View Full Version : Post your West Eurasian Cline (WEAC2) Results for PCAs & Graphs
The Weac calculators are low k calculators by Dienekes which place individuals on the West Eurasian cline. This cline is the first-order description of variation in West Eurasians, with populations from northern and western Europe falling on one end, and those from the Near East on the other. The spreadsheet & FST table is available at http://dodecad.blogspot.com/2012/06/weac2-calculator.html
For those who don't want to fool around with R, the following is a link to a DIY Dodecad Wrapper calculator http://www.y-str.org/2013/07/diy-dodecad-21-wrapper.html, which is a very user friendly and simple to use calculator. All you have to do is point it to your unzipped genome file, and select the calculator you would like to use. You will be able to choose any of a list of Dodecad World calculators, one of the Weac calculators, or the Eurasia 7 calculator. Just point to your genome file, select the calculator from the drop down menu and click run.
The wrapper is a simple program which makes it easy for users to use DIYDodecad 2.1 with default parameters.
Also, in case Dienekes has used the samples in his admixture run, the so-called "calculator effect", I believe will be minimized because the Weac2 is a relatively low K calculator with larger FST distances between the admixture components. This is in case I decide to incorporate his spreadsheet references into the analysis.
You can right click on your resulting pie chart, save as image, and post here.
Here are my Weac2 results:
http://i.imgur.com/fdpBDgE.jpg
http://i.imgur.com/Ra8g7kZ.jpg
BalkanKiwi
05-10-2015, 01:28 AM
Did a run with R.
0.00% Palaeoafrican
73.75% Atlantic_Baltic
0.17% Northeast_Asian
20.68% Near_East
0.00% Sub_Saharan
5.02% South_Asian
0.38% Southeast_Asian
Chad Rohlfsen
05-10-2015, 01:34 AM
Here's mine.
4549
How do I make it larger?
Tolan
05-10-2015, 01:58 AM
Here's mine:
Weac 2
Population
0.01% Palaeoafrican
72.86% Atlantic_Baltic
0.00% Northeast_Asian
24.10% Near_East
0.13% Sub_Saharan
2.88% South_Asian
0.01% Southeast_Asian
Kandhari Pashtun, Zahra's dad. Not very different from mine. Our total E Asian is quite similar, his 6.52% is split 6.09% NE Asian, 0.43% SE Asian. My 6.93% E Asian is split 4.65 NE Asian, 2.28% SE Asian. His S Asian is a little higher, 36.02 vs 27.88% for me. Overall he is a little more east on the cline.
http://i.imgur.com/GcIFRwS.jpg
Hanna
05-10-2015, 02:52 AM
I was included in the spreadsheet
Weac 2
0.00% Palaeoafrican
22.8% Atlantic_Baltic
5.1% Northeast_Asian
59.8% Near_East
0.00% Sub_Saharan
12.0% South_Asian
0.03% Southeast_Asian
Rukha's:
http://i.imgur.com/9po64uL.jpg
randwulf
05-10-2015, 03:28 AM
I had just re-downloaded this one last night. The numbers:
Me:
0.46% Palaeoafrican
73.37% Atlantic_Baltic
1.31% Northeast_Asian
22.57% Near_East
0.00% Sub_Saharan
2.22% South_Asian
0.08% Southeast_Asian
Dad:
0.33% Palaeoafrican
72.04% Atlantic_Baltic
2.03% Northeast_Asian
22.38% Near_East
0.12% Sub_Saharan
2.79% South_Asian
0.32% Southeast_Asian
Mom:
0.00% Palaeoafrican
76.87% Atlantic_Baltic
0.13% Northeast_Asian
19.63% Near_East
0.00% Sub_Saharan
3.30% South_Asian
0.07% Southeast_Asian
Wife:
0.05% Palaeoafrican
74.68% Atlantic_Baltic
1.37% Northeast_Asian
21.74% Near_East
0.38% Sub_Saharan
1.61% South_Asian
0.16% Southeast_Asian
Wife's Dad:
0.13% Palaeoafrican
77.41% Atlantic_Baltic
1.01% Northeast_Asian
18.34% Near_East
0.95% Sub_Saharan
1.87% South_Asian
0.29% Southeast_Asian
Varun R
05-10-2015, 03:32 AM
My pie-chart
4550
firemonkey
05-10-2015, 03:32 AM
1 Atlantic Baltic 79.31
2 NearEast 15.99
3 South Asian 3.51
4 NorthEast Asian 1.02
5 SubSaharan 0.12
6 Southeast Asian 0.04
ZephyrousMandaru
05-10-2015, 03:32 AM
WEAC2
Dodecad WEAC2
0.32% Palaeoafrican
15.38% Atlantic_Baltic
0.51% Northeast_Asian
72.00% Near_East
0.01% Sub_Saharan
11.79% South_Asian
0.00% Southeast_Asian
My daughter's:
http://i.imgur.com/zaxPPKU.jpg
Jesus's WEAC2:
http://i.imgur.com/xhYlwbQ.jpg
Kaido
05-10-2015, 05:20 AM
http://i.imgur.com/VoSW2gT.jpg
CelticGerman
05-10-2015, 06:23 AM
My result WEAC2: 4551
BalkanKiwi
05-10-2015, 06:30 AM
My result WEAC2: 4551
That's some impressive Atlantic Baltic.
CelticGerman
05-10-2015, 06:52 AM
Here are my uncle's results:4552
vettor
05-10-2015, 07:08 AM
I am close to celticGerman
http://i103.photobucket.com/albums/m153/vicpret/weac2_zpsosvyoj0t.jpg (http://s103.photobucket.com/user/vicpret/media/weac2_zpsosvyoj0t.jpg.html)
As a T1 ( ydna )........my ancestors must have been in Europe a very long time, or am I misreading this?
Bulut
05-10-2015, 08:00 AM
WEAC2 with R;
0.00% Palaeoafrican
29.04% Atlantic_Baltic
8.69% Northeast_Asian
47.73% Near_East
1.14% Sub_Saharan
12.01% South_Asian
1.39% Southeast_Asian
0.17% Palaeoafrican
19.97% Atlantic_Baltic
2.30% Northeast_Asian
59.81% Near_East
0.01% Sub_Saharan
17.73% South_Asian
0.01% Southeast_Asian
Anabasis
05-10-2015, 08:27 AM
Here i am
Palaeoafrican 0
Atlantic_Baltic 19,8
Northeast_Asian 0
Near_East 69,17
Sub_Saharan 0,07
South_Asian 10,95
Southeast_Asian 0
CelticGerman
05-10-2015, 08:40 AM
Here are WEAC2 family results for comparison (German, French ...): 4553
0.17% Palaeoafrican
19.97% Atlantic_Baltic
2.30% Northeast_Asian
59.81% Near_East
0.01% Sub_Saharan
17.73% South_Asian
0.01% Southeast_Asian
My Ezidi Kurdish friend's results
0.13% Palaeoafrican
19.14% Atlantic_Baltic
1.54% Northeast_Asian
59.61% Near_East
0.02% Sub_Saharan
19.56% South_Asian
0.01% Southeast_Asian
I am close to celticGerman
As a T1 ( ydna )........my ancestors must have been in Europe a very long time, or am I misreading this?
