieval_Collegno
0.02272190 0.02295370 0.02694174
Poprad_Medieval England_MBA Hungary_Medieval_Szolad
0.02750614 0.02752038 0.02915247
England_IA Scotland_MBA
0.03043150 0.03290277

[1] "distance%=1.0867 / distance=0.010867"

English_Cornwall

Italy_Medieval_Collegno 26.90
Nordic_IA 15.75
Wales_CA_EBA 15.00
Scotland_LBA 12.50
Iberia_BA 10.85
Poprad_Medieval 8.15
Ireland_EBA 5.50
England_MBA 5.35
Altai_IA 0.00
Anatolia_BA 0.00
Anatolia_EBA 0.00
Anatolia_IA 0.00
Anatolia_MLBA 0.00
Anatolia_Ottoman 0.00
Armenia_EBA 0.00
Armenia_MLBA 0.00
Avar_Hungary_Szolad 0.00
Balkans_BA 0.00
Balkans_IA 0.00
Baltic_BA 0.00
Baltic_IA 0.00
Beaker_Britain 0.00
Cimmerian_Moldova 0.00
England_CA_EBA 0.00
England_IA 0.00
England_LBA 0.00
England_N 0.00
Finn_ancient 0.00
Gepid_Serbia_ACD 0.00
Germany_Medieval 0.00
Germany_Roman 0.00
Golden_Horde_Asian 0.00
Golden_Horde_European 0.00
Hallstatt_Bylany 0.00
Hun-Sarmatian 0.00
Hun_Kazakh_Steppe 0.00
Hun_Tian_Shan 0.00
Hungary_IA 0.00
Hungary_Medieval 0.00
Hungary_Medieval_Szolad 0.00
Iberia_Central_CA 0.00
Iberia_ChL 0.00
Iberia_Southwest_CA 0.00
Iran_recent 0.00
Ireland_MN 0.00
Italian_Tuscan 0.00
Minoan_Lasithi 0.00
Mycenaean 0.00
Netherlands_BA 0.00
Nomad_IA 0.00
Nomad_Medieval 0.00
Ostrogothic_Crimea_ACD 0.00
Poland_BA 0.00
Remedello_BA 0.00
Saami_ancient 0.00
Sarmatian 0.00
Sarmatian_Pokrovka 0.00
Sarmatian_Urals 0.00
Sarmatian_West 0.00
Scotland_CA_EBA 0.00
Scotland_MBA 0.00
Scotland_N 0.00
Scythian_Hungary 0.00
Scythian_Moldova 0.00
Scythian_Samara 0.00
Scythian_Ukraine 0.00
Slavic_Bohemia 0.00
Spain_LNCA 0.00
Sweden_Viking_Age 0.00
Turk_Medieval 0.00
Unetice 0.00
Wales_N 0.00

Modern AUSTRIANS:

[1] "1. CLOSEST SINGLE ITEM DISTANCES"
Hungary_Medieval_Szolad Scythian_Hungary Italy_Medieval_Collegno
0.02219305 0.02739925 0.02911182
Poprad_Medieval Germany_Medieval Hallstatt_Bylany
0.03278360 0.03507051 0.03555555
Scythian_Ukraine England_IA
0.03607362 0.03870974

[1] "distance%=0.5128 / distance=0.005128"

Austrian

England_IA 21.05
Avar_Hungary_Szolad 19.35
Nordic_IA 11.65
Minoan_Lasithi 7.15
Baltic_IA 6.20
Anatolia_BA 5.85
Hungary_Medieval 5.60
Hallstatt_Bylany 5.25
Poprad_Medieval 3.95
Balkans_IA 3.85
Germany_Medieval 2.90
Netherlands_BA 2.30
Golden_Horde_European 2.05
Wales_CA_EBA 1.60
Baltic_BA 0.75
Hungary_IA 0.50
Altai_IA 0.00
Anatolia_EBA 0.00
Anatolia_IA 0.00
Anatolia_MLBA 0.00
Anatolia_Ottoman 0.00
Armenia_EBA 0.00
Armenia_MLBA 0.00
Balkans_BA 0.00
Beaker_Britain 0.00
Cimmerian_Moldova 0.00
England_CA_EBA 0.00
England_LBA 0.00
England_MBA 0.00
England_N 0.00
Finn_ancient 0.00
Gepid_Serbia_ACD 0.00
Germany_Roman 0.00
Golden_Horde_Asian 0.00
Hun-Sarmatian 0.00
Hun_Kazakh_Steppe 0.00
Hun_Tian_Shan 0.00
Hungary_Medieval_Szolad 0.00
Iberia_BA 0.00
Iberia_Central_CA 0.00
Iberia_ChL 0.00
Iberia_Southwest_CA 0.00
Iran_recent 0.00
Ireland_EBA 0.00
Ireland_MN 0.00
Italian_Tuscan 0.00
Italy_Medieval_Collegno 0.00
Mycenaean 0.00
Nomad_IA 0.00
Nomad_Medieval 0.00
Ostrogothic_Crimea_ACD 0.00
Poland_BA 0.00
Remedello_BA 0.00
Saami_ancient 0.00
Sarmatian 0.00
Sarmatian_Pokrovka 0.00
Sarmatian_Urals 0.00
Sarmatian_West 0.00
Scotland_CA_EBA 0.00
Scotland_LBA 0.00
Scotland_MBA 0.00
Scotland_N 0.00
Scythian_Hungary 0.00
Scythian_Moldova 0.00
Scythian_Samara 0.00
Scythian_Ukraine 0.00
Slavic_Bohemia 0.00
Spain_LNCA 0.00
Sweden_Viking_Age 0.00
Turk_Medieval 0.00
Unetice 0.00
Wales_N 0.00

[Peterski]

I think I cannot use Italy_Medieval_Collegno, they were probably mixed with local Celts (Cisalpine Gauls) and Romans already.

Will remove Lombards (Collegno) and Minoans (too old) and try again.

[Token]

That's not a very good fit though. Either way his recent genetic drift sensitive PCA shows modern Irish and Welsh well within the variation of the earlier BA samples from Britain and Ireland. They have no particular shift in comparison with these 4000 year old genomes.
https://2.bp.blogspot.com/-optx810q0Ww/W6wo_3TtzTI/AAAAAAAAHP0/SBpyoC044r8yEr7xR_E7XidiMxUhIRABACLcBGAs/s1600/Prehistoric_Britain_%2526_Ireland_PCA.png
This PCA is screwed by ethnic-specific genetic drift. Look where Icelanders plot for example, you'd think they are more Germanic than Norwegians based on their proximity to Nordic_IA, but when you exclude recent genetic drift (or replicate this in a tri-dimensional PCA), they are actually somewhere in between Norwegians and Irish. Davidski made it for a very specific purpose: distinguishing modern-day Germanic and Celtic-speakers. Ireland, as a insular and until recently, a quite isolated population, did experienced extensive drift, specially after the drastic population reduction during the Great Famine.



First of the Iberian Beakers are quite diverse. We get this low 20% number cause of some of the Beakers were not Steppe derived and differed not only autosomally but also Y-DNA(ofc) and by the material they were buried in. I think it's more accurate to distinct these culturally only Beaker types from the Steppe derived ones probably originating in East-Central Europe.One of the six Steppe Iberian Beakers is almost 35% Yamnaya, higher than modern Iberians. Add to that in Southern France of the three Beakers, one is 40% Steppe the other is 45% and the third basically no Steppe.

"distance%=2.8103"

Beaker_Southern_France:I3875

Globular_Amphora,47.4
Yamnaya_Samara,45.6
France_MLN,7

"distance%=4.0694"

Beaker_Iberia:I6623

Globular_Amphora,62.8
Yamnaya_Samara,34
Iberia_Central_CA,3.2

Again wrong, Beaker_Iberia average includes only individuals with steppe ancestry. Exceptions are what they are - exceptions, one of the Beaker_Hungary samples is almost fully Yamnaya, should we use it as premise to suggest Beakers were initially almost fully Yamnaya, even though it is pretty obvious that the material culture originated in Central Europe, in the GAC-Yamnaya contact zone? Should we suggest that Bronze Age Hungarians were more steppe than Bronze Age Balts?


Individuals with Steppe related ancestry doesn't necesarrily mean the Beakers we think of. Either way some of the Beakers from Northern Europe were over 60% Steppe autosomally speaking. I think I saw a vid of some of the authors of this new paper describing the new Iberian Beakers as 40% Yamnaya admixed but maybe they meant Steppe_MLBA and not Yamnaya.
During the early Bronze Age, Beakers were the only collective in Western Europe with substantial steppe ancestry. In this particular case, Beaker and 'individuals with steppe ancestry' is interchangeable.


Only very slightly higher, but any kind of specific regional genetic profile would be almost gone by know due to internal migrations and mixing in Iberia. I disagree with the notion that there's substantial amount of Germanic admixture anywhere in Iberia.

And regarding Irish, they have very little Scandinavian or Germanic admixture, as shown by the PCA above they are not drifting towards the Germanic samples.
Irish did received substantial Germanic admixture as shown by recent peer-reviewed genetic studies, and the estimated dates of these events (using the still dubious GLOBETROTTER, but whatever) corresponds exactly to the Norse Viking and Anglo-Norman migrations. You can't overlook a entire study based on a screwed PCA interpretation.

[Peterski]

These models are with 70 ancient populations used (just like above minus Minoans and Collegno):

[1] "distance%=1.0544 / distance=0.010544"

English

Scotland_LBA 19.50
Germany_Medieval 17.35
Wales_CA_EBA 16.40
England_IA 15.25
Nordic_IA 12.90
Iberia_BA 8.40
Balkans_IA 4.70
England_MBA 2.55
Ireland_EBA 2.45
Iberia_Central_CA 0.50
Altai_IA 0.00
Anatolia_BA 0.00
Anatolia_EBA 0.00
Anatolia_IA 0.00
Anatolia_MLBA 0.00
Anatolia_Ottoman 0.00
Armenia_EBA 0.00
Armenia_MLBA 0.00
Avar_Hungary_Szolad 0.00
Balkans_BA 0.00
Baltic_BA 0.00
Baltic_IA 0.00
Beaker_Britain 0.00
Cimmerian_Moldova 0.00
England_CA_EBA 0.00
England_LBA 0.00
England_N 0.00
Finn_ancient 0.00
Gepid_Serbia_ACD 0.00
Germany_Roman 0.00
Golden_Horde_Asian 0.00
Golden_Horde_European 0.00
Hallstatt_Bylany 0.00
Hun-Sarmatian 0.00
Hun_Kazakh_Steppe 0.00
Hun_Tian_Shan 0.00
Hungary_IA 0.00
Hungary_Medieval 0.00
Hungary_Medieval_Szolad 0.00
Iberia_ChL 0.00
Iberia_Southwest_CA 0.00
Iran_recent 0.00
Ireland_MN 0.00
Italian_Tuscan 0.00
Mycenaean 0.00
Netherlands_BA 0.00
Nomad_IA 0.00
Nomad_Medieval 0.00
Ostrogothic_Crimea_ACD 0.00
Poland_BA 0.00
Poprad_Medieval 0.00
Remedello_BA 0.00
Saami_ancient 0.00
Sarmatian 0.00
Sarmatian_Pokrovka 0.00
Sarmatian_Urals 0.00
Sarmatian_West 0.00
Scotland_CA_EBA 0.00
Scotland_MBA 0.00
Scotland_N 0.00
Scythian_Hungary 0.00
Scythian_Moldova 0.00
Scythian_Samara 0.00
Scythian_Ukraine 0.00
Slavic_Bohemia 0.00
Spain_LNCA 0.00
Sweden_Viking_Age 0.00
Turk_Medieval 0.00
Unetice 0.00
Wales_N 0.00
[1] "distance%=1.1793 / distance=0.011793"

English_Cornwall

Scotland_LBA 21.65
Wales_CA_EBA 18.85
Germany_Medieval 15.70
Iberia_BA 13.65
Hungary_Medieval_Szolad 11.90
Nordic_IA 9.75
Poprad_Medieval 8.30
Ireland_EBA 0.20
Altai_IA 0.00
Anatolia_BA 0.00
Anatolia_EBA 0.00
Anatolia_IA 0.00
Anatolia_MLBA 0.00
Anatolia_Ottoman 0.00
Armenia_EBA 0.00
Armenia_MLBA 0.00
Avar_Hungary_Szolad 0.00
Balkans_BA 0.00
Balkans_IA 0.00
Baltic_BA 0.00
Baltic_IA 0.00
Beaker_Britain 0.00
Cimmerian_Moldova 0.00
England_CA_EBA 0.00
England_IA 0.00
England_LBA 0.00
England_MBA 0.00
England_N 0.00
Finn_ancient 0.00
Gepid_Serbia_ACD 0.00
Germany_Roman 0.00
Golden_Horde_Asian 0.00
Golden_Horde_European 0.00
Hallstatt_Bylany 0.00
Hun-Sarmatian 0.00
Hun_Kazakh_Steppe 0.00
Hun_Tian_Shan 0.00
Hungary_IA 0.00
Hungary_Medieval 0.00
Iberia_Central_CA 0.00
Iberia_ChL 0.00
Iberia_Southwest_CA 0.00
Iran_recent 0.00
Ireland_MN 0.00
Italian_Tuscan 0.00
Mycenaean 0.00
Netherlands_BA 0.00
Nomad_IA 0.00
Nomad_Medieval 0.00
Ostrogothic_Crimea_ACD 0.00
Poland_BA 0.00
Remedello_BA 0.00
Saami_ancient 0.00
Sarmatian 0.00
Sarmatian_Pokrovka 0.00
Sarmatian_Urals 0.00
Sarmatian_West 0.00
Scotland_CA_EBA 0.00
Scotland_MBA 0.00
Scotland_N 0.00
Scythian_Hungary 0.00
Scythian_Moldova 0.00
Scythian_Samara 0.00
Scythian_Ukraine 0.00
Slavic_Bohemia 0.00
Spain_LNCA 0.00
Sweden_Viking_Age 0.00
Turk_Medieval 0.00
Unetice 0.00
Wales_N 0.00
[1] "distance%=0.53 / distance=0.0053"

