Eurogenes Biogeographic Ancestry Project

Gaita asked me to run the new k36 for him, here are his results in order:

----------------------------
FINAL ADMIXTURE PROPORTIONS:
----------------------------
33.09% Iberian
13.82% Italian
----------------------------
8.18% North_Sea
5.59% West_Med
4.80% East_Central_Euro
4.52% Basque
4.52% North_Atlantic
4.41% East_Balkan
4.35% North_African
3.79% East_Med
3.67% Central_Euro
2.42% French
2.16% Volga-Ural
1.37% Near_Eastern
1.16% Eastern_Euro
----------------------------
0.69% Indo-Chinese
0.61% Arabian
0.27% Northeast_African
0.23% West_African
0.18% Armenian
0.13% Fennoscandian
0.02% Amerindian
0.01% Malayan
----------------------------
0.00% Central_African
0.00% East_African
0.00% East_Asian
0.00% East_Central_Asian
0.00% North_Caucasian
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% West_Caucasian

K36 seems to be up and running at Gedmatch.

Mine is taking forever. lol Admixed people problems.

Some of my my numbers are slightly different from the first two K36 runs via 'R' and DIYdodecad. Also, the pie chart image to the right is not displaying correctly and appears to be broken. Could be because of all the sections it needs to account for. Anyone else experience this?

My picture for the pie chart is glitchy as hell though. :/

Happened 3 times already.
http://ww2.gedmatch.com:8006/autosom...917_75EC98.gif

These are my results though.
Amerindian 15.37%
Arabian 1.14%
Armenian 1.18%
Basque 3.25%
Central_African 1.29%
Central_Euro 6.09%
East_Central_Asian 0.06%
East_Central_Euro 1.86%
East_Med 0.48%
Fennoscandian 0.88%
French 3.20%
Iberian 17.72%
Italian 4.68%
Near_Eastern 8.31%
North_African 4.56%
North_Atlantic 4.92%
North_Sea 8.41%
Northeast_African 1.73%
Oceanian 0.15%
Pygmy 0.34%
Siberian 1.45%
South_Central_Asian 0.56%
West_African 7.17%
West_Med 5.17%

So 72.41 % Caucasian , 17.37% Amerindian and 10.19% SSA African roughly. Accurate.

Quote:

---

Originally Posted by mark

Some of my my numbers are slightly different from the first two K36 runs via 'R' and DIYdodecad. Also, the pie chart image to the right is not displaying correctly and appears to be broken. Could be because of all the sections it need to account for. Anyone else experience this?

---

My image was broken too, I'm running it in a different browser right now.

Im running it on my Macbook. Im running them on Google Chrome and Safari to see.

K36 - from GEDmatch

http://i.imgur.com/V0964oR.png

okay here we go:

http://img46.imageshack.us/img46/5124/k36results.png

Percentages are almost exactly the same, so that's good.

Edit to add:


Chromosome painting, reduced size:

http://img515.imageshack.us/img515/189/crhpaintk36.png
http://imageshack.us/a/img818/357/colorassign.png

http://i1065.photobucket.com/albums/...pscc3fbf91.png


http://i1065.photobucket.com/albums/...pse5c1d7cc.png


Very slight differences from before.... seem insignificant to compare:


North_Sea 20.60%
Iberian 17.13%
Fennoscandian 12.32%
North_Atlantic 11.91%
French 7.76%
Italian 7.46%
Central_Euro 5.88%
East_Balkan 3.85%
Eastern_Euro 3.53%
East_Central_Euro 3.27%
North_Caucasian 3.24%
Basque 1.28%
Armenian 1.22%
Amerindian 0.27%
South_Central_Asian 0.26%

I think I'm going to run mine bychr in diydodecad, because my painting is too jumbled - it's difficult to really see what's what.

Results seem to be very similar to the manual runs. Although i now have the chance to see what the phased data files look like which is interesting.

My Paternal Phased Data File:

21.33% North_Sea
14.39% North_Atlantic
12.13% Italian
11.21% Central_Euro
9.18% French
8.26% East_Central_Euro
7.56% Iberian
7.34% Fennoscandian
2.60% West_Med
2.09% Volga-Ural
1.61% Basque
1.25% East_Med
0.89% Eastern_Euro
0.15% Indo-Chinese

My Maternal Phased Data File:
(The mystery ancestry seems to be from the Balkans looking at this)

20.58% North_Sea
14.56% North_Atlantic
9.62% Iberian
8.79% East_Central_Euro
7.84% Central_Euro
6.88% East_Balkan
6.48% Eastern_Euro
5.86% West_Caucasian
5.69% French
4.18% Basque
3.22% Italian
2.02% Fennoscandian
1.78% East_Med
1.66% South_Central_Asian
0.60% West_African
0.19% Oceanian


Dad's Paternal Phased Data File:

