Dodecad Ancestry Project

[Quasimodem]

Not true. The tests many of them ARE conflicting. Take 23andme, dienekes, and polako's. On polako's tests the british islanders and scandinavians show almost no non northern european admixture. In dienekes test they show something like 20-30% non northern european admixture. That is pretty substantial if you ask me. 23andme labels almost every euro as 100% or 99%. Again SUBSTANIAL.

That's because the allele frequencies for these same-name components are different, so they're in effect different guessed ancestral populations.

[Frederick]

Not true. The tests many of them ARE conflicting. Take 23andme, dienekes, and polako's. On polako's tests the british islanders and scandinavians show almost no non northern european admixture. In dienekes test they show something like 20-30% non northern european admixture. That is pretty substantial if you ask me. 23andme labels almost every euro as 100% or 99%. Again SUBSTANIAL.

Thats just different interpretation.
The "non-European admixture" that Dienekes or Eurogenes show is like 7.000 or 10.000 years old. Neolithic farmers DNA and all that.

Specially "West-Asian" could be interpreted as "Early farmer" aDNA.
Others claim, "West-Asian" admixture would be the "Indoeuropean" one (Caucasus/Anatolian Indoeuropean origin hyphothesis)

One needs to alter ones views on these "Non-European" components.
All Europeans somehow came from outside of Europe and its totaly normal, that they carry DNA material that exists outside of Europe or even has its distributional center outside of Europe.

[Frederick]

Since they are covered deeply in this thread I repost the maps of the dienekes components I made:

Red = no other component is higher that this one.
thats why Hungary is red in "West European" but Spain is not.
Its because in Hungary, "Westeuropean" is the largest component while in Spain its "Mediteranean".




If you ask: "DOH! Why are those Orkneys so Eastern Europen?"
Answere: Because they have Norwegian VIKING admixture





The component some claim as "Farmers DNA" or even "Indoeuropean DNA" if it comes to Europe....






And I add some, wich I seem to not have uploaded last time?

[Scrapple]

Not true. The tests many of them ARE conflicting. Take 23andme, dienekes, and polako's. On polako's tests the british islanders and scandinavians show almost no non northern european admixture. In dienekes test they show something like 20-30% non northern european admixture. That is pretty substantial if you ask me. 23andme labels almost every euro as 100% or 99%. Again SUBSTANIAL.

That is the problem with labelling allele frequencies geographically, the person doing the test chooses the labels.

[Amapola]

Per chromosome:

[1,] "Portuguese_D" "5.1126"
[2,] "IBS" "6.24"
[3,] "Spaniards" "6.605"
[4,] "Spanish_D" "7.0589"
[5,] "N_Italian_D" "11.5477"
[6,] "North_Italian" "12.0684"
[7,] "French_Basque" "14.9499"
[8,] "Sardinian" "15.8149"
[9,] "French_D" "16.2501"
[10,] "French" "16.7247"
[11,] "Tuscan_H" "17.4539"
[12,] "TSI" "17.7588"
[13,] "Tuscan_X" "18.385"
[14,] "Tuscan" "18.9673"
[15,] "O_Italian_D" "19.4393"
[16,] "C_Italian_D" "25.2012"
[17,] "CEU" "25.8471"
[18,] "Orcadian" "27.7598"
[19,] "N._European" "27.7935"
[20,] "Orkney_1KG" "28.1339"
[21,] "Argyll_1KG" "28.5554"
[22,] "Ashkenazy_Jews" "30.0512"
[23,] "Dutch_D" "30.3059"
[24,] "Slovenian" "30.4614"
[25,] "German_D" "30.5844"
[26,] "Mixed_Germanic_D" "30.5883"
[27,] "Ashkenazi_D" "30.6159"
[28,] "S_Italian_Sicilian_D" "31.515"
[29,] "Kent_1KG" "31.8929"
[30,] "Sicilian_D" "32.5025"

Nice variation.