Austrian

Avar_Hungary_Szolad 20.05
England_IA 18.50
Nordic_IA 14.10
Balkans_IA 9.90
Anatolia_BA 8.20
Hallstatt_Bylany 7.85
Poprad_Medieval 7.25
Hungary_Medieval 5.30
Baltic_IA 4.55
Golden_Horde_European 3.15
Netherlands_BA 0.65
Hungary_Medieval_Szolad 0.30
Remedello_BA 0.10
Slavic_Bohemia 0.05
Sweden_Viking_Age 0.05
Altai_IA 0.00
Anatolia_EBA 0.00
Anatolia_IA 0.00
Anatolia_MLBA 0.00
Anatolia_Ottoman 0.00
Armenia_EBA 0.00
Armenia_MLBA 0.00
Balkans_BA 0.00
Baltic_BA 0.00
Beaker_Britain 0.00
Cimmerian_Moldova 0.00
England_CA_EBA 0.00
England_LBA 0.00
England_MBA 0.00
England_N 0.00
Finn_ancient 0.00
Gepid_Serbia_ACD 0.00
Germany_Medieval 0.00
Germany_Roman 0.00
Golden_Horde_Asian 0.00
Hun-Sarmatian 0.00
Hun_Kazakh_Steppe 0.00
Hun_Tian_Shan 0.00
Hungary_IA 0.00
Iberia_BA 0.00
Iberia_Central_CA 0.00
Iberia_ChL 0.00
Iberia_Southwest_CA 0.00
Iran_recent 0.00
Ireland_EBA 0.00
Ireland_MN 0.00
Italian_Tuscan 0.00
Mycenaean 0.00
Nomad_IA 0.00
Nomad_Medieval 0.00
Ostrogothic_Crimea_ACD 0.00
Poland_BA 0.00
Saami_ancient 0.00
Sarmatian 0.00
Sarmatian_Pokrovka 0.00
Sarmatian_Urals 0.00
Sarmatian_West 0.00
Scotland_CA_EBA 0.00
Scotland_LBA 0.00
Scotland_MBA 0.00
Scotland_N 0.00
Scythian_Hungary 0.00
Scythian_Moldova 0.00
Scythian_Samara 0.00
Scythian_Ukraine 0.00
Spain_LNCA 0.00
Turk_Medieval 0.00
Unetice 0.00
Wales_CA_EBA 0.00
Wales_N 0.00

[Peterski]

My guess is that Davidski's English average includes many individuals from Northern England.

And that's why Scotland_LBA is preferred over England_IA (which are Belgae from the south).

[Gründig]

What is _LBA, _CA_EBA and _MBA?

[Peterski]

What is _LBA, _CA_EBA and _MBA?

LBA = Late Bronze Age
CA = Copper Age
EBA = Early Bronze Age
MBA = Middle Bronze Age

=====

Now the same model for modern Irish people. Single Distances (modern Irish are closest to Scotland_LBA):

[1] "1. CLOSEST SINGLE ITEM DISTANCES"
Scotland_LBA England_MBA Germany_Medieval Scotland_MBA
0.01638603 0.02010704 0.02375932 0.02561753
England_IA England_CA_EBA Wales_CA_EBA Ireland_EBA
0.02644739 0.02819667 0.02852785 0.02875479

Mixed Mode (Non-Germanic populations highlighted):

[1] "distance%=1.2363 / distance=0.012363"

Irish

Scotland_LBA 44.65
England_IA 19.25
Wales_CA_EBA 15.90
Nordic_IA 5.60
England_MBA 4.85
Iberia_BA 4.35
Germany_Medieval 2.80
Balkans_IA 1.60
Hungary_Medieval_Szolad 1.00
(...)

[Grace O'Malley]

LBA = Late Bronze Age
CA = Copper Age
EBA = Early Bronze Age
MBA = Middle Bronze Age

=====

Now the same model for modern Irish people. Single Distances (modern Irish are closest to Scotland_LBA):

[1] "1. CLOSEST SINGLE ITEM DISTANCES"
Scotland_LBA England_MBA Germany_Medieval Scotland_MBA
0.01638603 0.02010704 0.02375932 0.02561753
England_IA England_CA_EBA Wales_CA_EBA Ireland_EBA
0.02644739 0.02819667 0.02852785 0.02875479

Mixed Mode (Non-Germanic populations highlighted):

[1] "distance%=1.2363 / distance=0.012363"

Irish

Scotland_LBA 44.65
England_IA 19.25
Wales_CA_EBA 15.90
Nordic_IA 5.60
England_MBA 4.85
Iberia_BA 4.35
Germany_Medieval 2.80
Balkans_IA 1.60
Hungary_Medieval_Szolad 1.00
(...)

Could you do one for Scotland_LBA if it's not too much trouble? I'd be interested to see what comes up for them. Only if you've got the time.

Thanks

[Peterski]

Here is the file I'm using if someone wants to experiment with it. These are scaled averages:

81207

[Grace O'Malley]

Here is the file I'm using if someone wants to experiment with it. These are scaled averages:

81207

I'll have to teach myself to do this one day. I'd waste more time. :)

[Peterski]

Could you do one for Scotland_LBA

Scotland_LBA:

[1] "1. CLOSEST SINGLE ITEM DISTANCES"
England_MBA Scotland_MBA England_CA_EBA Germany_Medieval
0.01600244 0.02191330 0.02233530 0.02447231
Unetice Ireland_EBA Beaker_Britain Wales_CA_EBA
0.02451382 0.02590203 0.02619936 0.02696142

[1] "distance%=1.2229 / distance=0.012229"

Scotland_LBA

England_MBA 29.80
Wales_CA_EBA 21.90
Beaker_Britain 19.95
Nordic_IA 11.95
Iberia_BA 10.20
Netherlands_BA 3.20
Armenia_EBA 1.55
Scotland_MBA 1.40
Hungary_IA 0.05
(...)

[Peterski]

Or if you prefer nMonte3 with pen=0, here it is:

[1] "1. CLOSEST SINGLE ITEM DISTANCE%"
England_MBA Scotland_MBA England_CA_EBA Germany_Medieval
1.600244 2.191330 2.233530 2.447231
Unetice Ireland_EBA Beaker_Britain Wales_CA_EBA
2.451382 2.590203 2.619936 2.696142

[1] "distance%=1.2233"

Scotland_LBA

England_MBA,31.6
Wales_CA_EBA,21.8
Beaker_Britain,18
Nordic_IA,11.8
Iberia_BA,9.2
Netherlands_BA,3.6
Scotland_MBA,2
Armenia_EBA,1.6
Iberia_Southwest_CA,0.4

Interesting, that even Late Bronze Age Scotland shows some "Germanic" affinity (Nordic IA).

But as you can see also Iberian is very consistent in all of these models.

And when you remove Iberian probably fits will be significantly worse.

=====

Of course it is impossible that Nordic IA travelled to LBA Scotland - as they had no time machine.

We can remove Nordic IA, replace it with Nordic BA or CWC/Battle Axe and see what happens then.

Maybe Beaker can be removed too, as it was ancestral to Wales_CA_EBA and England_MBA as well.

[Grace O'Malley]

Or if you prefer nMonte3 with pen=0, here it is:

[1] "1. CLOSEST SINGLE ITEM DISTANCE%"
England_MBA Scotland_MBA England_CA_EBA Germany_Medieval
1.600244 2.191330 2.233530 2.447231
Unetice Ireland_EBA Beaker_Britain Wales_CA_EBA
2.451382 2.590203 2.619936 2.696142

[1] "distance%=1.2233"

Scotland_LBA

England_MBA,31.6
Wales_CA_EBA,21.8
Beaker_Britain,18
Nordic_IA,11.8
Iberia_BA,9.2
Netherlands_BA,3.6
Scotland_MBA,2
Armenia_EBA,1.6
Iberia_Southwest_CA,0.4

Interesting, that even Late Bronze Age Scotland shows some "Germanic" affinity (Nordic IA).

But as you can see also Iberian is very consistent in all of these models.

And when you remove Iberian probably fits will be significantly worse.

=====

Of course it is impossible that Nordic IA travelled to LBA Scotland - as they had no time machine.

We can remove Nordic IA, replace it with Nordic BA or CWC/Battle Axe and see what happens then.

Maybe Beaker can be removed too, as it was ancestral to Wales_CA_EBA and England_MBA as well.

A lot of categories mask other components. This is why you have to choose realistic models so it can be difficult getting a good fit for a population.

[J. Ketch]

Here is the file I'm using if someone wants to experiment with it. These are scaled averages:

81207
What I get from your sheet:

[1] "distance%=1.7105"

Scotland_LBA,46.6
Wales_CA_EBA,28.8
Nordic_IA,18.8
Remedello_BA,3.8
Ireland_MN,2

[Gründig]

What I get from your sheet:

[1] "distance%=1.7105"

Scotland_LBA,46.6
Wales_CA_EBA,28.8
Nordic_IA,18.8
Remedello_BA,3.8
Ireland_MN,2

What exactly is wales_CA_CBA? I now know what the initials stand for but thats it.

[Peterski]

A lot of categories mask other components. This is why you have to choose realistic models so it can be difficult getting a good fit for a population.

I can choose a realistic model but at first you should include many populations and let the algorithm choose.

If the algorithm chooses something unrealistic, remove it and try again.

Anyway in my opinion it chose realistic models on its own even with 70 populations available to choose from.

[XYZ.2018]

A lot of categories mask other components. This is why you have to choose realistic models so it can be difficult getting a good fit for a population.

And this is why nMonte author Huijbregts recommends models with few best fitted pops and at least one outlier like Nganassan for example.

[Peterski]

What exactly is wales_CA_CBA? I now know what the initials stand for but thats it.

Wales_CA_EBA = Wales Copper Age and Early Bronze Age.


And this is why nMonte author Huijbregts recommends models with few best fitted pops

At first you need to run all populations to see what is best fitted and then if you want you can remove/add some pops.

[Gründig]

Wales_CA_EBA = Wales Copper Age and Early Bronze Age.

Do you know who they are genetically closest to? Or place with?

Also do you know anything about poprad medieval?

[Peterski]

Do you know who they are genetically closest to? Or place with?

Closest to Scotland_LBA and then to Unetice and Ireland_EBA:

Wales_CA_EBA:

[1] "1. CLOSEST SINGLE ITEM DISTANCE%"
Scotland_LBA Unetice Ireland_EBA England_MBA
2.696142 2.772665 2.832599 2.840973
Germany_Medieval England_CA_EBA Beaker_Britain Scotland_MBA
2.971424 3.147246 3.261999 3.487246

Mixed Mode (but distance is large in this case, so it's not the best fit):

[1] "distance%=2.1323"

Wales_CA_EBA

Scotland_LBA,52.4
Ireland_EBA,24.6
Baltic_IA,9.4
Poprad_Medieval,9.4
Sweden_Viking_Age,2.4
Armenia_MLBA,1.8

[Grace O'Malley]

Closest to Scotland_LBA and then to Unetice and Ireland_EBA:

Wales_CA_EBA:

[1] "1. CLOSEST SINGLE ITEM DISTANCE%"
Scotland_LBA Unetice Ireland_EBA England_MBA
2.696142 2.772665 2.832599 2.840973
Germany_Medieval England_CA_EBA Beaker_Britain Scotland_MBA
2.971424 3.147246 3.261999 3.487246

Mixed Mode (but distance is large in this case, so it's not the best fit):

[1] "distance%=2.1323"

Wales_CA_EBA

Scotland_LBA,52.4
Ireland_EBA,24.6
Baltic_IA,9.4
Poprad_Medieval,9.4
Sweden_Viking_Age,2.4
Armenia_MLBA,1.8

Could you please see who is closest to Ireland_EBA and then I'll give you a break? :) I think Ireland_EBA would be Rathlin and I'd love to see a breakdown of them. Thank you for doing this for everyone.

[Peterski]

Could you please see who is closest to Ireland_EBA and then I'll give you a break? :) I think Ireland_EBA would be Rathlin and I'd love to see a breakdown of them. Thank you for doing this for everyone.