21.68% North_Sea
13.49% Central_Euro
11.55% North_Atlantic
11.16% Italian
10.35% Fennoscandian
9.95% East_Central_Euro
7.99% Iberian
7.77% French
2.97% Basque
1.80% South_Central_Asian
0.67% Volga-Ural
0.24% West_Med
0.23% Eastern_Euro
0.07% East_Med
0.07% Near_Eastern

Dad's Maternal Phased Data File:

22.74% North_Sea
17.65% Iberian
14.09% North_Atlantic
8.99% French
8.33% East_Central_Euro
8.11% Italian
8.02% Central_Euro
4.38% Eastern_Euro
2.36% Fennoscandian
2.16% West_Med
2.11% Volga-Ural
1.08% East_Balkan

Btw, now the "by chromosome" utility works as well with K36 at gedmatch ;)

K36 by each chromosome

Me

http://s21.postimg.org/ihp8gl3t3/bychrj.jpg

My grandpa

http://s21.postimg.org/b0g11de9z/bychre.jpg


I have 42.6% Iberian at chr 20, my grandfather got 29.8% near_eastern at chr 22 :p

Quote:

---

Originally Posted by Mazik

Btw, now the "by chromosome" utility works as well with K36 at gedmatch ;)

---

Lol, and I just finished sorting mine.

K36 ByChr
http://img24.imageshack.us/img24/1192/bychrk36.png

I don't know if there's anything interesting about mine.

You probably know already, but "by chromosome" calculations are available through DIYDodecad too.
Int he .par files, like "admix.par", change the parameter "genomewide" to "bychr".


On a sidenote, it seems gedmatch is down...maybe the masses trying out K36 kiled it :D

http://i50.tinypic.com/2vkbaef.jpg

Screenshot missed out 22, but nothing special happens there.

http://ww2.gedmatch.com:8006/autosom...917_93DED5.png

Typical admixed person. One hell of a paint ballon fight all in my genes.

My chromosome painting:

http://s22.postimg.org/yb8iwq0sx/k36chrpt.jpg

Quote:

---

Originally Posted by Mazik

My chromosome painting:

http://s22.postimg.org/yb8iwq0sx/k36chrpt.jpg

---

Am I colourblind, or why do you have so much East-Asian? :confused:

Quote:

---

Originally Posted by Loki

Am I colourblind, or why do you have so much East-Asian? :confused:

---

It's north sea :)
the same colors are used for different components, it's a bit messed up, gedmatch is not made for that many components :p

Atlantic Islander posted this pic:

http://imageshack.us/a/img818/357/colorassign.png

Quote:

---

Originally Posted by Jackson

Results seem to be very similar to the manual runs. Although i now have the chance to see what the phased data files look like which is interesting.

---

Your Grandad has different results, a little more. South Central Asian for a start.

Paternal grandmother gedmatch k36 result:

Armenian 1.10%
Basque 2.36%
Central_Euro 7.20%
East_Balkan 3.44%
East_Central_Euro 3.90%
East_Med 6.64%
Eastern_Euro 1.95%
Fennoscandian 2.40%
French 5.73%
Iberian 13.87%
Italian 22.04%
North_African 0.39%
North_Atlantic 6.78%
North_Caucasian 0.28%
North_Sea 12.98%
South_Central_Asian 0.97%
West_Caucasian 1.98%
West_Med 5.97%

Paternal Grandfather's result

Amerindian 0.34%
Basque 4.91%
Central_Euro 6.18%
East_Balkan 2.70%
East_Central_Euro 7.05%
Eastern_Euro 4.80%
Fennoscandian 2.91%
French 8.96%
Iberian 13.17%
Italian 12.22%
North_African 0.40%
North_Atlantic 12.72%
North_Caucasian 3.99%
North_Sea 15.14%
Volga-Ural 0.67%
West_Med 3.84%




Mom's result

Armenian 0.07%
Basque 4.49%
Central_Euro 9.44%
East_Balkan 1.41%
East_Central_Euro 4.58%
East_Med 3.45%
Eastern_Euro 5.93%
Fennoscandian 5.51%
French 6.17%
Iberian 19.20%
Italian 10.50%
Near_Eastern 1.75%
North_Atlantic 7.35%
North_Sea 12.90%
West_Caucasian 1.41%
West_Med 5.85%

http://i1065.photobucket.com/albums/...psaa668dd9.png

39.8% Italian on Chromosome 3.... on 23andMe's Ancestry composition's speculative mode, the 0.5% "Southern European" that I get is on the same chromosome. Maybe it is real?

Do I get to join the med supremacy club now?