How are you close to Celticgerman. He has 85% Atlantic Baltic & 11% Near East and yours are 61%/35%. You are kidding, right? You are a lot closer to French member Tolan than to him. In fact, your results are consistent with the N Italian average of 60.4%/37.4%.
CelticGerman
05-10-2015, 10:35 AM
How are you close to Celticgerman. He has 85% Atlantic Baltic & 11% Near East and yours are 61%/35%. You are kidding, right? You are a lot closer to French member Tolan than to him. In fact, your results are consistent with the N Italian average of 60.4%/37.4%.
Kurd, the 85%/11% is my uncle's result. My own "CelticGerman" result is around 83%/13%. Besides this you are right. It would be nice to see typical figures from Irish, British, Dutch and Scandinavian people as well for comparison. However, I imagine where it will go.
Kurd, the 85%/11% is my uncle's result. My own "CelticGerman" result is around 83%/13%. Besides this you are right. It would be nice to see typical figures from Irish, British, Dutch and Scandinavian people as well for comparison. However, I imagine where it will go.
You can check the spreadsheet under the 1st link I provided.
@Chad
You can take a screen shot of your result, or right click on it and choose save image, and then upload it to a photo sharing site such as imgur.com. You would have to copy the link ( I believe 4th option down), and insert image into the post as a URL.
firemonkey
05-10-2015, 11:23 AM
Kurd, the 85%/11% is my uncle's result. My own "CelticGerman" result is around 83%/13%. Besides this you are right. It would be nice to see typical figures from Irish, British, Dutch and Scandinavian people as well for comparison. However, I imagine where it will go.
I am British. Atlantic Baltic 79.31 NearEast 15.99
John Doe
05-10-2015, 11:46 AM
Here are my results (will you make a PCA out of them?):
4554
CelticGerman
05-10-2015, 01:16 PM
Here are WEAC2 family results for comparison (German, French ...): 4553
I had a look on the spreadsheet:
At first glance my results seem to be closest to Ireland, Orkney and Argyll (no known ancestry there).
My uncle maternal side is close to Norway or Poland.
My uncle paternal side fits well with Germans.
I am not sure the samples used are exactly what we need for Germany. Probably it is a mix of different German regions, which are quite different from each other, the Hamburg area in comparison with Bavaria for example. I am sure North German samples alone would be much closer to my result. Nevertheless, there could be some Irish or British in my genes too according to an analysis made by Davidski last year.
0.04% Palaeoafrican
5.37% Atlantic_Baltic
1.25% Northeast_Asian
22.59% Near_East
0.00% Sub_Saharan
63.00% South_Asian
7.74% Southeast_Asian
0.00% Palaeoafrican
6.97% Atlantic_Baltic
0.87% Northeast_Asian
15.66% Near_East
0.00% Sub_Saharan
66.90% South_Asian
9.60% Southeast_Asian
Sapporo's:
http://i.imgur.com/gKsNK0M.jpg
The following are various sorted graphs for the community. If you don't see yourself on there, you will be included in the next set. PCAs & dendograms will follow later today or tomorrow:
http://i.imgur.com/YOioxFT.jpg
http://i.imgur.com/Mfsc3XM.jpg
http://i.imgur.com/kBXu7tB.jpg
http://i.imgur.com/K5c419S.jpg
The Barnacle
05-10-2015, 06:29 PM
What is the South Asian made up of? It's pretty high among South Central Asians and near easterners.
Hanna
05-10-2015, 06:36 PM
What is the South Asian made up of? It's pretty among South Central Asians and near easterners.
I think it includes Gedrosia. Depending on the calculator I usually don't score South Asian however, in this calculator it shows that I have 12% South Asian. I am assuming it's actually Gedrosia for me.
The Barnacle
05-10-2015, 07:17 PM
I think it includes Gedrosia. Depending on the calculator I usually don't score South Asian however, in this calculator it shows that I have 12% South Asian. I am assuming it's actually Gedrosia for me.
We'll to be fair, the "South Asian component " in pretty much every calculator is a hybrid between west Eurasian ( gedrosian , Caucasus etc) and actual ASI, which in the other thread by Chad, is being broken down finally.
What is the South Asian made up of? It's pretty high among South Central Asians and near easterners.
The S Asian component peaks in S Indian groups such as the Pulayar and Hakipiki (80%), and not in SC Asian Gedrosian high groups. The Near Eastern component peaks in Saudis (85%) and Samaritians.
When groups such as Saudis are used as references in supervised runs, and there are no Caucasus groups (my guess), then it is reasonable to expect that there will be a stronger S Indian pull on W & SC Asian groups.
The Barnacle
05-10-2015, 07:23 PM
The S Asian component peaks in S Indian groups such as the Pulayar and Hakipiki (80%), and not in SC Asian Gedrosian high groups. The Near Eastern component peaks in Saudis (85%) and Samaritians.
When groups such as Saudis are used as references in supervised runs, and there are no Caucasus groups (my guess), then it is reasonable to expect that there will be a stronger S Indian pull on W & SC Asian groups.
So why does it reach upto 36% for zahras dad ( 14% South Indian on harappa) and Rukha ( 15% South Indian)?
So why does it reach upto 36% for zahras dad ( 14% South Indian on harappa) and Rukha ( 15% South Indian)?
When groups such as Saudis are used as references in supervised runs, and there are no Caucasus groups (my guess), then it is reasonable to expect that there will be a stronger S Indian pull on W & SC Asian groups.
Here Is a simplistic example:
If Zahra D is compared against S Indian vs Saudi, his S Indian may be 36%
But if he is compared against S Indian vs Armenian, his S Indian may drop to 16%, because now he has a greater pull (overlap) with armenian than he had with Saudi
Low k calculators have less component references as they are not meant to refine admixture to a detailed sub regional level
Chad Rohlfsen
05-10-2015, 09:07 PM
The only problem with using Caucasus pops in admixture runs is that they already have some ASI, in small amounts, so they can exaggerate the Caucasus/West Asian"ness", of those SC and S Asian pops. There was obviously some small amount of ASI that drifted into West Asia and the Balkans. I suppose it's anyone's guess as to when. All we know is that it wasn't in the first farmers. There's too much back and forth. We need an early Caucasus farmer and a Harappan farmer. It's a guessing game until then.