Closest to Ireland_EBA are:

[1] "1. CLOSEST SINGLE ITEM DISTANCE%"
England_CA_EBA England_MBA Beaker_Britain Scotland_CA_EBA Scotland_MBA
1.467497 1.524877 1.703544 2.047650 2.117615
Netherlands_BA Unetice Scotland_LBA
2.356201 2.366064 2.590203

Mixed Mode:

[1] "distance%=1.2423"

Ireland_EBA

England_MBA,28.6
Wales_CA_EBA,18.2
England_CA_EBA,13.6
Scotland_CA_EBA,8.6
Finn_ancient,8
England_IA,7.4
Beaker_Britain,6.4
Netherlands_BA,4.4
England_LBA,2.6
Sarmatian_Pokrovka,2.2

[J. Ketch]

I always find this PCA from Anthrogenica handy.
https://postimg.cc/zLfH8yZx

[Peterski]

and then I'll give you a break?

I'm stuck at the computer because I'm working on something else right now.

But running these "on the side" is several clicks of my mouse for each run.

So go on, I can run some more if you want.

[Peterski]

Lombards from Collegno:

[1] "1. CLOSEST SINGLE ITEM DISTANCE%"
Hungary_Medieval_Szolad Germany_Medieval Poprad_Medieval
1.371117 3.477097 3.484249
Hallstatt_Bylany Scythian_Hungary Scotland_LBA
3.524792 3.819605 4.048138
England_IA England_MBA
4.142336 4.617811

Mixed Mode:

[1] "distance%=0.7239"

Italy_Medieval_Collegno

Nordic_IA,17.8
Wales_CA_EBA,17
Germany_Medieval,15.8
Scotland_LBA,15.2
Balkans_IA,10
Italian_Tuscan,10 ---> modern population (I included as a possible proxy for Roman settlers in Britain)
Anatolia_MLBA,4.4
Hungary_Medieval_Szolad,3.2
Iberia_BA,2.2
Iberia_Southwest_CA,2.2
Iberia_Central_CA,1.6
Ostrogothic_Crimea_ACD,0.6

Seems obvious that they were mixed with local Cisalpine Gauls and Romans.

[Grace O'Malley]

I always find this PCA from Anthrogenica handy.
https://postimg.cc/zLfH8yZx

That's a very useful map. :thumb001:

[Peterski]

Modern Tuscans removed:

[1] "distance%=0.7298"

Italy_Medieval_Collegno

Nordic_IA,18
Wales_CA_EBA,16.6
Germany_Medieval,16
Scotland_LBA,14.6
Balkans_IA,12.2
Anatolia_MLBA,6.6
Hungary_Medieval_Szolad,6.6
Iberia_BA,4.6
Iberia_Central_CA,2.4
Ostrogothic_Crimea_ACD,1.6
Iberia_Southwest_CA,0.8

[Gründig]

Modern Tuscans removed:

[1] "distance%=0.7298"

Italy_Medieval_Collegno

Nordic_IA,18
Wales_CA_EBA,16.6
Germany_Medieval,16
Scotland_LBA,14.6
Balkans_IA,12.2
Anatolia_MLBA,6.6
Hungary_Medieval_Szolad,6.6
Iberia_BA,4.6
Iberia_Central_CA,2.4
Ostrogothic_Crimea_ACD,1.6
Iberia_Southwest_CA,0.8


Did you remove the southern lombard samples?

[Peterski]

England Anglo-Saxons (again the same model with ~70 ancient populations included):

[1] "1. CLOSEST SINGLE ITEM DISTANCE%"
England_MBA Ireland_EBA Scotland_MBA England_CA_EBA
1.785792 2.190510 2.190599 2.298404
Scotland_LBA Unetice Germany_Medieval Beaker_Britain
2.414383 2.576042 2.613092 2.660588

[1] "distance%=1.1682"

England_Anglo-Saxon

England_IA,31.4
Nordic_IA,23.8
Ireland_EBA,13
Finn_ancient,12.4
Wales_CA_EBA,11.6
Netherlands_BA,5.6
Iberia_Central_CA,1.4
Poprad_Medieval,0.8

[Peterski]

Did you remove the southern lombard samples?

You should ask Davidski if he removed them, because I'm using the spreadsheet with averages.

There are two Global25 spreadsheets - one with individual samples, one with population averages.

The one with averages has 3 averages for Collegno:

Italy_Medieval_Collegno
Italy_Medieval_Collegno_o1
Italy_Medieval_Collegno_o2

I'm using the first one, so without "outliers1" and "outliers2".

Maybe one group of outliers is actually more Germanic?

========

As for Anglo-Saxons:

The spreadsheet with averages has only one average for Anglo-Saxons, which is weird (because some of them were mixed with Celts - so in my opinion Davidski should have split them into 2 averages; one with more Germanic-leaning samples and one with heavily Celtic-admixed samples).

[Peterski]

Lombard outliers:

Collegno_outliers1:

[1] "1. CLOSEST SINGLE ITEM DISTANCE%"
Anatolia_EBA Anatolia_BA Anatolia_MLBA
4.060575 4.091294 4.101854
Mycenaean Ostrogothic_Crimea_ACD Balkans_IA
5.054249 5.340195 5.718183
Balkans_BA Scythian_Moldova
8.444893 8.742732

[1] "distance%=1.289"

Italy_Medieval_Collegno_o1

Anatolia_EBA,49.2
Anatolia_BA,15.2
Anatolia_MLBA,12.6
Sweden_Viking_Age,6.6
Iberia_BA,4.8
Baltic_IA,3.4
Ostrogothic_Crimea_ACD,3.4
Iran_recent,2.6
Nordic_IA,2.2

Collegno_outliers2:

[1] "1. CLOSEST SINGLE ITEM DISTANCE%"
Gepid_Serbia_ACD Anatolia_IA Ostrogothic_Crimea_ACD
7.671005 7.689554 12.297804
Scythian_Moldova Hungary_IA Anatolia_Ottoman
12.478282 12.804174 13.196364
Cimmerian_Moldova Iran_recent
13.369975 14.341648

[1] "distance%=4.0641"

Italy_Medieval_Collegno_o2

Anatolia_EBA,31
Gepid_Serbia_ACD,23
Hun-Sarmatian,16.8
Iberia_Central_CA,12.6
Armenia_EBA,11
Iberia_Southwest_CA,4.2
Baltic_IA,1
Ireland_MN,0.4

[Gründig]

Double post

[Gründig]

You should ask Davidski if he removed them, because I'm using the spreadsheet with averages.

There are two Global25 spreadsheets - one with individual samples, one with population averages.

The one with averages has 3 averages for Collegno:

Italy_Medieval_Collegno
Italy_Medieval_Collegno_o1
Italy_Medieval_Collegno_o2

I'm using the first one, so without "outliers1" and "outliers2".

Maybe one group of outliers is actually more Germanic?

The spreadsheet with averages has only one average for Anglo-Saxons, which is weird (because some of them were mixed with Celts - so in my opinion Davidski should have split them into 2 averages; one with more Germanic-leaning samples and one with heavily Celtic-admixed samples).

Unfortunately they aren't split up on the spread sheets. However, David and the nMonte creator posted the southern samples on anthrogenica. It's a pain but you have to manually remove each sample that are on the non average individual data sheet. I think around 15 samples total between both lombard groups (hungary and Italy).

I can't seem to get on anthrogenica at the moment though, but I have ran the a model without them against my own coordinates.

[Peterski]

Okay, Lombard outliers:

Collegno_outliers1:

[1] "1. CLOSEST SINGLE ITEM DISTANCE%"
Anatolia_EBA Anatolia_BA Anatolia_MLBA
4.060575 4.091294 4.101854
Mycenaean Ostrogothic_Crimea_ACD Balkans_IA
5.054249 5.340195 5.718183
Balkans_BA Scythian_Moldova
8.444893 8.742732

[1] "distance%=1.289"

Italy_Medieval_Collegno_o1

Anatolia_EBA,49.2
Anatolia_BA,15.2
Anatolia_MLBA,12.6
Sweden_Viking_Age,6.6
Iberia_BA,4.8
Baltic_IA,3.4
Ostrogothic_Crimea_ACD,3.4
Iran_recent,2.6
Nordic_IA,2.2

Collegno_outliers2:

[1] "1. CLOSEST SINGLE ITEM DISTANCE%"
Gepid_Serbia_ACD Anatolia_IA Ostrogothic_Crimea_ACD
7.671005 7.689554 12.297804
Scythian_Moldova Hungary_IA Anatolia_Ottoman
12.478282 12.804174 13.196364
Cimmerian_Moldova Iran_recent
13.369975 14.341648

[1] "distance%=4.0641" ---> I need to include more of Mongoloid pops to get a good fit for this one

Italy_Medieval_Collegno_o2

Anatolia_EBA,31
Gepid_Serbia_ACD,23
Hun-Sarmatian,16.8
Iberia_Central_CA,12.6
Armenia_EBA,11
Iberia_Southwest_CA,4.2
Baltic_IA,1
Ireland_MN,0.4

[Gründig]

Okay, Lombard outliers:

Collegno_outliers1:

[1] "1. CLOSEST SINGLE ITEM DISTANCE%"
Anatolia_EBA Anatolia_BA Anatolia_MLBA
4.060575 4.091294 4.101854
Mycenaean Ostrogothic_Crimea_ACD Balkans_IA
5.054249 5.340195 5.718183
Balkans_BA Scythian_Moldova
8.444893 8.742732

[1] "distance%=1.289"

Italy_Medieval_Collegno_o1

Anatolia_EBA,49.2
Anatolia_BA,15.2
Anatolia_MLBA,12.6
Sweden_Viking_Age,6.6
Iberia_BA,4.8
Baltic_IA,3.4
Ostrogothic_Crimea_ACD,3.4
Iran_recent,2.6
Nordic_IA,2.2

Collegno_outliers2:

[1] "1. CLOSEST SINGLE ITEM DISTANCE%"
Gepid_Serbia_ACD Anatolia_IA Ostrogothic_Crimea_ACD
7.671005 7.689554 12.297804
Scythian_Moldova Hungary_IA Anatolia_Ottoman
12.478282 12.804174 13.196364
Cimmerian_Moldova Iran_recent
13.369975 14.341648

[1] "distance%=4.0641" ---> I need to include more of Mongoloid pops to get a good fit for this one

Italy_Medieval_Collegno_o2

Anatolia_EBA,31
Gepid_Serbia_ACD,23
Hun-Sarmatian,16.8
Iberia_Central_CA,12.6
Armenia_EBA,11
Iberia_Southwest_CA,4.2
Baltic_IA,1
Ireland_MN,0.4

For all the tests you've been running, are you using scaled or unscaled?

Also, what distance do you personally consider a good fit (I know it depends on whether you are using scaled or unscaled)? It seems there are many different opinions on this.

[Aren]

This PCA is screwed by ethnic-specific genetic drift. Look where Icelanders plot for example, you'd think they are more Germanic than Norwegians based on their proximity to Nordic_IA, but when you exclude recent genetic drift (or replicate this in a tri-dimensional PCA), they are actually somewhere in between Norwegians and Irish. Davidski made it for a very specific purpose: distinguishing modern-day Germanic and Celtic-speakers. Ireland, as a insular and until recently, a quite isolated population, did experienced extensive drift, specially after the drastic population reduction during the Great Famine.

Screwed? These kind of specific genetic drift PCA:s are very valuable when looking at populations who are extremely close atusomally. Why would we wanna exlcude it? Whatever genetic drift Irish experienced so did the Scots, Welsh(no potato famine there) and to some extent the English. And somehow it made them cluster very tightly with the Bronze Age samples from Britain we have so far. Strange coincidence huh?
And what would the problem be with Icelanders position in this plot? There's documented migration of Germans, Dutch, Belgians, Brits etc all over Scandinavia(Norway included) starting from the middle ages well into the modern era. Some big cities in Scandinavia were almost half German during the Hansa period. Iceland wasn't effected of this which is why Norwegians are deviating slightly towards Germans and Dutch in comparison with the Nordic_IA sample. This is also supported by the existance of lineages such as J, G and E in modern Norwegians, Swedes and Danes but absent in Icelanders and the medieval Germanic genomes we have so far.

Again wrong, Beaker_Iberia average includes only individuals with steppe ancestry. Exceptions are what they are - exceptions, one of the Beaker_Hungary samples is almost fully Yamnaya, should we use it as premise to suggest Beakers were initially almost fully Yamnaya, even though it is pretty obvious that the material culture originated in Central Europe, in the GAC-Yamnaya contact zone? Should we suggest that Bronze Age Hungarians were more steppe than Bronze Age Balts?
Point is there are samples from Beaker-period Iberia who have more Steppe than modern day Iberians, and just north of the Pyrenees in Southern France out of the three samples, two of them are significantly more Steppe than the Iberian Beakers and ofc modern Iberians.
Btw BA_Hungarians were more Steppe than Bronze Age Balts, the CWC samples from Lithuania and Estonia are very heavily HG-admixed, with the exception of some very early CWC samples. We see a continuation with that but with even more HG-admix in Baltic_BA and Baltic_IA. Hungary was in fact part of the (Western)Yamnaya cluster and there are samples of Hungarian Yamnaya coming soon.

During the early Bronze Age, Beakers were the only collective in Western Europe with substantial steppe ancestry. In this particular case, Beaker and 'individuals with steppe ancestry' is interchangeable.
This is all of the pre-print of the Iberia paper, soon to be published.

