Eurasia K10 CHG Calculator

Angela

Elite member
Messages
21,823
Reaction score
12,327
Points
113
Ethnic group
Italian
The population averages have been posted.

You can find them here:

https://drive.google.com/file/d/0B-ObXiVfL-RzQzVxR0F0RDF1WlU/view

Does anyone know why on earth he doesn't have a northern Italian sample? I'm not going to be greedy and ask for four like one of the calculators provides, but would it be too much trouble to at least include the academic Bergamo sample?

Without it, the calculator is useless for northern Italians.
 
The population averages have been posted.

You can find them here:

https://drive.google.com/file/d/0B-ObXiVfL-RzQzVxR0F0RDF1WlU/view

Does anyone know why on earth he doesn't have a northern Italian sample? I'm not going to be greedy and ask for four like one of the calculators provides, but would it be too much trouble to at least include the academic Bergamo sample?

Without it, the calculator is useless for northern Italians.


That's not the only shortcoming. No Belgian, Dutch, German, Swiss, Austrian or Irish sample either. I can't make a map with such big holes over central and north-west Europe. Although there is a good chance that their CHG is all around 19% like the North French, English, Scots, Icelandic, Norwegians and Hungarians. A remarkable homogeneity overall.
 
Going in the right direction but still has the typical flaws. Still using Samara_Eneolithic as 100% EHG refference while Satsurbila is 88% CHG and 12% EHG(E_Samara). Eneolithic_Samara has Satsurbila like ancestry and not the opposite.

Again a good portion of CHG ancestry is getting eaten up by EHG. And again a whole chunk of EF ancestry is getting eaten up by WHG. Thats why populations such as Tajiks turn up with as much as 13%! WHG. And the Scythian sample which is said to be Andronovo like(20% EF) is turning out with 0% of it while having 37% extra WHG!.
He has more WHG than EHG and CHG because the Anatolian Farmer DNA is getting eaten up here.


This calculator has similar problems as Eurogenes K8 CHG.

That's not the only shortcoming. No Belgian, Dutch, German, Swiss, Austrian or Irish sample either. I can't make a map with such big holes over central and north-west Europe. Although there is a good chance that their CHG is all around 19% like the North French, English, Scots, Icelandic, Norwegians and Hungarians. A remarkable homogeneity overall.

Thats rather the smaller issues. The percentages are not accurate enough. All populations have by far more WHG and EHG as in reality. And the CHG frequency is too low.

We have Eurogenes K8 CHG where the average Northwest European has ~30% and here it is the opposite with ~20%. But looking at Jones et al. Something inbetween (25%) seems correct.

The calculators go in the right direction they are not but by any means but they have still flaws which turn some of the percentages completely upside down.

Since we have now "kurd" and "Davids" calculators I am wondering why we haven't seen any new calculator of "Dienekes", to see his results and how much they will differ.
 
Last edited:
That's not the only shortcoming. No Belgian, Dutch, German, Swiss, Austrian or Irish sample either. I can't make a map with such big holes over central and north-west Europe. Although there is a good chance that their CHG is all around 19% like the North French, English, Scots, Icelandic, Norwegians and Hungarians. A remarkable homogeneity overall.

You're right. Also, because of that the population assignment and fits are terrible. A lot of Northwest Europeans are coming out as Croatian or Hungarian. Northern Italians are coming out as Bulgarians or Spaniards, Tuscans as Albanians.

It's really not useful for Europeans.
 
That's not the only shortcoming. No Belgian, Dutch, German, Swiss, Austrian or Irish sample either. I can't make a map with such big holes over central and north-west Europe. Although there is a good chance that their CHG is all around 19% like the North French, English, Scots, Icelandic, Norwegians and Hungarians. A remarkable homogeneity overall.

I agree wih all comments on this thread ...................better to use the other new one on gedmatch

below is mine ..............far more logical for what I know about my line .............I just think he should not have used Spanish_basque, but just Basque to cover the french_basques

Bergamo with bulgarian is also in the Laz and haak data.



[h=2]puntDNAL K10 Ancient Oracle results:[/h]puntDNAL K10 Ancient Oracle



Admix Results (sorted):

#PopulationPercent
1WHG38.75
2ENF36.31
3CHG22.07
4ASI1.4
5Amerindian0.83
6Oceanian0.63

Single Population Sharing:

#Population (source)Distance
1Italian_Bergamo3.84
2Bulgarian6.08
3French8.97
4Tuscan9.02
5Spanish_Northeast9.38
6Spanish_Southwest9.59
7Albanian10.03
8Croatian10.06
9Greek10.1
10Hungarian12.61
11English_South13.9
12Czech16.06
13Sicilian17.2
14Scottish_West17.59
15Ashkenazi_Jew18.06
16Basque_Spanish19.79
17Ukrainian20.58
18Icelandic20.61
19Norwegian20.63
20Sephardic_Jew25.36

Mixed Mode Population Sharing:

