Eurogenes K36 admixtures of Hinxton genomes ( Iron Age Britons & Anglo-Saxons)

[Maciamo]

The Hinxton genomes were released in October 2014 and I already analysed their admixtures using the Dodecad dv3 and K12b calculators at the time.

Here is another look at them using the Eurogenes K36 admixtures instead. It's a good opportunity to see which component of K36 correlate most with Celtic or Germanic (Anglo-Saxon) ancestry.

Keep in mind that the component names for the Eurogenes K36 admixture are not accurate and do not represent the region of origin, nor even the region where it is the most common today. One of the reasons fro comparing ancient genomes is to try to determine what these components really mean. For more information check the analysis of the K36 components.


Hinxton 1 (Iron-age Celt, R1b-L21, K1a1b1b)

0.00% Amerindian
0.00% Arabian
0.00% Armenian
0.18% Basque
0.00% Central_African
6.95% Central_Euro
0.00% East_African
0.00% East_Asian
2.83% East_Balkan
0.00% East_Central_Asian
0.11% East_Central_Euro
0.00% East_Med
8.63% Eastern_Euro
7.28% Fennoscandian
7.69% French
20.20% Iberian
0.00% Indo-Chinese
5.01% Italian
0.00% Malayan
0.00% Near_Eastern
0.09% North_African
18.73% North_Atlantic
1.49% North_Caucasian
20.11% North_Sea
0.00% Northeast_African
0.00% Oceanian
0.00% Omotic
0.00% Pygmy
0.00% Siberian
0.00% South_Asian
0.00% South_Central_Asian
0.00% South_Chinese
0.71% Volga-Ural
0.00% West_African
0.00% West_Caucasian
0.00% West_Med

Hinxton 4 (Iron Age Celt, R1b-L21, H1ag1)

0.00% Amerindian
0.00% Arabian
0.00% Armenian
1.01% Basque
0.00% Central_African
8.46% Central_Euro
0.00% East_African
0.00% East_Asian
0.24% East_Balkan
0.00% East_Central_Asian
6.52% East_Central_Euro
0.00% East_Med
1.22% Eastern_Euro
12.00% Fennoscandian
10.18% French
18.53% Iberian
0.00% Indo-Chinese
5.53% Italian
0.00% Malayan
0.00% Near_Eastern
0.00% North_African
19.48% North_Atlantic
3.90% North_Caucasian
12.92% North_Sea
0.00% Northeast_African
0.00% Oceanian
0.00% Omotic
0.00% Pygmy
0.00% Siberian
0.00% South_Asian
0.00% South_Central_Asian
0.00% South_Chinese
0.00% Volga-Ural
0.00% West_African
0.00% West_Caucasian
0.01% West_Med


Hinxton 2 (Anglo-Saxon, H2a2b)

0.00% Amerindian
0.00% Arabian
0.00% Armenian
2.78% Basque
0.00% Central_African
7.27% Central_Euro
0.00% East_African
0.00% East_Asian
0.04% East_Balkan
0.00% East_Central_Asian
2.25% East_Central_Euro
0.00% East_Med
5.78% Eastern_Euro
10.38% Fennoscandian
5.66% French
0.01% Iberian
0.00% Indo-Chinese
10.57% Italian
0.00% Malayan
0.00% Near_Eastern
0.00% North_African
24.92% North_Atlantic
0.00% North_Caucasian
27.69% North_Sea
0.00% Northeast_African
0.00% Oceanian
0.00% Omotic
0.00% Pygmy
0.00% Siberian
0.00% South_Asian
0.74% South_Central_Asian
0.00% South_Chinese
1.90% Volga-Ural
0.00% West_African
0.00% West_Caucasian
0.00% West_Med


Hinxton 3 (Anglo-Saxon, K1a4a1a2)

0.00% Amerindian
0.00% Arabian
0.01% Armenian
4.67% Basque
1.04% Central_African
8.10% Central_Euro
0.00% East_African
0.00% East_Asian
0.00% East_Balkan
0.00% East_Central_Asian
0.25% East_Central_Euro
0.00% East_Med
0.18% Eastern_Euro
11.54% Fennoscandian
6.34% French
5.18% Iberian
0.00% Indo-Chinese
2.77% Italian
0.00% Malayan
0.00% Near_Eastern
0.02% North_African
21.30% North_Atlantic
2.22% North_Caucasian
31.83% North_Sea
0.00% Northeast_African
0.00% Oceanian
0.00% Omotic
0.00% Pygmy
0.00% Siberian
0.00% South_Asian
1.04% South_Central_Asian
0.00% South_Chinese
0.24% Volga-Ural
0.01% West_African
3.25% West_Caucasian
0.01% West_Med