Here is a plot of PC1/PC2 which captures approx. 70% of the variation. The population icons represent the averages of their respective groups and were taken from the Weac2 spreadsheet. All the plots are the same, just at different zoom levels to show detail. The eigenvector plot shows the admixture component axes. I have also included a screen shot of the eigenvalues of each component to give you an idea of percentage of each component is represented in the PCA. I had to remove a couple of relatives from the European members, because they were obstructing the view of the members
http://i.imgur.com/K4CBBgf.jpg
http://i.imgur.com/hi5Jb2v.jpg
http://i.imgur.com/zUnmTTm.jpg
http://i.imgur.com/otb7dDT.jpg
http://i.imgur.com/Xkq8aab.jpg
http://i.imgur.com/Bcnlo0s.jpg
Here are the accompanying dendograms. The Euclidean method was used for member to member distance measurement. I am posting 2 plots, one using the "furthest neighbor" method of intra clusteroid measurement, the other the "group averages" method. Please refer to tutorials/explanation of these methods that I have posted under other threads showing dendograms. The populations represent their average and were taken from the spreadsheet.
http://i.imgur.com/i67tWIC.jpg
http://i.imgur.com/nMPRdpu.jpg
thrax
05-11-2015, 01:43 PM
Atlantic Baltic 49.86
Near East 44.23
South Asian 5.67
Northeast Asian 0.16
Sub Saharan 0.03
Southeast Asian 0.03
Palaeoafrican 0.02
John Doe
05-11-2015, 01:47 PM
Here are the accompanying dendograms. The Euclidean method was used for member to member distance measurement. I am posting 2 plots, one using the "furthest neighbor" method of intra clusteroid measurement, the other the "group averages" method. Please refer to tutorials/explanation of these methods that I have posted under other threads showing dendograms. The populations represent their average and were taken from the spreadsheet.
http://i.imgur.com/i67tWIC.jpg
http://i.imgur.com/nMPRdpu.jpg
So... According to this, I'm closer to Pommies than Spaniards?
So... According to this, I'm closer to Pommies than Spaniards?
At this level of k indistiguishable from S Italians/ Sicilians
John Doe
05-11-2015, 02:39 PM
At this level of k indistiguishable from S Italians/ Sicilians
I can see that, what surprises me is that the English seem to be genetically closer to me than the Spaniards.
P.S On the PCA I seem to again be a little closer to Europe than the main Ashenazi cluster, just like in the previous one.
I can see that, what surprises me is that the English seem to be genetically closer to me than the Spaniards.
P.S On the PCA I seem to again be a little closer to Europe than the main Ashenazi cluster, just like in the previous one.
We must not be looking at the same PCA then. I think you are confusing JTom with yourself
CelticGerman
05-11-2015, 03:23 PM
Interesting to see I'm close to the British Isles (CELTICG) whereas my uncle (CELTICG U) is with the Swedes in the dendogram.
icebreaker
05-11-2015, 03:46 PM
0.20% Palaeoafrican
25.04% Atlantic_Baltic
7.06% Northeast_Asian
52.85% Near_East
0.04% Sub_Saharan
13.18% South_Asian
1.63% Southeast_Asian
MJost
05-11-2015, 06:01 PM
WEAC2
MJost
CSJ*
DPJ*
JWJ*
LFJ*
KW(J2b2)
SMcKaig (I2b1)
PAfr
0.53
0
0
0
0
0
0.07
AltBalt
77.51
76.53
76.41
72.65
74.92
76.38
76.18
NEAsian
1.53
0.2
0.27
0.65
0.02
0.02
1.04
NearEast
17.9
20.31
19.82
22.21
19.96
21.27
19.08
SubSah
0.01
0.13
0.23
0.21
0.18
0
0.56
SAsian
2.33
2.83
3.26
4.27
4.92
2.31
3.04
SEAsian
0.19
0
0
0.01
0
0.02
0.03
Total
100
100
99.99
100
100
100
100
* NW German 1/2 Maternal brothers U106
MJost
Here are the accompanying dendograms. The Euclidean method was used for member to member distance measurement. I am posting 2 plots, one using the "furthest neighbor" method of intra clusteroid measurement, the other the "group averages" method. Please refer to tutorials/explanation of these methods that I have posted under other threads showing dendograms. The populations represent their average and were taken from the spreadsheet.
http://i.imgur.com/i67tWIC.jpg
http://i.imgur.com/nMPRdpu.jpg
The following shows the admixture similarity to Paniyas
http://i.imgur.com/yYDnodp.jpg
Sapporo
05-12-2015, 06:50 AM
@Kurd
Did you mix up my results with someone else such as yourself or jesus when comparing to Paniyas? How could I be more distant from Paniya than yourself, your daughter, jesus, rukha, adam and zahra's d? I'm the South Asian. :)
Maybe there is something I'm missing with regards to your similarity comparison?
Also, you, your daughter and jesus should be less similar to paniya than rukha, adam and zahra's d no?
@Kurd
Did you mix up my results with someone else such as yourself or jesus when comparing to Paniyas? How could I be more distant from Paniya than yourself, your daughter, jesus, rukha, adam and zahra's d? I'm the South Asian. :)
Maybe there is something I'm missing with regards to your similarity comparison?
Also, you, your daughter and jesus should be less similar to paniya than rukha, adam and zahra's d no?
I checked, and everything seems ok. Paniya's SE/S Asian admixture was about 75/21%, and had a small amount of SSA. You had 5%NE Asian, and no SE Asian or SSA. That is why your score was less than us.
MonkeyDLuffy
05-12-2015, 01:03 PM
Mine
Paleoafrican 0.0
AtlanticBaltic 19.41
NortheastAsian 4.82
Near east 19.45
Sub saharan 0.0
South Asian 54.06
South east asian 2.26
Here is a plot of PC1/PC2 which captures approx. 70% of the variation. The population icons represent the averages of their respective groups and were taken from the Weac2 spreadsheet. All the plots are the same, just at different zoom levels to show detail. The eigenvector plot shows the admixture component axes. I have also included a screen shot of the eigenvalues of each component to give you an idea of percentage of each component is represented in the PCA. I had to remove a couple of relatives from the European members, because they were obstructing the view of the members
http://i.imgur.com/K4CBBgf.jpg
http://i.imgur.com/hi5Jb2v.jpg
http://i.imgur.com/zUnmTTm.jpg
http://i.imgur.com/otb7dDT.jpg
http://i.imgur.com/Xkq8aab.jpg
http://i.imgur.com/Bcnlo0s.jpg
http://i.imgur.com/jKSvJnL.jpg
http://i.imgur.com/XOPoxw9.jpg
MonkeyDLuffy
05-12-2015, 01:07 PM
Is it possible if you can throw my results in there too kurd?
Is it possible if you can throw my results in there too kurd?
Ok, I will re-post tonight.
jesus
05-12-2015, 01:37 PM
2 Iranians from Hamadan/Khorasan. There results are very similar to mine and Kurd's
http://i.imgur.com/C33y3re.png
Iranian provinces
http://www.mapsofworld.com/iran/maps/iran-political-map.jpg
SwampThing27
05-12-2015, 05:47 PM
Paleoafrican 0.18
AtlanticBaltic 74.53
NortheastAsian 0.82
Near east 18.38
Sub saharan 0.88
South Asian 05.20
South east asian 00.01
5.2% South Asian- first calculator that has gone over 4% for me on this component.
Gray Fox
05-12-2015, 06:32 PM
Here's mine.