# Primary Population (source)Secondary Population (source)Distance
1 73.7% Italian_Bergamo + 26.3% Croatian @ 1.46
2 69.4% Basque_Spanish + 30.6% Abkhasian @ 1.5
3 50.1% French + 49.9% Tuscan @ 1.55
4 78.2% Italian_Bergamo + 21.8% Hungarian @ 1.59
5 85.5% Italian_Bergamo + 14.5% Ukrainian @ 1.64
6 74.2% French + 25.8% Sephardic_Jew @ 1.68
7 88.2% Italian_Bergamo + 11.8% Belarusian @ 1.74
8 66.1% French + 33.9% Sicilian @ 1.75
9 90% Italian_Bergamo + 10% Lithuanian @ 1.77
10 62.5% Basque_Spanish + 37.5% Kumyk @ 1.77
11 83.7% Italian_Bergamo + 16.3% Scottish_West @ 1.78
12 88.3% Italian_Bergamo + 11.7% Mordovian @ 1.79
13 80.3% Italian_Bergamo + 19.7% English_South @ 1.8
14 89.4% Italian_Bergamo + 10.6% Russian @ 1.8
15 89.5% Italian_Bergamo + 10.5% Estonian @ 1.8
16 69.9% Tuscan + 30.1% Icelandic @ 1.83
17 85.9% Italian_Bergamo + 14.1% Norwegian @ 1.84
18 80.6% Spanish_Northeast + 19.4% Lezgin @ 1.84
19 54.6% Sicilian + 45.4% Icelandic @ 1.85
20 82.6% Italian_Bergamo + 17.4% Czech @ 1.86
 
This calculator is creating weird and inconsistent results. Only 58% Anatolia Neolithic for EEF? Almost 100% EHG for Poltvka? And people from closely related populations are scoring very differently. The trends make sense though.
 
This calculator is creating weird and inconsistent results. Only 58% Anatolia Neolithic for EEF? Almost 100% EHG for Poltvka? And people from closely related populations are scoring very differently. The trends make sense though.

Thats because, as I said above, a whole chunk of WHG like farmer DNA is getting eaten up by "WHG". Stuttgart should at least be ~90% EF. Even Sardinians who are said to be predominantly EF(~70%) are only 48% Farmer.

The trend is right but as I said especially for CHG, the usage of E_Samara as refference for EHG and that the WHG like portion(~50%) of EF is getting eaten up causes some inconsistency. I assume that Eurogenes K8 is also using E_Samara as refference for EHG.

However to be fair, some of these components, especially the once which look like a "merge" of two other "components" and therefore have some overlaps and are very close to each other, can be hard to be modeled perfectly by some simple calculator algorythm.
 
That's not the only shortcoming. No Belgian, Dutch, German, Swiss, Austrian or Irish sample either. I can't make a map with such big holes over central and north-west Europe. Although there is a good chance that their CHG is all around 19% like the North French, English, Scots, Icelandic, Norwegians and Hungarians. A remarkable homogeneity overall.
Yep, all north and central Europe seems to be uniform from 17 to 20% from Russia to GB, except Balts and Finns somewhat less. Probably from dilution of CHG after mixing with Uralic tribes later on.
It is sort of weird that CHG is so uniform through northern Europe.
 
Thats because, as I said above, a whole chunk of WHG like farmer DNA is getting eaten up by "WHG". Stuttgart should at least be ~90% EF. Even Sardinians who are said to be predominantly EF(~70%) are only 48% Farmer.

The trend is right but as I said especially for CHG, the usage of E_Samara as refference for EHG and that the WHG like portion(~50%) of EF is getting eaten up causes some inconsistency. I assume that Eurogenes K8 is also using E_Samara as refference for EHG.

However to be fair, some of these components, especially the once which look like a "merge" of two other "components" and therefore have some overlaps and are very close to each other, can be hard to be modeled perfectly by some simple calculator algorythm.
I have same doubts. All these recent amateurish calculators making me a bit uneasy into giving too much credibility to these numbers. However at least they can point to approximate trends of spread and relations between populations.
 
Yep, all north and central Europe seems to be uniform from 17 to 20% from Russia to GB, except Balts and Finns somewhat less. Probably from dilution of CHG after mixing with Uralic tribes later on.
It is sort of weird that CHG is so uniform through northern Europe.

Actually going by the original paper Estonians score the most of CHG in Northeast Europe. And average North/Northeast/Northwest is most likely ~25% since this calculator shows slightly less while Eurogenes K8 shows ~30% of average for North. Thats another of these signs of miscalculation.

North Europe is most likely ~35/25/20/20% EF/CHG/EHG/WHG. I doubt that much of pre Indo European WHG remained in Europe. Even the studies spoke about a "rise of WHG" with the incoming Indo Europeans.
 
I have same doubts. All these recent amateurish calculators making me a bit uneasy into giving too much credibility to these numbers. However at least they can point to approximate trends of spread and relations between populations.