Hinxton 5 (Anglo-Saxon, H2a2a)

0.00% Amerindian
0.00% Arabian
0.00% Armenian
3.35% Basque
0.00% Central_African
6.27% Central_Euro
0.00% East_African
0.00% East_Asian
3.06% East_Balkan
0.00% East_Central_Asian
6.16% East_Central_Euro
0.00% East_Med
3.65% Eastern_Euro
12.55% Fennoscandian
6.15% French
15.42% Iberian
0.00% Indo-Chinese
6.28% Italian
0.00% Malayan
0.00% Near_Eastern
0.94% North_African
15.76% North_Atlantic
1.05% North_Caucasian
19.35% North_Sea
0.00% Northeast_African
0.00% Oceanian
0.00% Omotic
0.00% Pygmy
0.00% Siberian
0.00% South_Asian
0.00% South_Central_Asian
0.00% South_Chinese
0.00% Volga-Ural
0.00% West_African
0.00% West_Caucasian
0.00% West_Med

The results are not at all what one could expect based on the region's names. For example:

- All three have about as much Fennoscandian. Why would one of the Celtic Briton score as high as the Anglo-Saxons?
- There are huge discrepancies between Anglo-Saxon samples. One of them has considerably relatively high percentages of Basque and West Caucasian, for unexplained reasons.
- I expected was higher North_Sea in the Anglo-Saxons and higher North_Atlantic in the Celtic Britons. It's true for Anglo-Saxons, but one Celt also has more North_Sea than North_Atlantic.
- The Dodecad admixtures showed 7% and 3.5% of African in Hinxton1 and Hinxton2. These have disappeared here, but Hinxton3 and Hinxron4 still have 1% of it. It could be a misidentified or extinct admixture.

So it looks like there was still a lot of genetic heterogeneity 1300 years ago, much more than in the modern population.

[Bollox79]

Maciamo... would it be at all possible to run the Driffield guys... in particular perhaps 6drif-3 and 3drif-16 (both U106 and Z304-307 under that and 6rif-3 is DF98+ and 3drif-16 is DF96+) to see how close they are? Are their genomes available from the paper in Jan of 2016? I figure they have ancestry from the same tribe/tribes... both showing affinity with the Baltic (aka Lithuanian pop) more than the other Driffield samples tested at Trinity in Dublin... just curious!

Cheers!

[Maciamo]

Maciamo... would it be at all possible to run the Driffield guys... in particular perhaps 6drif-3 and 3drif-16 (both U106 and Z304-307 under that and 6rif-3 is DF98+ and 3drif-16 is DF96+) to see how close they are? Are their genomes available from the paper in Jan of 2016? I figure they have ancestry from the same tribe/tribes... both showing affinity with the Baltic (aka Lithuanian pop) more than the other Driffield samples tested at Trinity in Dublin... just curious!

Cheers!

Yes, I could. I just have to check if the genomes in the paper are already in .txt or .csv format. Otherwise I will need to find a way to convert them.

EDIT: I checked but the files are over 1 GB in size, so it would take forever to run them in the calculator.

[Bollox79]

Ok... thanks for checking though !

[MOESAN]

The Hinxton genomes were released in October 2014 and I already analysed their admixtures using the Dodecad dv3 and K12b calculators at the time.

Here is another look at them using the Eurogenes K36 admixtures instead. It's a good opportunity to see which component of K36 correlate most with Celtic or Germanic (Anglo-Saxon) ancestry.

The results are not at all what one could expect based on the region's names. For example:

- All three have about as much Fennoscandian. Why would a Celtic Briton score as high as Anglo-Saxons?
- There are huge discrepancies between Anglo-Saxon samples. One of them has considerably relatively high percentages of Basque and West Caucasian, for unexplained reasons. Hinxton1 has 20% Iberian, 4x more than the other Anglo-Saxon.
- I expected was higher North_Sea in the Anglo-Saxon and higher North_Atlantic in the Celtic Briton, but three have similar North_Atlantic, and only one Anglo-Saxon scores considerably higher than either the Celts or the other Anglo-Saxon. One Celt has more North_Sea than North_Atlantic, while the other Celt has just the opposite.
- The Dodecad admixtures showed 7% and 3.5% of Sub-Saharan African in Hinxton1 and Hinxton2. These have disappeared here, but Hinxton3 still has 1% of it. It could be a misidentified or extinct admixture.