4563
J Man
05-12-2015, 06:36 PM
This is quite an interesting calculator in that it splits West Eurasian ancestry up into the Atlantic_Baltic and Near_East components. It is tempting to see the Atlantic_Baltic as a European Mesolithic hunter-gatherer component and the Near_East component as a Neolithic farmer component but that is probably far to simple.
Sapporo
05-12-2015, 06:46 PM
I checked, and everything seems ok. Paniya's SE/S Asian admixture was about 75/21%, and had a small amount of SSA. You had 5%NE Asian, and no SE Asian or SSA. That is why your score was less than us.
Interesting. So, the lack of SE Asian admixture outweighed the heavier South Asian admixture. Thanks.
Interesting. So, the lack of SE Asian admixture outweighed the heavier South Asian admixture. Thanks.
I would put it as since the Paniya averaged 21% S Asian, your overlap was limited to 21% (think of it as you being given credit for only 21% of you S Asian). With all other components being the same, assuming you had my SE Asian and SSA levels, if the Paniya had anything more than 28% S Asian (my level), then you would have scored higher than me
Sapporo
05-12-2015, 10:08 PM
I would put it as since the Paniya averaged 21% S Asian, your overlap was limited to 21% (think of it as you being given credit for only 21% of you S Asian). With all other components being the same, assuming you had my SE Asian and SSA levels, if the Paniya had anything more than 28% S Asian (my level), then you would have scored higher than me
I see. What were their admixture results? I presume they scored heavily in the SE Asian category rather than South Asian, which is presumably heavily West Eurasian influenced on this admixture run.
I see. What were their admixture results? I presume they scored heavily in the SE Asian category rather than South Asian, which is presumably heavily West Eurasian influenced on this admixture run.
1.4 0.8 0.1 0 0.9 21.6 75.1
75% SE Asian. The S Asian peaks in S India at around 80%, with Fil, BMG, and Varun having high scores.
WEAC2 using the latest 4 way oracle posted on the Eurogenes blog:
Target = 34% Jatt_D + 35% Saudis + 30% Tajiks_Y + 1% Kurds_Y @ D = 0.8166
Target = 35% Balochi + 19% Saudis + 39% Tajiks_Y + 7% Kurds_Y @ D = 1.0853
Target = 42% Jatt_D + 20% Saudis + 0% Armenians + 38% Kurds_Y @ D = 4.5135
jesus
05-13-2015, 05:37 AM
My Oracle 4 using the latest 4 way oracle posted on the Eurogenes blog:
Target = 34% Jatt_D + 35% Saudis + 30% Tajiks_Y + 1% Kurds_Y @ D = 0.8166
Target = 35% Balochi + 19% Saudis + 39% Tajiks_Y + 7% Kurds_Y @ D = 1.0853
Target = 42% Jatt_D + 20% Saudis + 0% Armenians + 38% Kurds_Y @ D = 4.5135
Pretty cool!
That's the closest result I managed to get till now:
[1] Target = 0% Bedouin + 8% Tadjik_Ishkashim + 87% Iranian + 4.99999999999999% Maasai @ D = 0.0188
Pretty cool!
That's the closest result I managed to get till now:
[1] Target = 0% Bedouin + 8% Tadjik_Ishkashim + 87% Iranian + 4.99999999999999% Maasai @ D = 0.0188
That is a good fit using the K8 spreadsheet.
My oracle was using the WEAC2 spreadsheet, and was not able to get a closer distance after about 10 mins of playing around with it.
Here are the accompanying dendograms. The Euclidean method was used for member to member distance measurement. I am posting 2 plots, one using the "furthest neighbor" method of intra clusteroid measurement, the other the "group averages" method. Please refer to tutorials/explanation of these methods that I have posted under other threads showing dendograms. The populations represent their average and were taken from the spreadsheet.
http://i.imgur.com/i67tWIC.jpg
http://i.imgur.com/nMPRdpu.jpg
5 additional individuals have been added to these similarity graphs. Will post other similarity graphs over the next couple of days
http://i.imgur.com/JrCGRDk.jpg
http://i.imgur.com/YY9uc7g.jpg
http://i.imgur.com/TOdwiNb.jpg
SwampThing27
05-13-2015, 01:52 PM
5 additional individuals have been added to these similarity graphs. Will post other similarity graphs over the next couple of days
http://i.imgur.com/JrCGRDk.jpg
http://i.imgur.com/YY9uc7g.jpg
http://i.imgur.com/TOdwiNb.jpg
Thanks for adding me to the similarity graphs, Kurd. Any idea where I fall on the two Dendograms?
Thanks for adding me to the similarity graphs, Kurd. Any idea where I fall on the two Dendograms?
I am hoping to redo the dendograms tonight
Roaring
05-13-2015, 05:26 PM
Rounded:
80% A-B
10% NE.
5% S. Asian.
5% NE Asian.
It is interesting that I am closer to Varun on the dendogram than BMG who is from the same state in India as me.
As promised here is the revised dendogram which includes 5 additional individuals. Also a couple of double posts were removed. The clusteroid position of some Iranian members changes depending on the intra cluseroid distance method used (furthest neighbor vs group average). For a description of these methods, please refer to my other posts on dendograms under other threads. I am also working on a couple of community similarity graphs which I hope to post later tonight.
http://i.imgur.com/Or9cuVe.jpg
http://i.imgur.com/YxH8jo4.jpg
The following are genetic admixture similarity graphs comparing community members against Iranians and Tajiks. Apparently Europeans have more overlap (greater similarity) with Tajiks than Iranians. Also, interesting that 2 kurds, Ezidi & MFA scored highest with regards to similarity with the Iranian average, 97% & 95%, respectively
http://i.imgur.com/4zDef1K.jpg
http://i.imgur.com/tUrCoIA.jpg
Hanna
05-14-2015, 11:35 AM
The following are genetic admixture similarity graphs comparing community members against Iranians and Tajiks. Apparently Europeans have more overlap (greater similarity) with Tajiks than Iranians. Also, interesting that 2 kurds, Ezidi & MFA scored highest with regards to similarity with the Iranian average, 97% & 95%, respectively
http://i.imgur.com/4zDef1K.jpg
http://i.imgur.com/tUrCoIA.jpg
I think it's more interesting that I match with Iranians! Why is that??
The Barnacle
05-14-2015, 12:57 PM
I think it's more interesting that I match with Iranians! Why is that??
Why so surprised?
Hanna
05-14-2015, 02:00 PM
Why so surprised?
Actually I am surprised that I match with Iranians better than other West Asians and Iranic members here but in my oracle results Iranians aren't my top match.