K36 by Chromosome with reduced size painting

http://i.imgur.com/v3sbFrY.png

http://i.imgur.com/gZcca2y.png

No idea what it means, but here is mine!

http://i1162.photobucket.com/albums/...ps2dcd774f.png


And my blob of colours!

http://i1162.photobucket.com/albums/...psb560174e.png

http://i631.photobucket.com/albums/u...h1lus/k36a.jpg
http://i631.photobucket.com/albums/u...h1lus/k36b.jpg

My K37 Gedmatch results:

23.23% North_Sea
18.07% Fennoscandian
16.13% North_Atlantic
9.62% French
8.86% East_Central_Euro
7.74% Central_Euro
6.21% Eastern_Euro
6.09% Iberian
2.87% Volga-Ural
1.18% East_Balkan


rest 0%

Whats "North Sea" exactly? Northern Euro?

Kent (U.K.) and Utah (U.S.)

>> Link to K36 Populations

My K36 results through Gedmatch
http://i738.photobucket.com/albums/x...c19f773796.jpg

Quote:

---

Originally Posted by Graham

Your Grandad has different results, a little more. South Central Asian for a start.

---

True. And my mothers side clearly likes to show up Balkanic ancestry on this test with about 6.88% East Balkan, 5.86 West Caucasian and 1.78% East Med. This test seems to be good for picking up more recent ancestry as it seems to exaggerate it more than other test making it more clear.

Quote:

---

Originally Posted by mark

Kent (U.K.) and Utah (U.S.)

>> Link to K36 Populations

---

Ricky, North Sea is Eastern England, (Kent, Utah) while North Atlantic is Western (SW Scots, Cornish, Utah), I think it is meant to be a sort of Celtic fringe vs. Anglo Saxon(Germanic) divider.

Looking at what constitutes the Iberian component, there are samples from quite a variety of Spaniards, maybe that is why a Northern Euro such as myself scores so high for it? Many of the regions are represented... how diverse are the Spaniards genetically from each other?

Doesn't k37 divide the Iberian? If so, can someone give me a link to download it from?

The "Iberian" populations includes spanish basques, which in my opinion should be in the "Basque" group. The rest of "Iberians" have no great difference from each other.

Quote:

---

Originally Posted by Oneeye

Doesn't k37 divide the Iberian? If so, can someone give me a link to download it from?

---

I think you mean these files, I had a hard time trying to edit the F file, so I didn't run it for myself.

This is what he wrote:

Quote:

---

New Jewish and Iberian clusters are available at the link below. These are compatible with the new K36 and K35 (but the Iberian cluster has to be removed).

Link

I have my doubts that splitting Iberia into west and east clusters is useful, but it was a suggestion from a native of the region.

---

Iberian Reference
Andalucia -- 4 Samples
Aragon -- 5 Samples
Baleares -- 8 Samples
Cantabria -- 6 Samples
Castilla_La_Mancha -- 4 Samples
Castilla_Y_Leon -- 12 Samples
Cataluna -- 5 Samples
Extremadura -- 7 Samples
Spanish -- 12 Samples
Galicia -- 8 Samples
Murcia -- 6 Samples
Vasco -- 8 Samples
Valencia -- 10 samples



If you could divide them into clusters. What would anyone pick? I'm Curious

Quote:

---

Originally Posted by Atlantic Islander

I think you mean these files, I had a hard time trying to edit the F file, so I didn't run it for myself.

---

Is it simply a matter of adding the content/text of 'Extra_clusters.txt' to 'admix.36.F' and running as usual?

Quote:

---

Originally Posted by Graham

Iberian Reference
Andalucia -- 4 Samples
Aragon -- 5 Samples
Baleares -- 8 Samples
Cantabria -- 6 Samples
Castilla_La_Mancha -- 4 Samples
Castilla_Y_Leon -- 12 Samples
Cataluna -- 5 Samples
Spanish_Extremadura -- 7 Samples
Spanish -- 12 Samples
Galicia -- 8 Samples
Murcia -- 6 Samples
Vasco -- 8 Samples
Valencia -- 10 samples



If you could divide them into clusters. What would anyone pick? I'm Curious

---

Dienekes did something with that. The results he came up with were the following:

Quote:

---

The Iberian results are particularly interesting, showing the power of this approach compared to the one with unlinked data. There appear to be:

a Spanish Basque (#6),
French Basque (#11) cluster, as well as
a Portuguese/Galician/Castilla Y Leon (#9) cluster, and
a complementary Castilla La Manch/Cantabria/Andalucia/Murcia (#7) cluster, and
a smaller Aragon/Cataluna cluster (#8).

There is overlap between these clusters, but the geographical contrasts are quite evident. I did not go through the results of Spanish Project participants (all the Portuguese fall in the Galician cluster, and our Basque member in the Basque cluster as expeccted), so it would be interesting to hear whether they fall in the cluster(s) which exist in their regions of origin.

---

http://dodecad.blogspot.pt/2012/01/f...nce-italy.html

So if those were to be divided into clusters we probably would get:

- Galicia / Castille-Léon cluster - NW Iberia
- Aragon / Catalonia cluster - NE Iberia
- the rest - Main Iberian Cluster

The clusters correlate more or less with languages:

http://2.bp.blogspot.com/_0hasQbYINU...C3%A9ricas.jpg