So it looks like there was still a lot of genetic heterogeneity 1300 years ago, much more than in the modern population.

I'm sorry I have not been serious and never red the archeological aspect of this surveys; have we precise clues about the class level of these men? Because I still suppose the elites were more heterogenous concerning auDNA than the basic freemen?

[Maciamo]

I'm sorry I have not been serious and never red the archeological aspect of this surveys; have we precise clues about the class level of these men? Because I still suppose the elites were more heterogenous concerning auDNA than the basic freemen?

Why would you think that? It is true for later European monarchies, where royal families traded family members between countries. But the Anglo-Saxons didn't intermix much with foreigners, and neither would a Roman-age Briton. I also doubt that the Hinxton genomes came from royalties. It is certainly not the case for the two Iron-Age Celts as they might have lived under Roman rule. Perhaps someone has more information about their presumed social background?

[Tomenable]

- There are huge discrepancies between Anglo-Saxon samples. One of them has considerably relatively high percentages of Basque and West Caucasian, for unexplained reasons. Hinxton1 has 20% Iberian, 4x more than the other Anglo-Saxon.

This is probably because some of Anglo-Saxon samples were already of mixed Saxon-Briton descent.

Or - alternatively - they could have ancestry from different continental groups, rather than one group.

BTW, have you tried placing these samples on a map (compared to modern populations) using this tool?:

http://gen3553.pagesperso-orange.fr/ADN/Europe.htm

http://www.eupedia.com/forum/threads...s-K36-results)

EDIT: I checked but the files are over 1 GB in size, so it would take forever to run them in the calculator.

Where did you find the files?

[Tomenable]

Hinxton 2 (Iron Age Celtic genome):

Genetically predicted Latitude: 54.54749626200001
Genetically predicted Longitude: -1.3658863080000039

And this is a location in North-Eastern England, between Darlington and Middlesbrough:

https://www.google.pl/maps/place/54°...3!4d-1.3658863

Compare to Hinxton: https://www.google.pl/maps/place/Hin...2869!4d0.18269

Today people in North-Eastern England have more of Celtic ancestry than people in South-Eastern England.

So the fact that a Celt from Hinxton (South-Eastern England) was genetically predicted as a Northern English person (Darlington) is consistent with Northern English people being more similar to Hinxton Celts due to having less of Saxon ancestry.

===================

Later I will check other samples for genetically predicted location.

If Saxon samples were unmixed, then I expect them to "land" somewhere outside of Britain (in the North Sea?).

Modern Englishmen are autosomally only 1/3 Germanic (and more in terms of Y-DNA, but less in terms of mtDNA).

[Tomenable]

Genomes of unmixed Anglo-Saxon immigrants should be genetically predicted as either people from somewhere in South-Eastern or Easternmost England (which is the most Anglo-Saxon part of England in terms of ancestry), or as people from somewhere outside of Britain. Maybe they will be predicted in the North Sea, or even in Denmark, or in Northern Netherlands.

According to Eurogenes K15, Hinxton Anglo-Saxons are most genetically similar to Swedes and Norwegians.

But I am not sure if Danish and Dutch people are among reference populations in that calculator.

=============

In Eurogenes K15 calculator, Celts have more of "Atlantic", and Anglo-Saxons more of "North Sea":

1) Hinxton Iron Age Celtic Briton - 38% North Sea / 30% Atlantic / 6.5% Mediterranean

2) Hinxton Early Anglo-Saxon - 41.5% North Sea / 28.5% Atlantic / 6.5% Mediterranean

3) Modern Englishman from Kent - 35.5% North Sea / 30% Atlantic / 11.5% Mediterranean

[Tomenable]

By the way, Early Bronze Age (Corded Ware culture) inhabitants of what is now Denmark, were most genetically similar to modern Norwegians (especially Western Norwegians), not to modern Danes. This is a very consistent pattern and it can't be a coincidence. This must also be the reason why R1a - associated with CW - is much more common in modern Norway, than it is in modern Denmark. Most of CW samples from Sweden were also most similar to modern Norwegians - not Swedes.