Gedmatch isn't working now so these are my oracle results I found on other posts:
eurogenes V2 K=15
Using 1 population approximation:
1 Armenian @ 6.993361
2 Georgian_Jewish @ 9.080214
3 Azeri @ 9.625549
4 Kurdish @ 9.632879
5 Turkish @ 9.765220
6 Iranian @ 12.351965
7 Assyrian @ 12.826490
8 Kumyk @ 17.646044
9 Iranian_Jewish @ 18.370588
10 Kurdish_Jewish @ 18.727787
11 Lebanese_Muslim @ 20.009659
12 Turkmen @ 20.779509
13 Georgian @ 20.786568
14 Syrian @ 22.107666
15 Adygei @ 22.302145
16 Balkar @ 23.344023
17 Cyprian @ 23.389896
18 Kabardin @ 23.513817
19 Lezgin @ 23.908424
20 Ossetian @ 23.967974
MDLP K23b
Using 1 population approximation:
1 Turk_Kayseri @ 4.266644
2 Turk @ 6.815643
3 Turk_Adana @ 8.286249
4 Armenian @ 8.361481
5 Georgian_Jew @ 8.882082
6 Armenian_Yerevan @ 9.504634
7 Assyrian_Iraqi @ 9.563898
8 Jew_Tat @ 10.325316
9 Uzbek_Tashkent @ 10.710565
10 Uzbekistani_Jew @ 10.758110
11 Turk_Istanbul @ 10.797637
12 Azeri @ 10.807470
13 Assyrian_Arzni @ 10.868814
14 Kurd_North @ 11.274456
15 Iraqi_Chaldean @ 11.923533
16 Cirkassian @ 12.250082
17 Turk_Aydin @ 12.944167
18 Kurd_Jew @ 13.072672
19 Baku_WGA @ 13.332445
20 Iranian_Jew @ 13.429023
Dodecad world 9
Using 1 population approximation:
1 Armenians @ 4.586259
2 Turks @ 5.264636
3 Armenian @ 8.838160
4 Armenians @ 9.140386
5 Turkish @ 9.358273
6 Uzbekistan_Jews @ 9.376534
7 Kurds @ 9.540112
8 Kurd @ 9.809800
9 Georgians @ 10.797691
10 Iranian @ 11.100307
11 Iranians @ 12.122344
12 Azerbaijan_Jews @ 12.189021
13 Assyrian @ 12.593275
14 Kumyks @ 13.106474
15 Abhkasians @ 13.382278
16 Georgia_Jews @ 13.564471
17 North_Ossetians @ 13.639699
18 Balkars @ 14.067990
19 Adygei @ 14.531388
20 Iranian_Jews @ 16.955582
Eurogenes K13
Using 1 population approximation:
1 Armenian @ 7.759
2 Turkish @ 8.184
3 Georgian_Jewish @ 9.597
4 Kurdish @ 9.628
5 Azeri_Jewish @ 10.889
6 Assyrian @ 12.395
7 Iranian @ 12.771
8 Kumyk @ 16.815
9 Georgian @ 17.892
10 Adygei @ 19.347
175 iterations.
Genetically I appear to be a Turkish/ Armenian hybrid.
The Barnacle
05-14-2015, 02:12 PM
Actually I am surprised that I match with Iranians better than other West Asians and Iranic members here but in my oracle results Iranians aren't my top match.
Gedmatch isn't working now so these are my oracle results I found on other posts:
eurogenes V2 K=15
Using 1 population approximation:
1 Armenian @ 6.993361
2 Georgian_Jewish @ 9.080214
3 Azeri @ 9.625549
4 Kurdish @ 9.632879
5 Turkish @ 9.765220
6 Iranian @ 12.351965
7 Assyrian @ 12.826490
8 Kumyk @ 17.646044
9 Iranian_Jewish @ 18.370588
10 Kurdish_Jewish @ 18.727787
11 Lebanese_Muslim @ 20.009659
12 Turkmen @ 20.779509
13 Georgian @ 20.786568
14 Syrian @ 22.107666
15 Adygei @ 22.302145
16 Balkar @ 23.344023
17 Cyprian @ 23.389896
18 Kabardin @ 23.513817
19 Lezgin @ 23.908424
20 Ossetian @ 23.967974
MDLP K23b
Using 1 population approximation:
1 Turk_Kayseri @ 4.266644
2 Turk @ 6.815643
3 Turk_Adana @ 8.286249
4 Armenian @ 8.361481
5 Georgian_Jew @ 8.882082
6 Armenian_Yerevan @ 9.504634
7 Assyrian_Iraqi @ 9.563898
8 Jew_Tat @ 10.325316
9 Uzbek_Tashkent @ 10.710565
10 Uzbekistani_Jew @ 10.758110
11 Turk_Istanbul @ 10.797637
12 Azeri @ 10.807470
13 Assyrian_Arzni @ 10.868814
14 Kurd_North @ 11.274456
15 Iraqi_Chaldean @ 11.923533
16 Cirkassian @ 12.250082
17 Turk_Aydin @ 12.944167
18 Kurd_Jew @ 13.072672
19 Baku_WGA @ 13.332445
20 Iranian_Jew @ 13.429023
Dodecad world 9
Using 1 population approximation:
1 Armenians @ 4.586259
2 Turks @ 5.264636
3 Armenian @ 8.838160
4 Armenians @ 9.140386
5 Turkish @ 9.358273
6 Uzbekistan_Jews @ 9.376534
7 Kurds @ 9.540112
8 Kurd @ 9.809800
9 Georgians @ 10.797691
10 Iranian @ 11.100307
11 Iranians @ 12.122344
12 Azerbaijan_Jews @ 12.189021
13 Assyrian @ 12.593275
14 Kumyks @ 13.106474
15 Abhkasians @ 13.382278
16 Georgia_Jews @ 13.564471
17 North_Ossetians @ 13.639699
18 Balkars @ 14.067990
19 Adygei @ 14.531388
20 Iranian_Jews @ 16.955582
Eurogenes K13
Using 1 population approximation:
1 Armenian @ 7.759
2 Turkish @ 8.184
3 Georgian_Jewish @ 9.597
4 Kurdish @ 9.628
5 Azeri_Jewish @ 10.889
6 Assyrian @ 12.395
7 Iranian @ 12.771
8 Kumyk @ 16.815
9 Georgian @ 17.892
10 Adygei @ 19.347
175 iterations.
Genetically I appear to be a Turkish/ Armenian hybrid.
I thought you meant compared to Tajiks my bad
Actually I am surprised that I match with Iranians better than other West Asians and Iranic members here but in my oracle results Iranians aren't my top match.