Looking on all the calculators they seem to differ massively from each other. Therefore expect something inbetween to be right. So far puntDNAL K10 seems to be the most "accurate". But even this has some issue in comparison to the other calculators because it give a bit too much of EF ancestry. I am shown with 37 EF! In reality I should be around ~25%.

puntDNAL is showin slightly too much EF and too weak CHG

Eurogenes K8 CHG and Eurasia K10 CHG are showing too much WHG and EHG at the expense of especially EF (and to lesser extents CHG).
 
Last edited:
PuntDNAL k10 is crap because it lacks a South West Asian component. Plenty of MENAs are scoring inflated levels of ENF and EHG on there. It's obvious that 20% WHG admixed Anantolian farmers can't be used as a proxy for Red Sea ancestry in Georgians, Kurds, Iranians,....
 
PuntDNAL k10 is crap because it lacks a South West Asian component. Plenty of MENAs are scoring inflated levels of ENF and EHG on there. It's obvious that 20% WHG admixed Anantolian farmers can't be used as a proxy for Red Sea ancestry in Georgians, Kurds, Iranians,....

how important do you think the red sea is?

J1e covers the bulk of the red sea admixture anyway
 
how important do you think the red sea is?

J1e covers the bulk of the red sea admixture anyway

Exactly Red Sea component is too low in the North. It's around 3-9% . I suspect a South Farmer component in the North too. But it is rather around ~10-15%. And it overlaps with EF.

I had a mistypo. I am showed with ~37% EF I should however be around ~25%.
 
Last edited:
I asked on the other thread, but I'll try again here. Where is the spreadsheet for population averages for the newest Eurogenes calculator which is showing all this elevated CHG. How can it be evaluated if we can't see how the populations relate to one another.
 
I asked on the other thread, but I'll try again here. Where is the spreadsheet for population averages for the newest Eurogenes calculator which is showing all this elevated CHG. How can it be evaluated if we can't see how the populations relate to one another.

I posted it at the thread I started about Eurogenes CHG K8. It's inside my Spreadsheet with regional results.
 
I posted it at the thread I started about Eurogenes CHG K8. It's inside my Spreadsheet with regional results.

The info is in gedmatch after you test under the oracle boxes named spreadsheet

GEDmatch.Com

Population Spreadsheet for Eurasia K10 CHG.

the only "italian" for this test is Tuscan .............the only one


my wife's results come out as bulgarian with spanish which in the dendro seems to be a nearness to North-portuguese/Galician.

she does have some matches with Veneti who migrated to Galicia in ftdna. But she has many more matches with east-Austria than with the iberian peninsula

#PopulationPercent
1WHG33.58
2Anatolian_Farmers26.22
3CHG21.19
4EHG9.37
5SW_Asian8.16
6W_African0.81
7Papuan0.66

Single Population Sharing:

#Population (source)Distance
1Bulgarian4.14
2Spanish6.29
3Croatian7.38
4Albanian7.63
5French8.08
6Tuscan8.58
7Greek10.68
8French_South10.95
9Hungarian11.25
10English13.67
11Sicilian14.81
12Czech14.83
13Scottish15.29
14Maltese16.73
15Norwegian17.5
16Ukrainian18.02
17Icelandic18.35
18Belarusian21.74
19Sardinian22.06
20Russian23.55

Mixed Mode Population Sharing:

#Primary Population (source)Secondary Population (source)Distance
161.3% Bulgarian + 38.7% Spanish @ 1.5
 
I asked on the other thread, but I'll try again here. Where is the spreadsheet for population averages for the newest Eurogenes calculator which is showing all this elevated CHG. How can it be evaluated if we can't see how the populations relate to one another.

It shows many Europeans with more or at least equal amount of CHG as EF. That is unlikely imo. CHG should be at least ~1/5 or 1/4 less than EF.

It also shows EEF(Stuttgart) being ~20% additional WHG instead of 10% as per the Haak paper. Another strong indication that some of the EF ancestry is getting eaten up.

On the other hand Eurasia K10 CHG gives even far too low scores for EF as well CHG. Even the most EF group of Europe (Sardinians) score only 48%.

The Spredsheet for Eurogenes K8 CHG is on Fire Heads recent thread.
 
I posted it at the thread I started about Eurogenes CHG K8. It's inside my Spreadsheet with regional results.

Thank you, Fire-Haired,

I meant a proper spreadsheet with ALL the tested samples, and an Oracle function for all the samples so we can properly evaluate it, like Kurd's. It might be nice and respectful to add the academic samples for southern Italians too. Kurd had no problem getting them. It should be a couple of strokes on the computer.

Would their results go against some narrative? Or am I getting paranoid considering the history?

@Alan,
Just saw your post. Thanks to you too.

@Sile,
That's Kurd's calculator. I was asking about the Eurogenes K8 because I thought my question on the K8 thread might not have been seen.
 

This thread has been viewed 37400 times.

Back
Top