Tribes scores

[tomcat]

Tom,

I noticed on the Kerchner Tribes blog that an askenazi fellow got top match to Greek cypriot. In fact, his results look a bit like yours if I am remembering them correctly.

Thank you NormanGalway!

I had a post on this forum looking for a Jewish R1b1c with a Tribes report. I emailed the tribeslog guy. He got a lot of North Africa for an Ashkenazi.

[tomcat]

I would like to know , when is the next update for TribeDNA, Does anyone know?
Thank you.

You might ask Tribes. They have added populations to their database regularly but not on a fixed schedule. Suppose we are 'due' for an update.

[Noaide]

9, 13, 15 and 20 loci CODIS data reality check!

I have done some tests with different levels of markers and I find quite interesting results. I use Germans (1), Han Chinese (2) and Guinea Bissau (3) (Africa) populations. Only Guinea Bissau missed some loci so I had to use other African populations as approximation.

Test 1: 20 loci (CSFPO, D2S1338, D3S1358, D5S818, D7S820, D8S1179, D13S317, D16S539, D18S51, D19S433, D21S11, F13A01, F13B, FESFPS, FGA, Penta-D, Penta-E, TH01, TPOX, VWA)

The 20 loci ability to place the person in the right population:

20 LOCI

1: 0.052 0.920 0.028
2: 0.944 0.045 0.012
3: 0.019 0.028 0.953

The chance that a individual will be put into the correct population is rather high using 20 loci. All over 0.92.

TEST 2: 15 loci (CSFPO, D3S1358, D5S818, D7S820, D8S1179, D13S317, D16S539, D18S51, D21S11, FGA, Penta-D, Penta-E, TH01, TPOX, VWA)

The 15 loci ability to place the person in the right population:

15 LOCI

1: 0.113 0.831 0.057
2: 0.871 0.103 0.025
3: 0.034 0.056 0.911

The chance that a individual will be put into the correct population drops using 15 loci, this is especially the case for the Germans that is the European proximation.

TEST 3: 13 loci (CSFPO, D3S1358, D5S818, D7S820, D8S1179, D13S317, D16S539, D18S51, D21S11, FGA, TH01, TPOX, VWA)

The 13 loci ability to place the person in the right population:

13 LOCI

1: 0.104 0.754 0.143
2: 0.056 0.136 0.808
3: 0.856 0.090 0.054

The chance that a individual will be put into the correct population drops using 13 loci, this is especially the case for the Germans

The 9 loci ability to place the person in the right population:

TEST4: 9 loci (D3S1358, D5S818, D7S820, D8S1179, D13S317, D18S51, D21S11, FGA, VWA)

1: 0.169 0.614 0.216
2: 0.078 0.206 0.716
3: 0.781 0.148 0.071

The chance that a individual will be put into the correct population drops using 9 loci, this is especially the case for the Germans. If it drops more for Germans you may instead safe your money and flip a coin to determine if your German or Chinese/African.

SUMMARY: Europeans and East Asians "floats" into each other while Africans largely seperate from each of the other groups.

Enclosed illustrative triangle plots, the 20 loci test is very distinctive while the 9 loci test is messy.

Noaide

[Noaide]

Forgot to mention the color codes for the plot figures:

1. RED - GERMANS (EUROPE)
2. GREEN - HAN CHINESE (EAST ASIA)
3- BLUE - GUINEA BISSAU / OTHER AFRICAN POPULATIONS (AFRICA)

Noaide

[tomcat]

Noaide,

Are you limited to comparing three population in STRUCTURE or can you compare more?

It seems that the power of discrimination of these various marker sets is not only a function of numbers of markers but also involves specific markers as evidenced by the following:

For both 20 and 15 sets, the best scores are 1.-2, 2.-1 and 3.-3 whereas for 13 and 9 the best scores are 1.-2, 2.-3 and 3.-1

Penta D and E seem quite discriminating. Have your tested the 13 CODIS with D2 and D19? Or the ENFSI 10-marker standard?

[Noaide]

As far as I know there is no population limitation in STRUCTURE. I will try testing how "good" the ENSFI system is compared to the rest and how 13 CODIS + the 2 additional from ENFSI compare.