Gedmatch isn't working now so these are my oracle results I found on other posts:
eurogenes V2 K=15
Using 1 population approximation:
1 Armenian @ 6.993361
2 Georgian_Jewish @ 9.080214
3 Azeri @ 9.625549
4 Kurdish @ 9.632879
5 Turkish @ 9.765220
6 Iranian @ 12.351965
7 Assyrian @ 12.826490
8 Kumyk @ 17.646044
9 Iranian_Jewish @ 18.370588
10 Kurdish_Jewish @ 18.727787
11 Lebanese_Muslim @ 20.009659
12 Turkmen @ 20.779509
13 Georgian @ 20.786568
14 Syrian @ 22.107666
15 Adygei @ 22.302145
16 Balkar @ 23.344023
17 Cyprian @ 23.389896
18 Kabardin @ 23.513817
19 Lezgin @ 23.908424
20 Ossetian @ 23.967974
MDLP K23b
Using 1 population approximation:
1 Turk_Kayseri @ 4.266644
2 Turk @ 6.815643
3 Turk_Adana @ 8.286249
4 Armenian @ 8.361481
5 Georgian_Jew @ 8.882082
6 Armenian_Yerevan @ 9.504634
7 Assyrian_Iraqi @ 9.563898
8 Jew_Tat @ 10.325316
9 Uzbek_Tashkent @ 10.710565
10 Uzbekistani_Jew @ 10.758110
11 Turk_Istanbul @ 10.797637
12 Azeri @ 10.807470
13 Assyrian_Arzni @ 10.868814
14 Kurd_North @ 11.274456
15 Iraqi_Chaldean @ 11.923533
16 Cirkassian @ 12.250082
17 Turk_Aydin @ 12.944167
18 Kurd_Jew @ 13.072672
19 Baku_WGA @ 13.332445
20 Iranian_Jew @ 13.429023
Dodecad world 9
Using 1 population approximation:
1 Armenians @ 4.586259
2 Turks @ 5.264636
3 Armenian @ 8.838160
4 Armenians @ 9.140386
5 Turkish @ 9.358273
6 Uzbekistan_Jews @ 9.376534
7 Kurds @ 9.540112
8 Kurd @ 9.809800
9 Georgians @ 10.797691
10 Iranian @ 11.100307
11 Iranians @ 12.122344
12 Azerbaijan_Jews @ 12.189021
13 Assyrian @ 12.593275
14 Kumyks @ 13.106474
15 Abhkasians @ 13.382278
16 Georgia_Jews @ 13.564471
17 North_Ossetians @ 13.639699
18 Balkars @ 14.067990
19 Adygei @ 14.531388
20 Iranian_Jews @ 16.955582
Eurogenes K13
Using 1 population approximation:
1 Armenian @ 7.759
2 Turkish @ 8.184
3 Georgian_Jewish @ 9.597
4 Kurdish @ 9.628
5 Azeri_Jewish @ 10.889
6 Assyrian @ 12.395
7 Iranian @ 12.771
8 Kumyk @ 16.815
9 Georgian @ 17.892
10 Adygei @ 19.347
175 iterations.
Genetically I appear to be a Turkish/ Armenian hybrid.
Hanna, although your oracles show you to have a greater similarity with Armenians and Turks, you do have a high genetic similarity (-90%) with the Iranian average which is probably based on mostly N/NW Iranians. The Iranian members who have slightly less similarity, khorasanian Irani, Jesus, myself, and my daughter are probably gypsies.....just kidding :)
My daughterclearly has 25% Punjabi admixture. With regards to khorasani, jesus, and myself, if the average had included individuals from southern and eastern Iran, then our similarity would be quite a bit higher.
NK19191
05-14-2015, 03:23 PM
Here is mine:
00.01% Palaeoafrican
20.77% Atlantic_Baltic
02.92% Northeast_Asian
54.37% Near_East
00.48% Sub_Saharan
20.81% South_Asian
00.64% Southeast_Asian
http://i1309.photobucket.com/albums/s632/nk191919/WEAC2_zpshlucyyeq.png
Hanna
05-14-2015, 05:03 PM
Hanna, although your oracles show you to have a greater similarity with Armenians and Turks, you do have a high genetic similarity (-90%) with the Iranian average which is probably based on mostly N/NW Iranians. The Iranian members who have slightly less similarity, khorasanian Irani, Jesus, myself, and my daughter are probably gypsies.....just kidding :)
My daughterclearly has 25% Punjabi admixture. With regards to khorasani, jesus, and myself, if the average had included individuals from southern and eastern Iran, then our similarity would be quite a bit higher.
Even if they are Northern Iranians I still find it interesting but then it could be due to less K values. The more the Ks the less the similarity due to breakdown.
Could you do the same similarity chart with West Asian populations? Like Turks and Armenians.
NK19191 & Roaring have been added. I have also posted clusteroid info for the Group Averages clusteroid distance measurement method.
@Hanna, I posted the similarity graph with Armenians for you. Your genetic similarity at k7 is 91% with Armenian as well as Iranian. I guess that makes you genetically as much Armenian as you are Iranian :)
http://i.imgur.com/w1W6Nu5.jpg
http://i.imgur.com/rM9J03S.jpg
http://i.imgur.com/axGmoPI.jpg
http://i.imgur.com/vNe42Yc.jpg
http://i.imgur.com/I7CHNvR.jpg
Here is the PCA updated with additional members. Fpr eigenvectors & eigenvalues refer to the earlier post.
http://i.imgur.com/I5M9kJi.jpg
Hanna
05-15-2015, 01:54 PM
@Hanna, I posted the similarity graph with Armenians for you. Your genetic similarity at k7 is 91% with Armenian as well as Iranian. I guess that makes you genetically as much Armenian as you are Iranian :)
This is interesting so I am Turkish, Armenian and Iranian:p
That means Iranians and Armenians aren't very distinct in this calculator run. I would probably match Turks at around 95%.
Thanks for all your effort in making these graphs and PCAs :)
This is interesting so I am Turkish, Armenian and Iranian:p
That means Iranians and Armenians aren't very distinct in this calculator run. I would probably match Turks at around 95%.
Thanks for all your effort in making these graphs and PCAs :)
You are correct, I believe you were at 96% with Turks. At this level of k the distinction between Armenian and Iranian component is not very clear. On a simplified level low ks can be viewed as a snapshot of admixture further back in time, before the alleles associated with various components mutated and formed their own sub components.
If you look at the PCA you will notice that your distance from Iranian is not that different from your distance to the Armenian average
MJost
05-15-2015, 02:40 PM
@Kurd,
I am not seeing a close link for Europeans and the remaining groups when looking at your WEAC2 Group Averages Dendrogram. The chart does show an early autosomal split from the Chinese into southern and western Asia branch with European. Would you take time to discuss?
MJost
Passa
05-15-2015, 03:43 PM
Palaeoafrican 0.01%
AtlanticBaltic 46.91%
NortheastAsian -
NearEast 48.35%
SubSaharan 0.84%
SouthAsian 3.62%
SoutheastAsian 0.26%
4584
@Kurd,
I am not seeing a close link for Europeans and the remaining groups when looking at your WEAC2 Group Averages Dendrogram. The chart does show an early autosomal split from the Chinese into southern and western Asia branch with European. Would you take time to discuss?
MJost
Don't interpret the distance of 0.75 to be the distance from the European to the W Asian or Med cluster. It is the distance from the European to the average of the W Asian, C Asian, Med, and S Asian clusters. I am sure if I were to remove the S Asian, C Asian, and some W Asian members, the distance from the European cluster to the remaining avg of the Med and W Asian clusters would considerably decrease. I may do that to illustrate this point.