During the later Iron Age, the first genome-wide data from ancient non-Indo-European speakers showed that they were similar to contemporaneous Indo-European speakers and derived most of their ancestry from the earlier Bronze Age substratum. With the exception of Basques, who remain broadly similar to Iron Age populations, during the last 2500 years Iberian populations were affected by additional gene-flow from the Central/Eastern Mediterranean region, probably associated to the Roman conquest, and from North Africa during the Moorish conquest but also in earlier periods, probably related to the Phoenician-Punic colonization of Southern Iberia.
There's no mention of a Northern like input.

Irish did received substantial Germanic admixture as shown by recent peer-reviewed genetic studies, and the estimated dates of these events (using the still dubious GLOBETROTTER, but whatever) corresponds exactly to the Norse Viking and Anglo-Norman migrations. You can't overlook a entire study based on a screwed PCA interpretation.
You mean a study that showed Irish and Brits as a mixture of modern French, Scandinavians, Belgians, Spanish and what not, using ADMIXTURE? Well yes I will overlook those kind of studies with ease. ADMIXTURE to determine actual ancestry is as about as accurate as Gedmatch oracles.
There's no clear deviation in the case of the Irish from the aDNA we have so far from Britain. Hell even the BA sample from Ireland is autosomally closer to Scandinavians than the modern day Irish are. Both auDNA and genetic drift sensitve PCA:s as shown above tell us that they don't have much of non-British admixture. I would trust this much more than using ADMIXTURE and modelling Brits as a mixture of modern day Europeans.

[Grace O'Malley]

Screwed? These kind of specific genetic drift PCA:s are very valuable when looking at populations who are extremely close atusomally. Why would we wanna exlcude it? Whatever genetic drift Irish experienced so did the Scots, Welsh(no potato famine there) and to some extent the English. And somehow it made them cluster very tightly with the Bronze Age samples from Britain we have so far. Strange coincidence huh?
And what would the problem be with Icelanders position in this plot? There's documented migration of Germans, Dutch, Belgians, Brits etc all over Scandinavia(Norway included) starting from the middle ages well into the modern era. Some big cities in Scandinavia were almost half German during the Hansa period. Iceland wasn't effected of this which is why Norwegians are deviating slightly towards Germans and Dutch in comparison with the Nordic_IA sample. This is also supported by the existance of lineages such as J, G and E in modern Norwegians, Swedes and Danes but absent in Icelanders and the medieval Germanic genomes we have so far.

Point is there are samples from Beaker-period Iberia who have more Steppe than modern day Iberians, and just north of the Pyrenees in Southern France out of the three samples, two of them are significantly more Steppe than the Iberian Beakers and ofc modern Iberians.
Btw BA_Hungarians were more Steppe than Bronze Age Balts, the CWC samples from Lithuania and Estonia are very heavily HG-admixed, with the exception of some very early CWC samples. We see a continuation with that but with even more HG-admix in Baltic_BA and Baltic_IA. Hungary was in fact part of the (Western)Yamnaya cluster and there are samples of Hungarian Yamnaya coming soon.

This is all of the pre-print of the Iberia paper, soon to be published.

There's no mention of a Northern like input.

You mean a study that showed Irish and Brits as a mixture of modern French, Scandinavians, Belgians, Spanish and what not, using ADMIXTURE? Well yes I will overlook those kind of studies with ease. ADMIXTURE to determine actual ancestry is as about as accurate as Gedmatch oracles.
There's no clear deviation in the case of the Irish from the aDNA we have so far from Britain. Hell even the BA sample from Ireland is autosomally closer to Scandinavians than the modern day Irish are. Both auDNA and genetic drift sensitve PCA:s as shown above tell us that they don't have much of non-British admixture. I would trust this much more than using ADMIXTURE and modelling Brits as a mixture of modern day Europeans.

So Aren are you dismissing the People of the British Isles Study, Irish DNA Atlas study and Insular Celtic dna paper? It would be more odd if the Irish don't have any admixture. Vikings had settlements in Ireland so how would there be no Scandinavian admixture? There was the Norse-Gaels population that was created from mixture with the Vikings. Also Normans with many Irish today having a Norman surname and then English and Scots plantations in many areas of Ireland. The studies find evidence of this admixture and you disagree? Just odd that you dismiss these studies and history. Even Davidski said that the plot is for a specific use and that the G25 was more informative. He was specifically looking at Irish and British drift to separate Celtic populations. Davidski said this plot was sensitive to Irish and British drift. What do you think of the G25 plot then which doesn't show drift?

[Grace O'Malley]

This is the Global 25 plot which isn't affected by drift. Whether a population is drifted doesn't show their descent or admixture.

http://i67.tinypic.com/15oh3d3.jpg

If you want to know what drift does you can see this with the Irish Traveller population who form their own genetic cluster away from the settler Irish. They are still fully Irish though but have diverged from the Irish population in only about 400 - 500 years. So you can't use a map specifically showing drift to make a conclusion about a population's relatedness to another population.

http://compsoc.nuigalway.ie/~dubhthach/DNA/traveller-pca.png

The aim of Davidski's Celtic vs Germanic plot was to specifically use drift to separate those populations more because on most plots they are clustered together. He could do this with Icelanders as well as they are a drifted population but his aim in this plot was specifically centred on Irish and British drift. He also said that this PCA was mostly about relatively recent genetic drift as well.

[Ibericus]

Modeling spaniards and north-italians using Hallstatt gives actually a very good fit :

(Cypriots for post-East-Med/West-Asia) :

[1] "distance%=2.3788 / distance=0.023788"

Spanish_Aragon

Hallstatt_Bylany 63.80
Iberia_N 18.55
Cypriot 11.00
Iberia_ChL 4.25
Moroccan 2.40


1] "distance%=2.134 / distance=0.02134"

Italian_Bergamo

Hallstatt_Bylany 55.1
Cypriot 29.4
Iberia_N 15.4

[FilhoV]

Many thanks for the help Ibericus

[1] "distancia% = 2.3858/Distance = 0.023858"

FilhoV_scaled

Hallstatt_Bylany 43,0
Portugal_MBA 24,2
Anatolia_BA 21,0
Maroccan 11,8
Portugal_LNCA 0,0
Portugal_MN 0,0

[Ibericus]

Using Bronze Age Iberia instead of Iberian Bell-Beaker (beacause Bell Beakers is very heterogeneus, from EEF-like to Northern-Euro) :

[1] "distance%=1.6861 / distance=0.016861"

Spanish_Baleares

Hallstatt_Bylany 46.8
Iberia_BA 29.7
Anatolia_BA 21.1
Moroccan 2.5

[1] "distance%=1.9934 / distance=0.019934"

Spanish_Cantabria

Hallstatt_Bylany 41.40
Iberia_BA 37.90
Anatolia_BA 12.35
Portugal_MBA 6.15
Moroccan 2.20

[FilhoV]

Using Bronze Age Iberia instead of Iberian Bell-Beaker (beacause Bell Beakers is very heterogeneus, from EEF-like to Northern-Euro) :

[1] "distance%=1.6861 / distance=0.016861"

Spanish_Baleares

Hallstatt_Bylany 46.8
Iberia_BA 29.7
Anatolia_BA 21.1
Moroccan 2.5

[1] "distance%=1.9934 / distance=0.019934"

Spanish_Cantabria

Hallstatt_Bylany 41.40
Iberia_BA 37.90
Anatolia_BA 12.35
Portugal_MBA 6.15
Moroccan 2.20

Mind using those ancient populations with the Canary islander on the Global25 Spreadsheet I’m curious to see what he comes oit

[Aren]

So Aren are you dismissing the People of the British Isles Study, Irish DNA Atlas study and Insular Celtic dna paper? It would be more odd if the Irish don't have any admixture. Vikings had settlements in Ireland so how would there be no Scandinavian admixture? There was the Norse-Gaels population that was created from mixture with the Vikings. Also Normans with many Irish today having a Norman surname and then English and Scots plantations in many areas of Ireland. The studies find evidence of this admixture and you disagree? Just odd that you dismiss these studies and history. Even Davidski said that the plot is for a specific use and that the G25 was more informative. He was specifically looking at Irish and British drift to separate Celtic populations. Davidski said this plot was sensitive to Irish and British drift. What do you think of the G25 plot then which doesn't show drift?
Modern day Brits with the Bronze-Iron Age British/Irish samples + Nordic_IA with Davids PCA 25 spreadsheet

"distance%=1.1771"

Irish

England_Roman,51.6
Scotland_LBA,30
England_IA,18.4

"distance%=1.0441"

Scottish

England_Roman,67.8
England_IA,18
Scotland_LBA,8.8
Nordic_IA,5.4

"distance%=1.4109"

Welsh

England_IA,43.2
Scotland_LBA,27.6
England_Roman,20.6
Nordic_IA,8.6

So it's not just the very informative genetic drift sensitive PCA that David created, but also nMonte with his spreadsheet shows no substantial post-Roman era input from Scandinavia either.

[Aren]

This is the Global 25 plot which isn't affected by drift. Whether a population is drifted doesn't show their descent or admixture.

http://i67.tinypic.com/15oh3d3.jpg

If you want to know what drift does you can see this with the Irish Traveller population who form their own genetic cluster away from the settler Irish. They are still fully Irish though but have diverged from the Irish population in only about 400 - 500 years. So you can't use a map specifically showing drift to make a conclusion about a population's relatedness to another population.

http://compsoc.nuigalway.ie/~dubhthach/DNA/traveller-pca.png

The aim of Davidski's Celtic vs Germanic plot was to specifically use drift to separate those populations more because on most plots they are clustered together. He could do this with Icelanders as well as they are a drifted population but his aim in this plot was specifically centred on Irish and British drift. He also said that this PCA was mostly about relatively recent genetic drift as well.

What does the first plot show really? Ancient samples from Britain aren't included so we can't make any conclusion whether Irish drift towards Scandinavia in relation to them. But as shown in my post above, nMonte doesn't pick up any Scandinavian(Nordic_IA).
Also regarding your try to disqualify Davids genetic drift sensitive PCA. How do you explain that the same kind of genetic drift that happend in Ireland also seems to be shared with Scots and Welshmen, even the English to some degree? And why are the Irish samples plotting well within the earlier Bronze and Iron Age samples from the British Isles and not drifting towards Norwegians? Is this a coincidence? No ofc not, it just shows a great continuity in Britain and Ireland since that era. More than 70% of all modern Irish men carry the same Y-DNA of the earlier British Beakers for crying out loud. I don't understand how you can argue for your cause so surely. Anyone who's not biased can see that there's no clear Scandinavian input(or atleast substantial amount of it) in modern day Brits.

[J. Ketch]

Modern day Brits with the Bronze-Iron Age British/Irish samples + Nordic_IA with Davids PCA 25 spreadsheet

"distance%=1.1771"

Irish

England_Roman,51.6
Scotland_LBA,30
England_IA,18.4

"distance%=1.0441"

Scottish

England_Roman,67.8
England_IA,18
Scotland_LBA,8.8
Nordic_IA,5.4

"distance%=1.4109"

Welsh

England_IA,43.2
Scotland_LBA,27.6
England_Roman,20.6
Nordic_IA,8.6

So it's not just the very informative genetic drift sensitive PCA that David created, but also nMonte with his spreadsheet shows no substantial post-Roman era input from Scandinavia either.
Using your method with my father's data:

[1] "distance%=2.0605"

England_Roman,70.8
Nordic_IA,29.2

Removing Germanic admixed England Roman, adding ENF heavy component.

[1] "distance%=2.2888"

Scotland_LBA,40.4
Nordic_IA,38.4
England_MBA,9
Wales_CA_EBA,8.4
Remedello_BA,3.8

[1] "distance%=2.2708"

Nordic_IA,39.8
Scotland_LBA,39.4
Wales_CA_EBA,9.2
England_MBA,6.8
Ireland_MN,4.8

[Grace O'Malley]

Modern day Brits with the Bronze-Iron Age British/Irish samples + Nordic_IA with Davids PCA 25 spreadsheet

"distance%=1.1771"

Irish

England_Roman,51.6
Scotland_LBA,30
England_IA,18.4

"distance%=1.0441"

Scottish

England_Roman,67.8
England_IA,18
Scotland_LBA,8.8
Nordic_IA,5.4

"distance%=1.4109"

Welsh

England_IA,43.2
Scotland_LBA,27.6
England_Roman,20.6
Nordic_IA,8.6

So it's not just the very informative genetic drift sensitive PCA that David created, but also nMonte with his spreadsheet shows no substantial post-Roman era input from Scandinavia either.

I'll still go with what the scientists say and the peer reviewed studies. Everyone here are amateurs so we're no experts although we have a great interest in the topic. But I think the scientists who have been working in this field and have more knowledge about the subject such as Prof Dan Bradley and Prof Walter Bodmer are the people I'll give more weight to. Interesting discussion though.

There is a study coming out sometime by Lara Cassidy who was heavily involved in the Rathlin paper, the Insular Celtic paper and the population genomics of archaeological transition in Western Iberia with genomes from different periods in Ireland so I'm sure a lot of this stuff will be solved down the track.