[Barreldriver]

Curious with the DIY Dodecad what's the gauge to indicate relatively recent admix with the percentages? Still trying to interpret my results in the context of my genealogy so any help would be appreciated.

[Frederick]

Curious with the DIY Dodecad what's the gauge to indicate relatively recent admix with the percentages? Still trying to interpret my results in the context of my genealogy so any help would be appreciated.

In that case, I would run the results in "Oracle". Thats another "R" based software by Dienekes with 2 major functions:
- what populations average results are the most similiar to ones own.
- if mathematically "mixing" 2 nations average results, what combination comes closest to ones results.

Plus I would additional look in 23andMe what people you share long segments with (other than the "expected" own group... (that theory that long segments of shared DNA are a sign of beeing "recently" (last couple centuries maybe) related and not just having the same DNA by random re-combining. This would help to pick, what suggested combinations of "Oracle" may be likely, if there is no "Family Lore".

I for example, have 3 out of 4 Grandparents from "Prussia" (beeing from there, mixing with Poles is likely) and one Great Great Grandparent from Lithuania.

In fact 23andMe finds quiet a lot Poles in the 5-7cM region, suggesting "recent" (last couple century) shared anchestors. Of course, nobody knows WHAT anchestors I may share with them. Maybe they had German ones. Who knows. But you need something to start with.

"Oracle" suggests that my results are close to what is expected from 95% average German mixed by 5% average Pole.
Or same with Lithuanian.

[Barreldriver]

In that case, I would run the results in "Oracle". Thats another "R" based software by Dienekes with 2 major functions:
- what populations average results are the most similiar to ones own.
- if mathematically "mixing" 2 nations average results, what combination comes closest to ones results.

Plus I would additional look in 23andMe what people you share long segments with (other than the "expected" own group... (that theory that long segments of shared DNA are a sign of beeing "recently" (last couple centuries maybe) related and not just having the same DNA by random re-combining. This would help to pick, what suggested combinations of "Oracle" may be likely, if there is no "Family Lore".

I for example, have 3 out of 4 Grandparents from "Prussia" (beeing from there, mixing with Poles is likely) and one Great Great Grandparent from Lithuania.

In fact 23andMe finds quiet a lot Poles in the 5-7cM region, suggesting "recent" (last couple century) shared anchestors. Of course, nobody knows WHAT anchestors I may share with them. Maybe they had German ones. Who knows. But you need something to start with.

"Oracle" suggests that my results are close to what is expected from 95% average German mixed by 5% average Pole.
Or same with Lithuanian.

I'd assume there's a link to oracle at Dienekes blog? Will give it a wing some time.

[Frederick]

I tried it for you. (All one needs is the DIY Dodecad results)

By Oracle, these are the closest populations:

(so more small the number behind it, so more close the results)
Barreldriver

[1,] "CEU" "2.9717" <--------- white American of predominantly North-West European anchestry
[2,] "N._European" "5.0694" (a group with "Northern Europeans" not closer defined what people exactly)
[3,] "Orcadian" "5.2101"
[4,] "Argyll_1KG" "5.391" (South-West-Scottland)
[5,] "Orkney_1KG" "5.6548"
[6,] "German_D" "8.5726"
[7,] "French" "9.6303"
[8,] "French_D" "10.0577"
[9,] "Mixed_Germanic_D" "10.6703" (people of mixed background, but all known anchestors of Germanic speaking parts of Europe)
[10,] "Dutch_D" "11.7807"
[11,] "Kent_1KG" "14.129" (South-East England)
[12,] "British_Isles_D" "16.2479"
[13,] "British_D" "16.2505"
[14,] "Cornwall_1KG" "16.8781" (South-West England)
[15,] "Slovenian" "17.4146"
[16,] "Irish_D" "19.6356"
[17,] "Swedish_D" "20.1689"
[18,] "Norwegian_D" "20.9766"
[19,] "Hungarians" "21.9821"
[20,] "French_Basque" "22.4489"
[21,] "Portuguese_D" "24.3769"
[22,] "Spaniards" "25.2005"
[23,] "N_Italian_D" "26.3565"
[24,] "Spanish_D" "26.5371"
[25,] "IBS" "26.6198"
[26,] "Tuscan_H" "27.2357"
[27,] "TSI" "27.2763"
[28,] "Tuscan_X" "27.6064"
[29,] "FIN" "29.219"
[30,] "North_Italian" "30.4805"