The same applies to the other cluster/clusteroids. As an example, you will notice that member Jesus is part of the SC Asian cluster using the group average method, and the distance from his clusteroid to the other SC Asian clusteroids does not exceed 0.2. However, using the furthest neighbor method, he is pulled to the W Asian cluster by Iranian Hamadan. Does that mean that his distance to the SC Asian cluster has now increased to 1.1. Of course not, the distance of 1.1 should be interpreted as the distance from the European cluster to the S Asian cluster (furthest neighbor). In all likelihood Jesus and Iranian K would be pulled back to the SC Asian cluster by me if I were to remove Iranian H
ashwinb
05-17-2015, 10:28 PM
Here's mine:
AtlanticBaltic 10.42
NortheastAsian 2.37
Near East 15.28
South Asian 66.25
Southeast Asian 5.66
The Euclidean distance is a very realistic distance measurement between 2 vectors in n dimensional space. A member's admixture result can be considered a vector with n admixture comp values. I have come up with a method to perform the computations of Euclidean distance between a sample admixture result and a set of references relatively quickly, and maybe incorporating this into analyses moving forward. I could probably do the computation on 20 members within minutes.
I took my WEAC2 results and compared them against the calculator references, and against other members. The ones against calculator references will be limited by the calculator effect as is the case with any oracle. This may become problematic with high k calculators as the calculator effect magnifies with smaller FST distances between comps, but for low k it should not be too bad. The ones against other members obviously do not suffer from the calculator effect.
Here is what I got:
http://i.imgur.com/8vvD8Xy.jpg
I will be posting Euclidean distance measurements for a few members tonight
rockman
05-23-2015, 12:29 AM
Here is mine
4619
The Euclidean distance is a very realistic distance measurement between 2 vectors in n dimensional space. A member's admixture result can be considered a vector with n admixture comp values. I have come up with a method to perform the computations of Euclidean distance between a sample admixture result and a set of references relatively quickly, and maybe incorporating this into analyses moving forward. I could probably do the computation on 20 members within minutes.
I took my WEAC2 results and compared them against the calculator references, and against other members. The ones against calculator references will be limited by the calculator effect as is the case with any oracle. This may become problematic with high k calculators as the calculator effect magnifies with smaller FST distances between comps, but for low k it should not be too bad. The ones against other members obviously do not suffer from the calculator effect.
Here is what I got:
http://i.imgur.com/8vvD8Xy.jpg
Here are a few more members. I am running batches, so the next few will be in an hour or two.
MEMBER DISTANCE FROM ZAHRA D
RUKHA 0.06
KURD D 0.09
ADAM 0.12
KURD 0.15
SAPPORO 0.21
JESUS 0.22
BULUT 0.28
EZIDI 0.30
MFA 0.31
VARUN 0.34
HANNA 0.34
FIL 0.35
BMG 0.41
J DOE 0.42
ANABASIS 0.43
ZEPH 0.45
VETTOR 0.50
TOLAN 0.62
JTOM 0.62
RAND 0.63
RAND W 0.64
CHAD 0.68
FIREMO 0.69
CELTICG 0.73
CELTICG U 0.75
MEMBER DISTANCE FROM SAPPORO
ADAM 0.09
VARUN 0.15
RUKHA 0.18
FIL 0.20
ZAHRA D 0.21
BMG 0.23
KURD D 0.25
KURD 0.34
JESUS 0.41
BULUT 0.49
EZIDI 0.50
MFA 0.51
HANNA 0.55
J DOE 0.62
ANABASIS 0.63
ZEPH 0.64
VETTOR 0.65
JTOM 0.72
TOLAN 0.73
RAND 0.73
RAND W 0.75
CHAD 0.77
FIREMO 0.77
CELTICG 0.81
CELTICG U 0.83
MEMBER DISTANCE FROM JESUS
KURD 0.08
EZIDI 0.12
MFA 0.13
KURD D 0.17
HANNA 0.18
BULUT 0.21
ZAHRA D 0.22
ANABASIS 0.24
ZEPH 0.25
RUKHA 0.28
ADAM 0.33
J DOE 0.35
SAPPORO 0.41
FIL 0.50
VARUN 0.52
VETTOR 0.53
BMG 0.57
TOLAN 0.68
RAND 0.69
JTOM 0.70
RAND W 0.71
CHAD 0.76
FIREMO 0.77
CELTICG 0.82
CELTICG U 0.84
MEMBER DISTANCE FROM ADAM
SAPPORO 0.09
RUKHA 0.11
ZAHRA D 0.12
KURD D 0.17
VARUN 0.22
FIL 0.25
KURD 0.26
BMG 0.30
JESUS 0.33
BULUT 0.40
EZIDI 0.41
MFA 0.43
HANNA 0.47
J DOE 0.54
ANABASIS 0.54
ZEPH 0.56
VETTOR 0.59
JTOM 0.68
TOLAN 0.68
RAND 0.69
RAND W 0.70
CHAD 0.73
FIREMO 0.74
CELTICG 0.78
CELTICG U 0.80
MEMBER DISTANCE FROM HANNA
MFA 0.07
EZIDI 0.09
ANABASIS 0.11
BULUT 0.14
ZEPH 0.15
JESUS 0.18
KURD 0.22
J DOE 0.23
KURD D 0.31
ZAHRA D 0.34
RUKHA 0.39
VETTOR 0.46
ADAM 0.47
SAPPORO 0.55
TOLAN 0.62
RAND 0.64
JTOM 0.65
RAND W 0.65
FIL 0.66
VARUN 0.68
CHAD 0.70
FIREMO 0.72
BMG 0.73
CELTICG 0.77
CELTICG U 0.80
MEMBER DISTANCE FROM JTOM
RAND 0.04
RAND W 0.04
TOLAN 0.04
CHAD 0.06
FIREMO 0.07
CELTICG 0.12
CELTICG U 0.15
VETTOR 0.19
J DOE 0.46
BULUT 0.53
RUKHA 0.59
ZAHRA D 0.62
HANNA 0.65
KURD 0.66
MFA 0.68
ADAM 0.68
KURD D 0.68
EZIDI 0.69
JESUS 0.70
SAPPORO 0.72
ANABASIS 0.73
ZEPH 0.78
VARUN 0.84
FIL 0.90
BMG 0.92
MEMBER
DISTANCE FROM RUKHA
ZAHRA D
0.06
ADAM
0.11
KURD D
0.14
SAPPORO
0.18
KURD
0.20
JESUS
0.28
BULUT
0.31
VARUN
0.32
EZIDI
0.35
FIL
0.35
MFA
0.36
HANNA
0.39
BMG
0.40
J DOE
0.44
ANABASIS
0.47
VETTOR
0.48
ZEPH
0.50
JTOM
0.59
TOLAN
0.59
RAND
0.60
RAND W
0.61
CHAD
0.64
FIREMO
0.65
CELTICG
0.69
CELTICG U
0.71
The Euclidean distance is a very realistic distance measurement between 2 vectors in n dimensional space. A member's admixture result can be considered a vector with n admixture comp values. I have come up with a method to perform the computations of Euclidean distance between a sample admixture result and a set of references relatively quickly, and maybe incorporating this into analyses moving forward. I could probably do the computation on 20 members within minutes.