[Peterski]

Modern day Brits with the Bronze-Iron Age British/Irish samples + Nordic_IA with Davids PCA 25 spreadsheet

"distance%=1.1771"

Irish

England_Roman,51.6
Scotland_LBA,30
England_IA,18.4

"distance%=1.0441"

Scottish

England_Roman,67.8
England_IA,18
Scotland_LBA,8.8
Nordic_IA,5.4

"distance%=1.4109"

Welsh

England_IA,43.2
Scotland_LBA,27.6
England_Roman,20.6
Nordic_IA,8.6

So it's not just the very informative genetic drift sensitive PCA that David created, but also nMonte with his spreadsheet shows no substantial post-Roman era input from Scandinavia either.

England_Roman average is Germanic-admixed though (specifically those two gladiators with U106).

[Peterski]

Without England_Roman I get this for modern Irish:

[1] "distance%=1.2732"

Irish

Scotland_LBA,51
England_IA,24
Wales_CA_EBA,16
Nordic_IA,5.8
Ireland_MN,3.2

[Peterski]

BTW here is a rather good fit I got for myself, I included 18 ancient populations:

[1] "distance%=1.9777"

Peterski_scaled

Golden_Horde_European,48.8
Ireland_EBA,17.4
Sweden_Viking_Age,14.4
Poprad_Medieval,10.6
Balkans_IA,7.8
Baltic_IA,1

^^^
This looks a lot like my DNA Land (55% NS, 36% NW, 8% Balkan, 1% Italian).

Other populations included in that run were:

Poland_BA
Nordic_IA
Globular_Amphora
England_Anglo-Saxon
Scotland_LBA
England_IA
Germany_Medieval
Hallstatt_Bylany
Unetice
Hungary_IA
Finn_ancient
Levant_BA_North

=====

After removing Sweden Viking Age, fit is worse (I'm trying to get fits below 2):

[1] "distance%=2.0226"

Peterski_scaled

Golden_Horde_European,51.4
Ireland_EBA,33
Balkans_IA,9.8
Poprad_Medieval,3.2
Baltic_IA,2.6

=====

After removing Ireland_EBA, fit is again worse. What can I add to improve fits?:

[1] "distance%=2.1555"

Peterski_scaled

Golden_Horde_European,45.8
Poprad_Medieval,21.4
Scotland_LBA,10.6
Poland_BA,8.4
Balkans_IA,8
Baltic_IA,5.8

^^^
At least now I'm getting populations geographically closer (Poprad, Poland_BA).

=====

Replacing Golden Horde European with Avar Hungary is only a slight improvement:

[1] "distance%=2.0575"

Peterski_scaled

Avar_Hungary_Szolad,58.6
Poland_BA,19.8
Balkans_IA,8.4
Baltic_IA,7
Poprad_Medieval,6.2

^^^
This is already after previous deletion of Sweden Viking Age and Ireland_EBA.

[Grace O'Malley]

I just don't know enough about all these ancient models and some obviously cancel each other out. Anyway as I've said before I'm just going to leave it up to the geneticists.

I'm going to have a read about using nMonte as it is fun to play around with. Wales only has 3 people so it is a very limited sample.

Anyway this is using nMonte runner. I've taken out the England_Roman but still have England IA. Myself and Irish_Average.

http://i68.tinypic.com/30w4ifo.jpg

[Peterski]

Adding Baltic Bronze Age (instead of just Baltic Iron Age) improved the fits for me:

[1] "distance%=1.973"

Peterski_scaled

Avar_Hungary_Szolad,32.8
Baltic_BA,15.6
Poland_BA,12.8
Balkans_IA,12
Poprad_Medieval,10.4
Golden_Horde_European,9.4
England_Anglo-Saxon,7

I don't think there was a real significant migration from the Balkans to Poland, though.

So I will remove Balkans_IA and check what happens then.

=====

After removing Balkans_IA fit was worse than 2. But adding back Viking Age improved it:

[1] "distance%=1.9772"

Peterski_scaled

Golden_Horde_European,29
Avar_Hungary_Szolad,28.8
Poprad_Medieval,24.4
Sweden_Viking_Age,15.8
Levant_BA_North,1.8
Poland_BA,0.2

But now I get a bit of Levant Bronze Age. So obviously something southern is there.

[Peterski]

I've taken out the England_Roman but still have England IA. Myself and Irish_Average.

It seems that Scotland_LBA (Late Bronze Age) is a better fit for modern Irish, than Ireland_EBA (Rathlin Island).

Scotland_LBA is also a rather solid average, because it includes four samples (while Ireland_EBA only 2 samples):

Scotland_LBA:I2859
Scotland_LBA:I2860
Scotland_LBA:I2861
Scotland_LBA:I3130

Versus:

Ireland_EBA:Rathlin1
Ireland_EBA:Rathlin3

[FilhoV]

This is mine when using England_Roman

"distance%=2.0901"

FilhoV_scaled

Italian_Tuscan,42.4
Iberia_BA,28.6
England_Roman,15.6
Mozabite,13.4

[FilhoV]

Also I have a Bronze Age Portuguese kit from the area around Alentejo if I recall

Kit Number: T277697 Elapsed Time: 4.91 seconds

Population
North_Atlantic 46.80
Baltic 4.86
West_Med 45.59
West_Asian -
East_Med 2.75
Red_Sea -
South_Asian -
East_Asian -
Siberian -
Amerindian -
Oceanian -
Northeast_African -
Sub-Saharan -

[Damião de Góis]

Also I have a Bronze Age Portuguese kit from the area around Alentejo if I recall

Kit Number: T277697 Elapsed Time: 4.91 seconds

Population
North_Atlantic 46.80
Baltic 4.86
West_Med 45.59
West_Asian -
East_Med 2.75
Red_Sea -
South_Asian -
East_Asian -
Siberian -
Amerindian -
Oceanian -
Northeast_African -
Sub-Saharan -

There's also pseudo-sardinian (to say the least, they are very far from Sardinia too) like samples from Bronze Age Portugal:

# Population Percent
1 West_Med 58.17
2 North_Atlantic 33.93
3 Red_Sea 5.61
4 Sub-Saharan 2.28

Single Population Sharing:

# Population (source) Distance
1 Sardinian 22.14
2 French_Basque 27.15
3 Southwest_French 30.02
4 Spanish_Aragon 30.15
5 Spanish_Cantabria 30.4
6 Spanish_Andalucia 30.64
7 Spanish_Castilla_La_Mancha 30.95
8 Spanish_Extremadura 32.26
9 Spanish_Valencia 32.6
10 Spanish_Galicia 32.84
11 Spanish_Castilla_Y_Leon 33.28
12 Spanish_Murcia 33.29
13 Portuguese 34.01
14 Spanish_Cataluna 34.13
15 North_Italian 36.09
16 Tuscan 39.82
17 French 40.35
18 West_Sicilian 43.43
19 South_Dutch 45.07
20 Italian_Abruzzo 45.34


Like i said it would be interesting to see results from Iron Age (we have none so far), especially from any of the Pre-roman people:

https://upload.wikimedia.org/wikipedia/commons/c/c8/Ethnographic_Iberia_200_BCE.PNG

[FilhoV]

There's also pseudo-sardinian (to say the least, they are very far from Sardinia too) like samples from Bronze Age Portugal:

# Population Percent
1 West_Med 58.17
2 North_Atlantic 33.93
3 Red_Sea 5.61
4 Sub-Saharan 2.28

Single Population Sharing:

# Population (source) Distance
1 Sardinian 22.14
2 French_Basque 27.15
3 Southwest_French 30.02
4 Spanish_Aragon 30.15
5 Spanish_Cantabria 30.4
6 Spanish_Andalucia 30.64
7 Spanish_Castilla_La_Mancha 30.95
8 Spanish_Extremadura 32.26
9 Spanish_Valencia 32.6
10 Spanish_Galicia 32.84
11 Spanish_Castilla_Y_Leon 33.28
12 Spanish_Murcia 33.29
13 Portuguese 34.01
14 Spanish_Cataluna 34.13
15 North_Italian 36.09
16 Tuscan 39.82
17 French 40.35
18 West_Sicilian 43.43
19 South_Dutch 45.07
20 Italian_Abruzzo 45.34


Like i said it would be interesting to see results from Iron Age (we have none so far), especially from any of the Pre-roman people:

https://upload.wikimedia.org/wikipedia/commons/c/c8/Ethnographic_Iberia_200_BCE.PNG

It would be interesting to say the least to have some Iron Age results hopefully we will see one soon

[Aren]

I'll still go with what the scientists say and the peer reviewed studies. Everyone here are amateurs so we're no experts although we have a great interest in the topic. But I think the scientists who have been working in this field and have more knowledge about the subject such as Prof Dan Bradley and Prof Walter Bodmer are the people I'll give more weight to. Interesting discussion though.

There is a study coming out sometime by Lara Cassidy who was heavily involved in the Rathlin paper, the Insular Celtic paper and the population genomics of archaeological transition in Western Iberia with genomes from different periods in Ireland so I'm sure a lot of this stuff will be solved down the track.

The people who worked on the British paper you are reffering to, show that Irish can be modelled(using ADMIXTURE) as 65% French and the rest being various Germanic input, mainly Scandinavian. I don't get how that is indicative of Scandinavian admixture in Ireland considering the fact that we do have now pre-Anglo Saxon, pre-Viking era samples from the British Isles and they are not like modern Frenchmen, they are significantly more Northern.

[Grace O'Malley]

The people who worked on the British paper you are reffering to, show that Irish can be modelled(using ADMIXTURE) as 65% French and the rest being various Germanic input, mainly Scandinavian. I don't get how that is indicative of Scandinavian admixture in Ireland considering the fact that we do have now pre-Anglo Saxon, pre-Viking era samples from the British Isles and they are not like modern Frenchmen, they are significantly more Northern.

They were modelled as Breton-like so Northern French. Breton's are close to populations from Ireland and Britain.

Anyway there are genomes from every period in Ireland in Prof Dan Bradley's lab so we should see more of these older genomes from Ireland which should give a clearer picture.

[Lucas]

The people who worked on the British paper you are reffering to, show that Irish can be modelled(using ADMIXTURE) as 65% French and the rest being various Germanic input, mainly Scandinavian. I don't get how that is indicative of Scandinavian admixture in Ireland considering the fact that we do have now pre-Anglo Saxon, pre-Viking era samples from the British Isles and they are not like modern Frenchmen, they are significantly more Northern.

Using Global25 it seems not valid. Proportions are opposite though:)

http://i67.tinypic.com/2ldq51.png

[Grace O'Malley]

Using Global25 it seems not valid. Proportions are opposite though:)

http://i67.tinypic.com/2ldq51.png

I don't know why they didn't use Dutch in both the PoBI and IDA. Dutch are always a close population. They didn't appear to use any Dutch samples.

[Aren]

England_Roman average is Germanic-admixed though (specifically those two gladiators with U106).
Lol Peterski, a year ago you were arguing over at Anthrogenica that the subclade of U106 found in the two gladiators is actually a specifc Celtic subclade of U106 and you made a very good case for it(I think this subclade peaks in Czechia right?).
What made you change your mind now?

Without England_Roman I get this for modern Irish:

[1] "distance%=1.2732"

Irish

Scotland_LBA,51
England_IA,24
Wales_CA_EBA,16
Nordic_IA,5.8
Ireland_MN,3.2
So even without it, there's very little Scandinavian input.

[Aren]

They were modelled as Breton-like so Northern French. Breton's are close to populations from Ireland and Britain.

Anyway there are genomes from every period in Ireland in Prof Dan Bradley's lab so we should see more of these older genomes from Ireland which should give a clearer picture.
Not just Bretons, they used French samples from all over France. And btw even Bretons are significantly southern shifted in comparison to modern day Brits, aswell as the earlier IA-BA samples ofc.

Lukasz has a Breton average for his K36 project.
https://i.imgur.com/zaZWjh9.png

[Lucas]

I don't know why they didn't use Dutch in both the PoBI and IDA. Dutch are always a close population. They didn't appear to use any Dutch samples.

Using German instead of Dutch. Fit is worser.

http://i64.tinypic.com/15ojwn9.png

[Peterski]

Lol Peterski, a year ago you were arguing over at Anthrogenica that the subclade of U106 found in the two gladiators is actually a specifc Celtic subclade of U106 and you made a very good case for it(I think this subclade peaks in Czechia right?).
What made you change your mind now?

I was arguing that their U106 subclade can be Celtic (and I just considered it a possibility, not a certainty) but I was not arguing that they are 100% autosomally Celtic.

Anyway during the last year we got a lot of new ancient Y-DNA samples including Migration Period Germanic samples so obviously there is a lot more evidence now to support the link between U106 and the spread of Germanics.

Yeah this one Unetice sample from Czechia still leaves some possibilities for Celtic U106 subclades.

[Aren]

Using Global25 it seems not valid. Proportions are opposite though:)

http://i67.tinypic.com/2ldq51.png

Dutch weren't used in that study. I don't think you will be able to replicate it, they used different averages than what Peterski has for sure. They had Scandinavia divided into regions aswell. And the Scandinavian input if I remember it correctly in Irish and Scots showed mostly as Northern Norwegian which is very odd, but maybe that's why they got so high French in the first place(around 65%) because Northern Scandinavians are very Finnish/NE Euro shifted.

[Aren]

I was arguing that their U106 subclade can be Celtic (and I just considered it a possibility, not a certainty) but I was not arguing that they are 100% autosomally Celtic.

Anyway during the last year we got a lot of new ancient Y-DNA samples including Migration Period Germanic samples so obviously there is a lot more evidence now to support the link between U106 and the spread of Germanics.