Noaide

[Noaide]

10 Loci ENSFI

1: 0.137 0.740 0.123
2: 0.719 0.153 0.127
3: 0.117 0.114 0.769

13 CODIS Loci + 2 ENSFI

1: 0.077 0.806 0.117
2: 0.046 0.117 0.837
3: 0.872 0.078 0.050

The plots enclosed.

Noaide

[tomcat]

Thanks Noaide. It does seem that for admixture more markers are better and better all 'round.

[Kathleen Carrow]

My husband's Autosmal markers all came back tonight..
Bear in mind that he is almost all Scottish, both of his parents came to US as adults. A known Irish maternal great-grandmother and a maternal grandfather( Gunn) who married another of the same name from Norway, which is where Clan Gunn( Gunni) originated..
CSF1PO
10, 11
D13S317
11, 12
D16S539
11, 12
D18S51
11, 12
D21S11
30, 32.2
D3S1358
15, 15
D5S818
11, 13
D7S820
8, 13
D8S1179
12, 15
FGA
22, 23
Penta D
10, 13
Penta E
7, 12
TH01
6, 6
TPOX
8, 8
vWA
18, 19

ENSFI says this:
#1 Italy
#2 Estonia
#3 Spain
#4 Belgium
#5 Portugal
#6 Denmark
#7 Austria
#8 North Ireland
#9 Switzerland
#10 Poland
#11 Scotland Dundee

His mt DNA is U5

Comments??

[tomcat]

...
ENSFI says this:
#1 Italy
#2 Estonia
#3 Spain
#4 Belgium
#5 Portugal
#6 Denmark
#7 Austria
#8 North Ireland
#9 Switzerland
#10 Poland
#11 Scotland Dundee
...
Comments??

Mediterranean
Northeast
Mediterranean
Northwest
Mediterranean
Northwest
Southcentral
Northwest
Southcentral
Northeast
Northwest

Wonder what the addition of D2 and D19 would do the pattern? Is there a score gradient discontinuity at 3-4 to go along with the change of pattern, another at 6-7 and again at 9-10?

[tomcat]

KC,

Your husband's pattern of results remind me of yours (page 149 of this thread) as they are both multi-focal. This raises the question as to whether others of primarily 'Celtic' ancestry produce multi-focal patterns on ENFSI?

[Kathleen Carrow]

KC,

Your husband's pattern of results remind me of yours (page 149 of this thread) as they are both multi-focal. This raises the question as to whether others of primarily 'Celtic' ancestry produce multi-focal patterns on ENFSI?

Tom ,
Thanks also for the reference to pg. 149 which I must have missed..looks like I was posting something at the same time..
that is why I was so interested in his results because I think they DO look Celtic..and probably rightly so..his Ingram male line is in Aberdeen Scotland forever and the Gunn female line which comes from the Norse..also he has (unbenownst to him..I had his great aunt send me the paper genealogy)the Gunn line into Antrim and back and married more recently into the Norwegian Gunns.
My Swanson is Clan Gunn..the original Gunni is son of Swen..both septs supposedly point to the same person or family as origin in 1100 or so..
The original George Gunn is "Crowner" of Scotland

[tomcat]

He has Southcentral, Northeast and 'Celtic' foci like yours. With a longer list of ENFSI matches the same pattern might be more fully evident.

[Kathleen Carrow]

I am going to get his D2 and D19 ..it definitely changed my equation..it seemed to beef up the Irish and Scots for me..

[tomcat]

Even in my list of ENFSI matches and with no known 'Celtic' ancestry the 'Celtic' to South/Southeast pattern appears among my E15's -

#12 Norway to #13 Slovenia to #14 Ireland to #15 Toulouse to #16 Wales to #17 Croatia

Then the line wanders to Czech Rep. Austria, Poland, Germany, Estonia and Finland and goes to ground in Glasgow at E16. Except for the 'Celtic'-Balkan 'oscillation' across Europe and a zig-zag from Iberia to Scandinavia my pattern traces a path around the periphery of Europe.

See also the thread 'Italian Results'