I took my WEAC2 results and compared them against the calculator references, and against other members. The ones against calculator references will be limited by the calculator effect as is the case with any oracle. This may become problematic with high k calculators as the calculator effect magnifies with smaller FST distances between comps, but for low k it should not be too bad. The ones against other members obviously do not suffer from the calculator effect.
Here is what I got:
http://i.imgur.com/8vvD8Xy.jpg
MEMBER
DISTANCE FROM RAND
RAND W
0.02
TOLAN
0.02
JTOM
0.04
CHAD
0.07
FIREMO
0.09
CELTICG
0.14
CELTICG U
0.16
VETTOR
0.18
J DOE
0.44
BULUT
0.52
RUKHA
0.60
ZAHRA D
0.63
HANNA
0.64
KURD
0.66
MFA
0.67
EZIDI
0.68
KURD D
0.68
ADAM
0.69
JESUS
0.69
ANABASIS
0.72
SAPPORO
0.73
ZEPH
0.77
VARUN
0.85
FIL
0.92
BMG
0.93
MEMBER
DISTANCE FROM ZEPH
ANABASIS
0.05
MFA
0.14
HANNA
0.15
EZIDI
0.15
JESUS
0.25
BULUT
0.29
KURD
0.31
J DOE
0.34
KURD D
0.40
ZAHRA D
0.45
RUKHA
0.50
ADAM
0.56
VETTOR
0.59
SAPPORO
0.64
FIL
0.72
TOLAN
0.75
VARUN
0.75
RAND
0.77
JTOM
0.78
RAND W
0.78
BMG
0.80
CHAD
0.84
FIREMO
0.85
CELTICG
0.90
CELTICG U
0.93
The Euclidean distance is a very realistic distance measurement between 2 vectors in n dimensional space. A member's admixture result can be considered a vector with n admixture comp values. I have come up with a method to perform the computations of Euclidean distance between a sample admixture result and a set of references relatively quickly, and maybe incorporating this into analyses moving forward. I could probably do the computation on 20 members within minutes.
I took my WEAC2 results and compared them against the calculator references, and against other members. The ones against calculator references will be limited by the calculator effect as is the case with any oracle. This may become problematic with high k calculators as the calculator effect magnifies with smaller FST distances between comps, but for low k it should not be too bad. The ones against other members obviously do not suffer from the calculator effect.
[SIZE=3]
MEMBER
DISTANCE FROM FIL
BMG
0.08
VARUN
0.10
SAPPORO
0.20
ADAM
0.25
RUKHA
0.35
KURD D
0.35
ZAHRA D
0.35
KURD
0.44
JESUS
0.50
EZIDI
0.59
MFA
0.61
BULUT
0.62
HANNA
0.66
ANABASIS
0.72
ZEPH
0.72
J DOE
0.77
VETTOR
0.82
JTOM
0.90
TOLAN
0.91
RAND
0.92
RAND W
0.93
CHAD
0.95
FIREMO
0.95
CELTICG
0.99
CELTICG U
1.01
I have posted results for some more members on the previous page
MEMBER
DISTANCE FROM TOLAN
RAND
0.02
RAND W
0.04
JTOM
0.04
CHAD
0.08
FIREMO
0.10
CELTICG
0.15
VETTOR
0.16
CELTICG U
0.18
J DOE
0.43
BULUT
0.51
RUKHA
0.59
ZAHRA D
0.62
HANNA
0.62
KURD
0.65
MFA
0.66
EZIDI
0.67
KURD D
0.67
ADAM
0.68
JESUS
0.68
ANABASIS
0.70
SAPPORO
0.73
ZEPH
0.75
VARUN
0.85
FIL
0.91
BMG
0.93
MEMBER
DISTANCE FROM VETTOR
TOLAN
0.16
RAND
0.18
JTOM
0.19
RAND W
0.20
CHAD
0.25
J DOE
0.26
FIREMO
0.27
CELTICG
0.31
CELTICG U
0.34
BULUT
0.36
HANNA
0.46
RUKHA
0.48
MFA
0.50
ZAHRA D
0.50
EZIDI
0.51
KURD
0.51
JESUS
0.53
ANABASIS
0.54
KURD D
0.55
ADAM
0.59
ZEPH
0.59
SAPPORO
0.65
VARUN
0.78
FIL
0.82
BMG
0.85
basmaci
05-26-2015, 12:31 PM
0.07% Palaeoafrican
32.24% Atlantic_Baltic
7.53% Northeast_Asian
44.42% Near_East
0.00% Sub_Saharan
11.46% South_Asian
4.28% Southeast_Asian
Paleo-African 0
Atlantic_Baltic 20.2
Northeast_Asian 3.8
Near_East 54.2
Sub_Saharan 0
South_Asian 21.1
Southeast_Asian 0.8
vettor
05-29-2015, 05:56 AM
I have posted results for some more members on the previous page
MEMBER
DISTANCE FROM TOLAN
RAND
0.02
RAND W
0.04
JTOM
0.04
CHAD
0.08
FIREMO
0.10
CELTICG
0.15
VETTOR
0.16
CELTICG U
0.18
J DOE
0.43
BULUT
0.51
RUKHA
0.59
ZAHRA D
0.62
HANNA
0.62
KURD
0.65
MFA
0.66
EZIDI
0.67
KURD D
0.67
ADAM
0.68
JESUS
0.68
ANABASIS
0.70
SAPPORO
0.73
ZEPH
0.75
VARUN
0.85
FIL
0.91
BMG
0.93
MEMBER
DISTANCE FROM VETTOR
TOLAN
0.16
RAND
0.18
JTOM
0.19
RAND W
0.20
CHAD
0.25
J DOE
0.26
FIREMO
0.27
CELTICG
0.31
CELTICG U
0.34
BULUT
0.36
HANNA
0.46
RUKHA
0.48
MFA
0.50
ZAHRA D
0.50
EZIDI
0.51
KURD
0.51
JESUS
0.53
ANABASIS
0.54
KURD D
0.55
ADAM
0.59
ZEPH
0.59
SAPPORO
0.65
VARUN
0.78
FIL
0.82
BMG
0.85
:) interesting my closest is Tolan .......a Brittany "Veneti" ( well, in the same area ) and to me, a Adiatic Veneti .............:eyebrows:
Helgenes50
05-31-2015, 06:27 AM
Here are WEAC2 family results for comparison (German, French ...): 4553
All your results are interesting, mine and those of my cousin are very close to those of your children, East French/ North German and very different to the French Average.
Mine
0.00% Palaeoafrican
76.09% Atlantic_Baltic
0.10% Northeast_Asian
21.55% Near_East
0.58% Sub_Saharan
1.66% South_Asian
0.02% Southeast_Asian
First cousin
0.00% Palaeoafrican
78.34% Atlantic_Baltic
0.00% Northeast_Asian
20.90% Near_East
0.29% Sub_Saharan
0.46% South_Asian
0.00% Southeast_Asian
Powered by vBulletin® Version 4.2.5 Copyright © 2021 vBulletin Solutions Inc. All rights reserved.