Not all U106, though. We already have a Bronze Age U106 sample from central Europe. It's the second oldest U106, no?
But it's not important, you showed that even without the Roman era samples Irish get basically noise level Nordic_IA.

[Grace O'Malley]

Not just Bretons, they used French samples from all over France. And btw even Bretons are significantly southern shifted in comparison to modern day Brits, aswell as the earlier IA-BA samples ofc.

Lukasz has a Breton average for his K36 project.
https://i.imgur.com/zaZWjh9.png

I've done that as well. This is the older version of that plot but you can see where I plot much the same place as my mother as well.

http://i65.tinypic.com/sq4mdd.jpg

[Peterski]

6% is not noise level come on...

To get 6% of Scandinavian there had to be dozens of thousands (or at least thousands) of Viking settlers in Ireland (basically 6% of the population).

[Grace O'Malley]

Dutch weren't used in that study. I don't think you will be able to replicate it, they used different averages than what Peterski has for sure. They had Scandinavia divided into regions aswell. And the Scandinavian input if I remember it correctly in Irish and Scots showed mostly as Northern Norwegian which is very odd, but maybe that's why they got so high French in the first place(around 65%) because Northern Scandinavians are very Finnish/NE Euro shifted.

It was northern and I think western Norwegian from memory but I have to check. I do agree though that the older genomes are the way to go so I'm really looking forward to these studies.

[Aren]

6% is not noise level come on...

To get 6% of Scandinavian there had to be dozens of thousands of Viking settlers in Ireland (basically 6% of the population).

I say noise level cause Nordic_IA is really close autosomally to the pre-Germanic British/Irish samples. 6% of a very similar source could very well be noise, even if it's not 6% to me is really low and not worth having a discussion over...

[Peterski]

I say noise level cause Nordic_IA is really close autosomally to the pre-Germanic British/Irish samples. 6% of a very similar source could very well be noise, even if it's not 6% to me is really low and not worth having a discussion over...

But as Grace mentioned there are documented Viking settlements, historical & archaeological evidence, surnames of Viking origin etc.

Were they all just acculturated Gaels LARP-ing as Norsemen? :)

[Lucas]

Dutch weren't used in that study. I don't think you will be able to replicate it, they used different averages than what Peterski has for sure. They had Scandinavia divided into regions aswell. And the Scandinavian input if I remember it correctly in Irish and Scots showed mostly as Northern Norwegian which is very odd, but maybe that's why they got so high French in the first place(around 65%) because Northern Scandinavians are very Finnish/NE Euro shifted.

Yes there aren't directly comparable, but still French 65% for Irish is too high to me. French has Roman input and other southern which is lacking in Irish.

[Aren]

But as Grace mentioned there are documented Viking settlements, historical & archaeological evidence, surnames of Viking origin etc.

Were they all just acculturated Gaels LARP-ing as Norsemen? :)
Historical records mean very little in the light of aDNA.

Yes there aren't directly comparable, but still French 65% for Irish is too high to me.
It's because the Scandinavian part is mainly Northern Scandinavian in the study and as you probably know Northern Swedes/Norwegians are very Finnish shifted so 30% North Scandinavian + the rest mainly being French makes sense.
That's how ADMIXTURE works...

[J. Ketch]

Irish:Average

Fit 0.839
Beaker Southern France 4.17
England IA 36.67
Germany Medieval 25.83
Ireland EBA 0
Scotland LBA 33.33

[Peterski]

Slightly off-topic but it's interesting how using Baltic_BA is a better fit when modeling myself than Baltic_IA. Probably a reflection of what happened in the Baltics during the transition from ~100% R1a in the Bronze Age to ~50% R1a and ~50% N1c later (something that only affected Proto-Balts but not Proto-Slavs).

[Grace O'Malley]

On the Global 25 my closest match is Icelandic. David explained this as due to Irish diversity not been captured by the Irish samples in the Global 25. He said that if samples from other areas of Ireland were added I would most likely match one of these Irish regions more than Iceland. It does however show that populations do have some diversity.

[Lucas]

It's because the Scandinavian part is mainly Northern Scandinavian in the study and as you probably know Northern Swedes/Norwegians are very Finnish shifted so 30% North Scandinavian + the rest mainly being French makes sense.
That's how ADMIXTURE works...

Ok, but it means this model was skewed. They can use instead of north Norway, for example Yoruba, and French would be rose to 100%.

[FilhoV]

I wonder what an Icelandic person would look using these populations

[Grace O'Malley]

Ok, but it means this model was skewed. They can use instead of north Norway, for example Yoruba, and French would be rose to 100%.

It's the same method they used in the PoBI study. This is the admixture graph and you can see where the Norwegian samples came from.

https://media.springernature.com/full/springer-static/image/art%3A10.1038%2Fs41598-018-24846-6/MediaObjects/41598_2018_24846_Fig1_HTML.jpg

[Grace O'Malley]

Modern day Brits with the Bronze-Iron Age British/Irish samples + Nordic_IA with Davids PCA 25 spreadsheet

"distance%=1.1771"

Irish

England_Roman,51.6
Scotland_LBA,30
England_IA,18.4

"distance%=1.0441"

Scottish

England_Roman,67.8
England_IA,18
Scotland_LBA,8.8
Nordic_IA,5.4

"distance%=1.4109"

Welsh

England_IA,43.2
Scotland_LBA,27.6
England_Roman,20.6
Nordic_IA,8.6

So it's not just the very informative genetic drift sensitive PCA that David created, but also nMonte with his spreadsheet shows no substantial post-Roman era input from Scandinavia either.

Okay Aren this is a bit coincidental but someone from Denmark contacted me on 23&Me as they share dna to the 4th Great Grandparent level with me and they have all their ancestry in Denmark. I've never looked before but what is interesting is that my mother, brother and myself are tested on 23andMe and the only countries other than Britain that we share ancestry with all 4 great grandparents that are born in the country are Ireland of course, then Britain, Sweden, Norway, Denmark and my mother also has one from Germany with all 4 great grandparents born there which neither me nor my brother have.

Either Irish ancestry is spread throughout all these countries or there is some Scandinavian connection in Ireland.

[Aren]

Okay Aren this is a bit coincidental but someone from Denmark contacted me on 23&Me as they share dna to the 4th Great Grandparent level with me and they have all their ancestry in Denmark. I've never looked before but what is interesting is that my mother, brother and myself are tested on 23andMe and the only countries other than Britain that we share ancestry with all 4 great grandparents that are born in the country are Ireland of course, then Britain, Sweden, Norway, Denmark and my mother also has one from Germany with all 4 great grandparents born there which neither me nor my brother have.

Either Irish ancestry is spread throughout all these countries or there is some Scandinavian connection in Ireland.

Denmark eh?
I thought the Scandinavian connection was strictly Norwegian according to you? ;)

One a more srious note, 4th great grandparent is rather recent. If the match is real then British/Irish ancestry in your Danish match is more likely imo. But then again I also have matches at that level or slightly below that according to Myheritage and 23andme who are 100% Northern European, which doesn't make any sense at all(I score 99,9% West Asian on 23andme).

[Grace O'Malley]

Denmark eh?
I thought the Scandinavian connection was strictly Norwegian according to you? ;)

One a more srious note, 4th great grandparent is rather recent. If the match is real then British/Irish ancestry in your Danish match is more likely imo. But then again I also have matches at that level or slightly below that according to Myheritage and 23andme who are 100% Northern European, which doesn't make any sense at all(I score 99,9% West Asian on 23andme).

It's a bit of a coincidence that those are the only countries that all 4 great grandparents are born in though. On MyHeritage I also have some from Iceland and the Faroes. The more I look into it the more I'm convinced. It is also a bit of a coincidence that on my mtdna I'm a one step match with people in Norway and Sweden. One of the matches I know from online and they are from Eastern Norway.

Even a sceptic like you :) has to admit that the weight of evidence is on the side of Scandinavian input into the Irish? Too much evidence stacked in that direction and not to mention that my top match on the Global 25 is Iceland. All these Scandinavian connections. :p

[Aren]

It's a bit of a coincidence that those are the only countries that all 4 great grandparents are born in though. On MyHeritage I also have some from Iceland and the Faroes. The more I look into it the more I'm convinced. It is also a bit of a coincidence that on my mtdna I'm a one step match with people in Norway and Sweden. One of the matches I know from online and they are from Eastern Norway.

Even a sceptic like you :) has to admit that the weight of evidence is on the side of Scandinavian input into the Irish? Too much evidence stacked in that direction and not to mention that my top match is Iceland. All these Scandinavian connections. :p

You have to look at it from a bigger picture though. How many say Germans, Frenchmen or Dutch or other Northern Euros also have some matches at that level with Scandinavians according to 23andme? Like I mentioned in my post it's hardly 100% accurate when I'm getting distant Northern European matches when I score 99,9% West Asian and they are 100% European on 23andme. Matches at that level indicate recent connections, not to the Viking era right?

And as you probably know all Northern Euros have very similar mtdna profiles. How far deep do you know your mtdna sublcade? What is the TMRCA between yours and the Scandinavian matches? It's the Y-DNA you should be looking at and Ireland has few typically Scandinavian subclades actually. What other kind of evidence is there exactly? The study showing that Irish can roughly be modelled as French+Scandinavian? Gedmatch already probably does that for you in the mixed oracle mode. That doesn't mean you are part French, or?

[Grace O'Malley]

You have to look at it from a bigger picture though. How many say Germans, Frenchmen or Dutch or other Northern Euros also have some matches at that level with Scandinavians according to 23andme? Like I mentioned in my post it's hardly 100% accurate when I'm getting distant Northern European matches when I score 99,9% West Asian and they are 100% European on 23andme. Matches at that level indicate recent connections, not to the Viking era right?

And as you probably know all Northern Euros have very similar mtdna profiles. How far deep do you know your mtdna sublcade? What is the TMRCA between yours and the Scandinavian matches? It's the Y-DNA you should be looking at and Ireland has few typically Scandinavian subclades actually. What other kind of evidence is there exactly? The study showing that Irish can roughly be modelled as French+Scandinavian? Gedmatch already probably does that for you in the mixed oracle mode. That doesn't mean you are part French, or?

If someone says black you'd say white. :) As I've said the only countries outside of Ireland and Britain that my family share dna that have all 4 great grandparents born in the country are Denmark, Norway, Sweden and my mother also has Germany as well. You do appear to argue against any Scandinavian connection with Ireland when there is a history of Viking invasions and Dublin alone was one of the largest Viking trading towns outside of Scandinavia. Is it so difficult to believe there is some shared connections?

I don't have any evidence of French connections even though I'd be thrilled because France is one of my favourite countries. Anyway I don't see what's so difficult to believe about that study when the majority of French dna is Northern French as is shown in the breakdown above. Northern French are close to Northwest Europeans genetically (well they are Northwest Europeans anyway).

The mtdna for a full match could be approx 250 years on HVR1, HVR2, & Coding Region match with a 1 step match could be a 1,000 years or more.

The Danish person contacted me so this is how I became aware and could see that it was at the 4th great grandparent level. I'd have to go though the others and see what level they are which I might sometime. I've never really looked at this before so wasn't expecting to find connections with people in Scandinavian countries that have all 4 sets of grandparents born there. If this is so common why don't I have this with other European countries or indeed other countries outside Europe besides the colonial ones of course?

Do you score Northern European matches that are at the 4th great grandparent level?

[Aren]

If someone says black you'd say white. :) As I've said the only countries outside of Ireland and Britain that my family share dna that have all 4 great grandparents born in the country are Denmark, Norway, Sweden and my mother also has Germany as well. You do appear to argue against any Scandinavian connection with Ireland when there is a history of Viking invasions and Dublin alone was one of the largest Viking trading towns outside of Scandinavia. Is it so difficult to believe there is some shared connections?

I don't have any evidence of French connections even though I'd be thrilled because France is one of my favourite countries. Anyway I don't see what's so difficult to believe about that study when the majority of French dna is Northern French as is shown in the breakdown above. Northern French are close to Northwest Europeans genetically (well they are Northwest Europeans anyway).

The mtdna for a full match could be approx 250 years on HVR1, HVR2, & Coding Region match with a 1 step match could be a 1,000 years or more.

The Danish person contacted me so this is how I became aware and could see that it was at the 4th great grandparent level. I'd have to go though the others and see what level they are which I might sometime. I've never really looked at this before so wasn't expecting to find connections with people in Scandinavian countries that have all 4 sets of grandparents born there. If this is so common why don't I have this with other European countries or indeed other countries outside Europe besides the colonial ones of course?

Do you score Northern European matches that are at the 4th great grandparent level?

We've been over this. Pre-Germanic/Pre-Viking era samples from Britain were not autosomally French-like, not even Northern French. So why would you think they are a good population to use to model the Irish? You haven't answered me on that yet.

[Grace O'Malley]

We've been over this. Pre-Germanic/Pre-Viking era samples from Britain were not autosomally French-like, not even Northern French. So why would you think they are a good population to use to model the Irish? You haven't answered me on that yet.

Because as I've said in my post above I don't see what the problem is? The French component is majority Northern French anyway and Northern French are very similar to British and Irish dna wise.