Just for comparation the results of "Average CEU". to help you, what to think about the results above:

[1,] "CEU" "0"
[2,] "Orcadian" "2.4839"
[3,] "Orkney_1KG" "2.924"
[4,] "Argyll_1KG" "2.9614"
[5,] "N._European" "3.3347"
[6,] "German_D" "6.9361"
[7,] "Mixed_Germanic_D" "9.3862"
[8,] "Dutch_D" "10.3315"
[9,] "French" "11.1736"
[10,] "French_D" "11.4066"

Thats the top 10 suggestion for 2 nation mixedmode:
As always: so more small the number at the end of the line is, so more close are the results)

And note how the tool is named: ORACLE.
It needs you to interpret this and pick what you guess is the most likely.

Another note: It only expects TWO nation mixes. Thats another weak point.

[1,] "90.4% Orcadian + 9.6% S_Italian_D" "0.4005"
[2,] "89.7% Orkney_1KG + 10.3% S_Italian_D" "0.4049"
[3,] "12.5% C_Italian_D + 87.5% Orkney_1KG" "0.4141"
[4,] "90.1% Orcadian + 9.9% Sicilian_D" "0.4439"
[5,] "89.7% Orcadian + 10.3% S_Italian_Sicilian_D" "0.4723"
[6,] "89.3% Orkney_1KG + 10.7% Sicilian_D" "0.4816"
[7,] "11.7% C_Italian_D + 88.3% Orcadian" "0.4921"
[8,] "14.5% O_Italian_D + 85.5% Orkney_1KG" "0.5639"
[9,] "13.5% O_Italian_D + 86.5% Orcadian" "0.5794"
[10,] "84.2% Orcadian + 15.8% Tuscan_X" "0.5924"

Here are some "aberations" from the average CEU:

[14,] "90.4% CEU + 9.6% TSI" "0.714"
[18,] "92% CEU + 8% O_Italian_D" "0.795"
[20,] "90.6% CEU + 9.4% Tuscan_X" "0.8094"
[27,] "94.4% CEU + 5.6% Sicilian_D" "0.9574"
[31,] "95.7% CEU + 4.3% Cypriots" "1.028"
[36,] "94.9% CEU + 5.1% Sephardic_Jews" "1.0779"
[39,] "94.3% CEU + 5.7% Greek_D" "1.1763"
[48,] "95.5% CEU + 4.5% Turkish_D" "1.3667"

some British Isles aberations:

[50,] "11.7% C_Italian_D + 88.3% Argyll_1KG" "1.3712"
[60,] "30% Balkans_D + 70% Kent_1KG" "1.5231"
[70,] "33.9% Balkans_D + 66.1% Cornwall_1KG" "1.6006"
[75,] "67.6% British_Isles_D + 32.4% Romanians_14" "1.6621"

Maybe its just this:
[134,] "46.9% French + 53.1% German_D" "2.3392"
[135,] "83.6% CEU + 16.4% French" "2.3399"
[145,] "5% Balkans_D + 95% CEU" "2.4219"

[Imperivm]

If anyone is feeling like being a very helpful chap - could they please run all these tests with my DOD scores? I'm DOD726.

I can't run any of the tests on my mac because it's not compatible. Although, I ran DiY 2.0, and got similar scores to everyone else.

It will be much appreciated, and met with hugs and kisses.