The only Early Bronze Age samples from Sweden most similar to modern Swedes was RISE98 (R1b-U106).

===============

Here are some ancient DNA Scandinavian / Fennoscandian samples on GEDmatch:

GEDmatch kit - sample (country):

M671253 - RISE71 (Denmark)
M130094 - RISE61 (Denmark)
F999956 - RISE94 (Sweden)
F999945 - RISE97 (Sweden)
F999941 - RISE98 (Sweden)
F999943 - RISE174 (Sweden)
F999955 - RISE00 (Estonia)

If I remember correctly, RISE71, 61, 94 and 97 were most similar to modern Norwegians.

Only RISE98 and RISE174 (but this one is only 1500 years old) were like modern Swedes.

====================

Some Central European samples on GEDmatch (later I will add some more to this list):

M733797 - RISE109 Unetice EBA Poland [1954-1772 BC] U4 -
M453254 - RISE154 Unetice EBA Poland [1925-1765 BC] K1a4a1 -
M239638 - RISE586 Unetice EBA Czech Republic [?] K1b1a -
M484113 - RISE139 Unetice Poland [2135-1923 BC] U2e1f1 n/a
M069028 - RISE145 Unetice Poland [2188-1958 BC] H6a1b -

M483824 - RISE598 Lithuania LBA [908-485 BC] H2a2 R1a-L62>M417

M669778 - RISE1 Corded Ware Poland [2865-2578 BC] K1b1a1 R1b-M343>L754
M191719 - RISE431 CW or Proto-Unetice Poland [2286-2048 BC] T2e R1a-M417

M631469 - RISE349 Hungary MBA [2034-1748 BC] T2b3 -
M375165 - RISE373 Maros Hungary [1886-1696 BC] K1a2a -
M974598 - RISE374 Maros Hungary [1866-1619 BC] T2 G2a-P287>P15>PF3177

Maros culture samples look like being mostly Neolithic-descended with not so much of Steppe ancestry.

Is Maros culture the same as Mures culture?:

https://en.wikipedia.org/wiki/Prehis....C5.9F_culture

[Tomenable]

Anglo-Saxon Hinxton 1:

Latitude: 52.643113694
Longitude: -2.7662528660000127

Anglo-Saxon Hinxton 3:

Latitude: 57.91280375000001
Longitude: -4.882640850000026

Surprising... One location is in Northern Scotland, the other one in Western England.

And according to the authors of the publication, they were not mixed with locals ???

To me it seems that they had to be mixed with locals. At least Hinxton 1.

[Maciamo]

Sorry I mistook in the OP. I wrote that Hinxton 1 was Anglo-Saxon and Hinxton 5 was Iron Age Celt, but it was the opposite. Please check again the names to be sure.

[Tomenable]

So Hinxton 1's genetically predicted location near the border of Wales actually makes sense. If he was Celtic.

[Northener]

@Tomenable, here my Northern Netherlands results, has similarities (but less divers as Maciamo stated)

K36.jpg
and K15

North_Sea	39.01
Atlantic	27.71
Baltic	12.19
Eastern_Euro	10.20
West_Med	5.86
West_Asian	2.80
East_Med	0.51
Red_Sea	-
South_Asian	1.02
Southeast_Asian	-
Siberian	-
Amerindian	-
Oceanian	0.48
Northeast_African	0.22
Sub-Saharan	-

Genomes of unmixed Anglo-Saxon immigrants should be genetically predicted as either people from somewhere in South-Eastern or Easternmost England (which is the most Anglo-Saxon part of England in terms of ancestry), or as people from somewhere outside of Britain. Maybe they will be predicted in the North Sea, or even in Denmark, or in Northern Netherlands.

According to Eurogenes K15, Hinxton Anglo-Saxons are most genetically similar to Swedes and Norwegians.

But I am not sure if Danish and Dutch people are among reference populations in that calculator.

=============

In Eurogenes K15 calculator, Celts have more of "Atlantic", and Anglo-Saxons more of "North Sea":

1) Hinxton Iron Age Celtic Briton - 38% North Sea / 30% Atlantic / 6.5% Mediterranean

2) Hinxton Early Anglo-Saxon - 41.5% North Sea / 28.5% Atlantic / 6.5% Mediterranean

3) Modern Englishman from Kent - 35.5% North Sea / 30% Atlantic / 11.5% Mediterranean