[Aren]

Because as I've said in my post above I don't see what the problem is? The French component is majority Northern French anyway and Northern French are very similar to British and Irish dna wise.

And I showed you that they actually aren't that close. And that the Iron Age and Roman era British samples actually plot with modern day Brits not with Northern French. A logical modelling would be with those pre-Germanic British samples + the Germanic samples we have so far you know like the nMonte run I showed you a couple pages back.

Here you see there's a significant gap between Bretons/Northern French and Brits.
https://i.imgur.com/zaZWjh9.png

[Grace O'Malley]

And I showed you that they actually aren't that close. And that the Iron Age and Roman era British samples actually plot with modern day Brits not with Northern French. A logical modelling would be with those pre-Germanic British samples + the Germanic samples we have so far you know like the nMonte run I showed you a couple pages back.

Here you see there's a significant gap between Bretons/Northern French and Brits.
https://i.imgur.com/zaZWjh9.png

I know that guy from Anthrogenica.

That's a plot by Lukasz and I've posted my position. If you put so much stock in that why do you think I plot where I do?

Anyway that is based on the K36. All these maps including Davidski's are very interesting and I think people like Lukasz and Davidski are very talented but they aren't qualified geneticists. You put a lot of stock in these plots and dismiss peer-reviewed scientific studies.

Anyway whether you agree or not on whether the Irish have genuine Scandinavian admixture you have to acknowledge that populations like the Irish and Scandinavians are quite close surely?

[Aren]

I know that guy from Anthrogenica.

That's a plot by Lukasz and I've posted my position. If you put so much stock in that why do you think I plot where I do?

Anyway that is based on the K36. All these maps including Davidski's are very interesting and I think people like Lukasz and Davidski are very talented but they aren't qualified geneticists. You put a lot of stock in these plots and dismiss peer-reviewed scientific studies.

Anyway whether you agree or not on whether the Irish have genuine Scandinavian admixture you have to acknowledge that populations like the Irish and Scandinavians are quite close surely?

You are one individual, why would I give more attention to where you plot when there's an Irish average?
It does not take a top notch geneticist to see that the aDNA samples we have from Britain so far aren't like modern Northern French. And you keep saying that it's Northern French but a significant amount of it is also Southern and Central French. Aswell as 5-10% Spanish.
But let's see if I understand your take on the British study correctly. You think that the Scandinavian shown in the ADMIXTURE run of the Irish is indicative of actual Scandinavian heritage, dismissing the fact that the more "southern" source in their run is French/Northern French, a bad proxy for how Ireland looked like pre-Viking era. You don't see the major flaw? If Northern French are significantly more southern shifted than the Roman era British samples then it's going to compensate with giving more Northern-like admixture ie Scandinavian(mainly Northern Scandinavian to be precise).

Obviously I believe that modern Brits and Scandinavians are very closely related autosomally speaking, I've argued for it many times here. But that doesn't necessarily mean that they have a shared heritage the past millenia. Some shared ancestry for sure, but not significant amount of it.

[J. Ketch]

You are one individual, why would I give more attention to where you plot when there's an Irish average?
It does not take a top notch geneticist to see that the aDNA samples we have from Britain so far aren't like modern Northern French. And you keep saying that it's Northern French but a significant amount of it is also Southern and Central French. Aswell as 5-10% Spanish.
But let's see if I understand your take on the British study correctly. You think that the Scandinavian shown in the ADMIXTURE run of the Irish is indicative of actual Scandinavian heritage, dismissing the fact that the more "southern" source in their run is French/Northern French, a bad proxy for how Ireland looked like pre-Viking era. You don't see the major flaw? If Northern French are significantly more southern shifted than the Roman era British samples then it's going to compensate with giving more Northern-like admixture ie Scandinavian(mainly Northern Scandinavian to be precise).

Obviously I believe that modern Brits and Scandinavians are very closely related autosomally speaking, I've argued for it many times here. But that doesn't necessarily mean that they have a shared heritage the past millenia. Some shared ancestry for sure, but not significant amount of it.
How can one individual be so far from the average though? If this is distant ancestry that should be relatively uniform across Ireland, is it false when someone gets 10-20+% more Nordic admixture than the average across calculators? The Irish can't be that diverse.

[Peterski]

Aren,

Germanic Y-DNA is ca. 10% and possibly up to 15% of Irish Y-DNA. Of course not all of it is Scandinavian, also Anglo-Saxon.

On the other hand, the English could bring also Celtic Y-DNA with them to Ireland (such as R1b-L21 lineages from England).

[Grace O'Malley]

You are one individual, why would I give more attention to where you plot when there's an Irish average?
It does not take a top notch geneticist to see that the aDNA samples we have from Britain so far aren't like modern Northern French. And you keep saying that it's Northern French but a significant amount of it is also Southern and Central French. Aswell as 5-10% Spanish.
But let's see if I understand your take on the British study correctly. You think that the Scandinavian shown in the ADMIXTURE run of the Irish is indicative of actual Scandinavian heritage, dismissing the fact that the more "southern" source in their run is French/Northern French, a bad proxy for how Ireland looked like pre-Viking era. You don't see the major flaw? If Northern French are significantly more southern shifted than the Roman era British samples then it's going to compensate with giving more Northern-like admixture ie Scandinavian(mainly Northern Scandinavian to be precise).

Obviously I believe that modern Brits and Scandinavians are very closely related autosomally speaking, I've argued for it many times here. But that doesn't necessarily mean that they have a shared heritage the past millenia. Some shared ancestry for sure, but not significant amount of it.

I was asking specifically about the Irish not the Brits? The reason why I queried about where I plot is because you appear to place a lot of credence in that plot. My mother plots in the same place as I do as well. There would be other Irish as well that would plot there.

The French component that they used for the Irish is majority Northern French. I understand what you have said about the Bronze Age samples being further north than the French but we should get a fuller picture when more ancient genome studies are released and there is one expected in the next year by Lara Cassidy with more ancient samples from Ireland so that will give a fuller picture. The people involved in the Insular Celtic paper (i.e. Prof Dan Bradley and Lara Cassidy) have ancient genomes at their disposal and did compare the Irish to both Rathlin and Ballynahatty. I find it hard to believe that the Geneticists are less knowledgeable than amateurs like yourself. Both Prof Bradley and Lara Cassidy were involved in the Rathlin study. We don't have the full picture like they do. They have genomes from every age in Ireland so I would suspect they would have tested these genomes. As I've said Lara Cassidy has put a paper forward for peer review so it should be published sometime next year.

I've enjoyed this discussion with yourself because it is always good to have different opinions. Some of these geneticists give quite a few talks and and a few people on Anthrogenica have attended these talks. I'm sure they are aware of what the ancient genomes are like as they wrote the papers. I'd be surprised if they haven't considered all of this. I find it hard to believe that so many scientists aren't aware of what you are discussing.

This is what the Insular Celtic paper said about the Scandinavian stuff.


Of all the European populations considered, ancestral influence in Irish genomes was best represented by modern Scandinavians and northern Europeans, with a significant single-date one-source admixture event overlapping the historical period of the Norse-Viking settlements in Ireland (p < 0.01; fit quality FQB > 0.985; Fig 6). This was recapitulated to varying degrees in specific genetically- and geographically-defined groups within Ireland, with the strongest signals in south and central Leinster (the largest recorded Viking settlement in Ireland was Dubh linn in present-day Dublin), followed by Connacht and north Leinster/Ulster (S5 Fig; S6 Table).

It will be interesting to compare the paper that is coming out with more samples from Scotland that Ed Gilbert is doing who was heavily involved in the Irish DNA Atlas.

Also if you dismiss the Scandinavian input in the Irish do you also dismiss the Scandinavian input in Orkney?

Personally I don't see why it is so difficult to believe that Vikings would leave some genetics behind. They have settlements in Dublin, Cork, Limerick, Wexford etc and were all over Ireland. It is more difficult to believe they that didn't leave any dna.

Anyway there will be more studies coming out and there is actually a Viking study coming out from Denmark where they have actual Viking dna and they are looking for countries with Viking descendants so I'm sure some of these questions will be more thoroughly answered as more of these studies come out.

[Peterski]

If this is distant ancestry that should be relatively uniform across Ireland

Why should it be uniform with no regional differences at all?

Most people married within their county in the past - usually marriages were between people from neighbouring villages/towns.

Only relatively recently, long-distance mobility greatly intensified.

Even today 35% of users on this forum have grandparents born within 80 kilometers from each other. Check my thread:

https://www.theapricity.com/forum/showthread.php?234514-Were-all-4-of-your-grandparents-born-within-80-km-(50-mi)-of-each-other

[Grace O'Malley]

Why should it be uniform with no regional differences at all?

Most people married within their county in the past - usually marriages were between people from neighbouring villages/towns.

Only relatively recently, long-distance mobility greatly intensified.

Even today 35% of users on this forum have grandparents born within 80 kilometers from each other. Check my thread:

https://www.theapricity.com/forum/showthread.php?234514-Were-all-4-of-your-grandparents-born-within-80-km-(50-mi)-of-each-other

Interestingly though that Danish match is paternal for me and not through my mother. My mother has other Scandinavian matches and has some I don't have.

[Peterski]

Interestingly though that Danish match is paternal for me and not through my mother. My mother has other Scandinavian matches and has some I don't have.

How close is this match, how many cM?

[Aren]

How can one individual be so far from the average though? If this is distant ancestry that should be relatively uniform across Ireland, is it false when someone gets 10-20+% more Nordic admixture than the average across calculators? The Irish can't be that diverse.
So far away? I hardly think that's far away in a plot with 36K. Scoring 10-20% of a very related population is very normal. Many pure Swedes can get something like 70% Swedish + 30% Scottish on Gedmatch it's very normal since these populations are very close to one and other. In other words she's well within the variation. There are Irish who plot on the opposite end aswell.

Aren,

Germanic Y-DNA is ca. 10% and possibly up to 15% of Irish Y-DNA. Of course not all of it is Scandinavian, also Anglo-Saxon.

On the other hand, the English could bring also Celtic Y-DNA with them to Ireland (such as R1b-L21 lineages from England).
10% is not much considering how skewed towards the male populations the admix must've been at that time. We can't rule out a signficant portion of it coming from only partially Germanic people such as the English. Especially much of the U106 and R1a-L664.

[J. Ketch]

Why should it? Most people married within their county in the past.

Only relatively recently long-distance mobility greatly intensified.
It's a small island with no real natural barriers, a supposedly minor (5-10%) admixture from over 1000 years ago. There shouldn't be a great variance therefore across Ireland. But my father gets ridiculous Nordic scores across Gedmatch and G25, while getting basically 100% Irish in the commercial tests. I don't understand it.

[Peterski]

how skewed towards the male populations the admix must've been at that time.

In Scotland it was pretty balanced, about as much Scandinavian Y-DNA as their mtDNA (so Viking women also settled):

https://www.researchgate.net/figure/Map-showing-the-proportions-of-Scandinavian-and-British-Irish-ancestry-for-mtDNA-Mt-and_fig1_7920315

https://www.researchgate.net/profile/Sara_Goodacre/publication/7920315/figure/fig1/AS:277873870032899@1443261701559/Map-showing-the-proportions-of-Scandinavian-and-British-Irish-ancestry-for-mtDNA-Mt-and.png

https://www.researchgate.net/profile/Sara_Goodacre/publication/7920315/figure/tbl1/AS:669125294686258@1536543310973/and-British-Irish-patrilineal-and-matrilineal-ancestry-based-on-the-mr-procedure-for.png

[Aren]

In Scotland it was pretty balanced, about as much Scandinavian Y-DNA as their mtDNA (so Viking women also settled):

https://www.researchgate.net/figure/Map-showing-the-proportions-of-Scandinavian-and-British-Irish-ancestry-for-mtDNA-Mt-and_fig1_7920315

https://www.researchgate.net/profile/Sara_Goodacre/publication/7920315/figure/fig1/AS:277873870032899@1443261701559/Map-showing-the-proportions-of-Scandinavian-and-British-Irish-ancestry-for-mtDNA-Mt-and.png
You mean NW Scottish coast? Mixing with people from the Shetlands and Orkney most likely. I doubt that's the case for the Lowlands. Look at the Western Isles much less Scandinavian mtdna. Shetlands and Orkney obviously have significant amount of genuine Viking era(and also from the Middle ages for sure) Scandinavian ancestry.
This is from 2005 though, how well would they have been able to tell apart Scandinavian and Celtic uniparental markers at that time?

[J. Ketch]

So far away? I hardly think that's far away in a plot with 36K. Scoring 10-20% of a very related population is very normal. Many pure Swedes can get something like 70% Swedish + 30% Scottish on Gedmatch it's very normal since these populations are very close to one and other. In other words she's well within the variation. There are Irish who plot on the opposite end aswell.

Is this within the variation?
https://i.postimg.cc/6Qt14Jdw/k36dad.jpg

[Aren]

Is this within the variation?
https://i.postimg.cc/6Qt14Jdw/k36dad.jpg

Depends. Can you post his scores?

[J. Ketch]

Depends. Can you post his scores?
K36 full nmonte

"distance%=0.471"
Scotland,40.8
Mecklenburg-Vorpommern,26
NL_Utrecht,17.2
Sachsen-Anhalt,6
Oberland(western_Ost_Preussen),4.2
Ireland,3
NW_England,1.2
Sathmarer_Schwaben,0.6
Mordva_Moksha,0.4
Pl_Upper_Silesia,0.2
Ukraine_Cherkassy,0.2
Ukraine_NE,0.2

Eurogenes K13

# Population Percent
1 North_Atlantic 53.57
2 Baltic 26.31
3 West_Med 10.06
4 West_Asian 4.17
5 East_Med 2.9
6 South_Asian 1.7
7 Sub-Saharan 0.63
8 Amerindian 0.48
9 Siberian 0.12
10 Oceanian 0.06

Single Population Sharing:

# Population (source) Distance
1 Danish 3.51
2 North_Dutch 3.58
3 Norwegian 3.69
4 Orcadian 3.84
5 West_Scottish 3.91
6 Irish 4.09
7 Southeast_English 5.63
8 Southwest_English 6.27
9 North_German 6.38
10 Swedish 6.52
11 South_Dutch 10.97
12 West_German 11.93
13 North_Swedish 12.71
14 East_German 15.96
15 Austrian 16.09
16 French 16.51
17 Hungarian 20.39
18 Southwest_Finnish 21
19 Spanish_Cataluna 23.49
20 Southwest_French 24.51

Mixed Mode Population Sharing:

# Primary Population (source) Secondary Population (source) Distance
1 52.8% Norwegian + 47.2% West_Scottish @ 2.62
2 56.6% Danish + 43.4% West_Scottish @ 2.85
3 68.9% West_Scottish + 31.1% Swedish @ 2.88
4 52.6% Norwegian + 47.4% Orcadian @ 2.97
5 83.1% West_Scottish + 16.9% North_Swedish @ 2.99
6 71.9% Norwegian + 28.1% Southwest_English @ 2.99
7 56.6% Norwegian + 43.4% Irish @ 3.01
8 69.3% Norwegian + 30.7% Southeast_English @ 3.03
9 89.9% West_Scottish + 10.1% Southwest_Finnish @ 3.13
10 91.9% West_Scottish + 8.1% La_Brana-1 @ 3.13

Using 2 populations approximation:
1 50% Norwegian +50% West_Scottish @ 2.908939

Eurogenes K15

1 North_Sea 36.75
2 Atlantic 30.91
3 Eastern_Euro 11.47
4 Baltic 11.44
5 West_Med 6.01
6 West_Asian 2.14
7 South_Asian 0.73
8 Sub-Saharan 0.55

Single Population Sharing:

# Population (source) Distance
1 West_Scottish 3.02
2 Danish 3.5
3 Irish 3.54
4 North_Dutch 4.53
5 Southeast_English 4.82
6 Orcadian 5.22
7 North_German 5.95
8 Southwest_English 6.46
9 Norwegian 7.18
10 West_Norwegian 7.65
11 Swedish 8.43
12 South_Dutch 9.78
13 North_Swedish 10.43
14 West_German 11.5
15 East_German 14.18
16 French 14.66
17 Southwest_Finnish 14.85
18 Austrian 18.13
19 Finnish 18.5
20 Hungarian 19.17

Mixed Mode Population Sharing:

# Primary Population (source) Secondary Population (source) Distance
1 89.3% West_Scottish + 10.7% Southwest_Finnish @ 2.46
2 93.5% West_Scottish + 6.5% Estonian @ 2.51
3 94.4% West_Scottish + 5.6% Kargopol_Russian @ 2.51
4 93.2% West_Scottish + 6.8% East_Finnish @ 2.52
5 94.3% West_Scottish + 5.7% Estonian_Polish @ 2.52
6 95% West_Scottish + 5% Lithuanian @ 2.56
7 94.5% West_Scottish + 5.5% Belorussian @ 2.56
8 95.3% West_Scottish + 4.7% Erzya @ 2.56
9 93.7% West_Scottish + 6.3% Polish @ 2.58
10 92.3% West_Scottish + 7.7% Finnish @ 2.6

Using 2 populations approximation:
1 50% Danish +50% West_Scottish @ 3.254816

EUTEST

# Population Percent
1 ATLANTIC 29.4
2 NORTH-CENTRAL_EURO 28.67
3 SOUTH_BALTIC 14.9
4 EAST_EURO 13.24
5 WEST_MED 8.39
6 WEST_ASIAN 3.58
7 EAST_MED 1.55
8 SOUTH_ASIAN 0.27

Single Population Sharing:

# Population (source) Distance
1 Orcadian 2.76
2 Scottish 3.88
3 IE 3.92
4 DK 4.4
5 NO 5.02
6 English 5.25
7 Cornish 5.44
8 NL 5.54
9 South_&_Central_Swedish 5.8
10 West_&_Central_German 6.45
11 North_Swedish 9.38
12 AT 13.52
13 FR 14.59
14 South_Finnish 18.24
15 HU 18.34
16 Serbian 21.37
17 PL 21.39
18 PT 21.95
19 ES 22.12
20 EE 23.12

Mixed Mode Population Sharing:

# Primary Population (source) Secondary Population (source) Distance
1 60.3% Scottish + 39.7% South_&_Central_Swedish @ 0.84
2 56.9% Scottish + 43.1% NO @ 1.17
3 72.4% Scottish + 27.6% North_Swedish @ 1.62
4 72.1% Orcadian + 27.9% South_&_Central_Swedish @ 1.76
5 69.7% Orcadian + 30.3% NO @ 1.88
6 82% Orcadian + 18% North_Swedish @ 1.89
7 54% Scottish + 46% DK @ 1.89
8 92.4% Orcadian + 7.6% EE @ 2.02
9 90.8% Orcadian + 9.2% South_Finnish @ 2.06
10 61.9% IE + 38.1% South_&_Central_Swedish @ 2.07
11 85% Scottish + 15% South_Finnish @ 2.18
12 94.4% Orcadian + 5.6% LIT @ 2.18
13 93.4% Orcadian + 6.6% East_Finnish @ 2.22
14 70.6% Orcadian + 29.4% DK @ 2.28
15 94.1% Orcadian + 5.9% Belorussian @ 2.28
16 58.4% IE + 41.6% NO @ 2.28
17 88% Scottish + 12% EE @ 2.29
18 94.7% Orcadian + 5.3% Northwest_Russian @ 2.29
19 95.4% Orcadian + 4.6% North_Russian @ 2.41
20 94% Orcadian + 6% PL @ 2.41

Using 2 populations approximation:
1 50% Scottish +50% South_&_Central_Swedish @ 1.420826

[Grace O'Malley]

How close is this match, how many cM?

I've sent you a PM.

[Aren]

K36 full nmonte

"distance%=0.471"
Scotland,40.8
Mecklenburg-Vorpommern,26
NL_Utrecht,17.2
Sachsen-Anhalt,6
Oberland(western_Ost_Preussen),4.2
Ireland,3
NW_England,1.2
Sathmarer_Schwaben,0.6
Mordva_Moksha,0.4
Pl_Upper_Silesia,0.2
Ukraine_Cherkassy,0.2
Ukraine_NE,0.2

Eurogenes K13

# Population Percent
1 North_Atlantic 53.57
2 Baltic 26.31
3 West_Med 10.06
4 West_Asian 4.17
5 East_Med 2.9
6 South_Asian 1.7
7 Sub-Saharan 0.63
8 Amerindian 0.48
9 Siberian 0.12
10 Oceanian 0.06

Single Population Sharing:

# Population (source) Distance
1 Danish 3.51
2 North_Dutch 3.58
3 Norwegian 3.69
4 Orcadian 3.84
5 West_Scottish 3.91
6 Irish 4.09
7 Southeast_English 5.63
8 Southwest_English 6.27
9 North_German 6.38
10 Swedish 6.52
11 South_Dutch 10.97
12 West_German 11.93
13 North_Swedish 12.71
14 East_German 15.96
15 Austrian 16.09
16 French 16.51
17 Hungarian 20.39
18 Southwest_Finnish 21
19 Spanish_Cataluna 23.49
20 Southwest_French 24.51

Mixed Mode Population Sharing:

# Primary Population (source) Secondary Population (source) Distance
1 52.8% Norwegian + 47.2% West_Scottish @ 2.62
2 56.6% Danish + 43.4% West_Scottish @ 2.85
3 68.9% West_Scottish + 31.1% Swedish @ 2.88
4 52.6% Norwegian + 47.4% Orcadian @ 2.97
5 83.1% West_Scottish + 16.9% North_Swedish @ 2.99
6 71.9% Norwegian + 28.1% Southwest_English @ 2.99
7 56.6% Norwegian + 43.4% Irish @ 3.01
8 69.3% Norwegian + 30.7% Southeast_English @ 3.03
9 89.9% West_Scottish + 10.1% Southwest_Finnish @ 3.13
10 91.9% West_Scottish + 8.1% La_Brana-1 @ 3.13

Using 2 populations approximation:
1 50% Norwegian +50% West_Scottish @ 2.908939

Eurogenes K15

1 North_Sea 36.75
2 Atlantic 30.91
3 Eastern_Euro 11.47
4 Baltic 11.44
5 West_Med 6.01
6 West_Asian 2.14
7 South_Asian 0.73
8 Sub-Saharan 0.55

Single Population Sharing:

# Population (source) Distance
1 West_Scottish 3.02
2 Danish 3.5
3 Irish 3.54
4 North_Dutch 4.53
5 Southeast_English 4.82
6 Orcadian 5.22
7 North_German 5.95
8 Southwest_English 6.46
9 Norwegian 7.18
10 West_Norwegian 7.65
11 Swedish 8.43
12 South_Dutch 9.78
13 North_Swedish 10.43
14 West_German 11.5
15 East_German 14.18
16 French 14.66
17 Southwest_Finnish 14.85
18 Austrian 18.13
19 Finnish 18.5
20 Hungarian 19.17

Mixed Mode Population Sharing:

# Primary Population (source) Secondary Population (source) Distance
1 89.3% West_Scottish + 10.7% Southwest_Finnish @ 2.46
2 93.5% West_Scottish + 6.5% Estonian @ 2.51
3 94.4% West_Scottish + 5.6% Kargopol_Russian @ 2.51
4 93.2% West_Scottish + 6.8% East_Finnish @ 2.52
5 94.3% West_Scottish + 5.7% Estonian_Polish @ 2.52
6 95% West_Scottish + 5% Lithuanian @ 2.56
7 94.5% West_Scottish + 5.5% Belorussian @ 2.56
8 95.3% West_Scottish + 4.7% Erzya @ 2.56
9 93.7% West_Scottish + 6.3% Polish @ 2.58
10 92.3% West_Scottish + 7.7% Finnish @ 2.6

Using 2 populations approximation:
1 50% Danish +50% West_Scottish @ 3.254816

EUTEST

# Population Percent
1 ATLANTIC 29.4
2 NORTH-CENTRAL_EURO 28.67
3 SOUTH_BALTIC 14.9
4 EAST_EURO 13.24
5 WEST_MED 8.39
6 WEST_ASIAN 3.58
7 EAST_MED 1.55
8 SOUTH_ASIAN 0.27

Single Population Sharing:

# Population (source) Distance
1 Orcadian 2.76
2 Scottish 3.88
3 IE 3.92
4 DK 4.4
5 NO 5.02
6 English 5.25
7 Cornish 5.44
8 NL 5.54
9 South_&_Central_Swedish 5.8
10 West_&_Central_German 6.45
11 North_Swedish 9.38
12 AT 13.52
13 FR 14.59
14 South_Finnish 18.24
15 HU 18.34
16 Serbian 21.37
17 PL 21.39
18 PT 21.95
19 ES 22.12
20 EE 23.12

Mixed Mode Population Sharing:

# Primary Population (source) Secondary Population (source) Distance
1 60.3% Scottish + 39.7% South_&_Central_Swedish @ 0.84
2 56.9% Scottish + 43.1% NO @ 1.17
3 72.4% Scottish + 27.6% North_Swedish @ 1.62
4 72.1% Orcadian + 27.9% South_&_Central_Swedish @ 1.76
5 69.7% Orcadian + 30.3% NO @ 1.88
6 82% Orcadian + 18% North_Swedish @ 1.89
7 54% Scottish + 46% DK @ 1.89
8 92.4% Orcadian + 7.6% EE @ 2.02
9 90.8% Orcadian + 9.2% South_Finnish @ 2.06
10 61.9% IE + 38.1% South_&_Central_Swedish @ 2.07
11 85% Scottish + 15% South_Finnish @ 2.18
12 94.4% Orcadian + 5.6% LIT @ 2.18
13 93.4% Orcadian + 6.6% East_Finnish @ 2.22
14 70.6% Orcadian + 29.4% DK @ 2.28
15 94.1% Orcadian + 5.9% Belorussian @ 2.28
16 58.4% IE + 41.6% NO @ 2.28
17 88% Scottish + 12% EE @ 2.29
18 94.7% Orcadian + 5.3% Northwest_Russian @ 2.29
19 95.4% Orcadian + 4.6% North_Russian @ 2.41
20 94% Orcadian + 6% PL @ 2.41

Using 2 populations approximation:
1 50% Scottish +50% South_&_Central_Swedish @ 1.420826

I meant the K36 values. But going by Lukasz nMonte it seems more so East-Central Euro(Mecklenburg, Saxony-Anhalt) pull rather than Scandinavian which makes him plot with Dutch.