Admixture or autosomal tests for percentages?

[rainbow]

KCat,

You didn't get many responses last night, and it could be because there has been much criticism of percentage-based testing.

Notice how quiet this results log has been: http://www.kerchner.com/cgi-kerchner/dnaprint.cgi

Jim

I had been looking at the dnaprint log and the dnatribes log again. I submitted my info for the dnatribes log in January, but it wasn't included.

[rainbow]

I was not specifically reffering to Tribes. I was reffering to autosomal tests in general, which I think it was what Rainbow was asking. Based on the many testimonies I have read from people who have taken the ABDNA test and Tribes and the many discrepancies they have seen, I tend to think that, up until now, many people may be even more confused after taking those tests. That is the reason why I am still waiting to see if there is a positive evolution in this field and tests become more universally reliable. I may have to wait a long time, it seems. I am always hoping to read about those who have had positive experiences with their results- in other words, that they make sense- but all I tend to see are what appear to be well-grounded complaints and lots of disappointed testees.

Thank you Juan Carlos. I am happy someone understands what I'm talking about.

[juan carlos]

Thank you Juan Carlos. I am happy someone understands what I'm talking about.

Yes, I understand and you give a very good example of these puzzling results. I mean 17% Native American and there are not even NAs among your ancestors! Consider that, in order to have that percentage, you should at least have a NA great-granparent or even a grandparent. Or several ancestors having some NA genes in such amounts they would have known about it. So I think these results should be not be taken that seriously. I know this is hard to do when you have spent some of your hard-earned money on a test which you expect to make sense, but raises more questions than before.

[rainbow]

Yes, I understand and you give a very good example of these puzzling results. I mean 17% Native American and there are not even NAs among your ancestors! Consider that, in order to have that percentage, you should at least have a NA great-granparent or even a grandparent. Or several ancestors having some NA genes in such amounts they would have known about it. So I think these results should be not be taken that seriously. I know this is hard to do when you have spent some of your hard-earned money on a test which you expect to make sense, but raises more questions than before.

Right. And to get Mozambique from DNATRIBES?!?
The extended report from DNATRIBES is even more bizzarre. It lists Maori, Javanese, Australian Aboriginal, West Polynesian, Oriya Brahmin, Egypt, Fang (Bioko Island), Canary Islands, Cape Verde Islands, South Africa (North & South Sotho), Kathmandu, and much more. Bizzarre.
I can understand all the European matches. I match every single European country/category on the map, except Lithuania.

[Paulie]

Mozambique has it's official language as Portuguese.

So you're probably seeing that some persons of Portugal decent live in Mozambique and you happen to have a similar match on markers. Does this mean you have Portugal ancestry? Maybe, maybe not. Could just be convergence (aka Luck).

I have a similar problem; I'm R1b1c and H, and I have very high scores for Mestizo and Iceland (and Nordic countries). I think my results show more about the Mestizo population admixture than it tells me anything... And no known Nordic ancestry...

Remember, the results tell you what populations your marker values are in, but that doesn't tell you what populations you actually BELONG to. That's Genealogy.

[Paulie]

Oh, I forgot to mention. I also have a high match in Guinea-Bissau, another African Portuguese speaking country.

[rainbow]

Oh, I forgot to mention. I also have a high match in Guinea-Bissau, another African Portuguese speaking country.

My original dnatribes report has Maputo, Mozambique and Mozambique.

My extended report from dnatribes also has Guinea-Bissau, and a bunch of other African matches. I wish the extended report was organized in separate parts, so I can just read the part that is an extension of "Part B" of the original. I wish the diaspora groups and the native groups weren't all lumped together like this. I emailed dnatribes about my "Amazon Region" asking if I really matched the Amazonian Indians, or if I was matched to that category because of my Spain & Portugal matches. The response seemed to be it was because of the latter. This was after I already went around telling people I matched Amazonian and a bunch of other things.

I thought of starting a thread asking who else got these African matches. Do you think I should?

[Paulie]

I thought of starting a thread asking who else got these African matches. Do you think I should?

If the countries are former European colonies, at least for autosomal tests using CODIS markers, it's probably a dead end.

Going back more than a couple generations is best left to y-dna and mt-dna. The "great middle" can only be found by testing known relatives.

[rainbow]

If the countries are former European colonies, at least for autosomal tests using CODIS markers, it's probably a dead end.

A dead end? The odd matches in my dnatribes report were European colonies. Mozambique was a Portuguese colony. Java was a Dutch colony.

[Paulie]

It's a dead end if you're expecting to be able to show any native ancestry for you from those countries. As your signature suggests. And you probably won't be able to show the european side either, as I mentioned earlier; What populations your makers match to does not indicate populations you belong to.

[rainbow]

Here is a quote from an email response I received from dnatribes.:

Thank you for contacting us regarding your results. I have reviewed all files and verified that the reports delivered to you by email include analysis for your own DNA sample labeled with your name and unique tracking number.

DNA Tribes analysis differs substantially from freely available programs such as OmniPop, and we cannot warrant or interpret OmniPop results. Our own algorithms were developed for DNA Tribes by Eduardas Valaitis (Ph.D. in Statistics, Yale University). We are not familiar with the algorithms which delivered your percentages below, but they are not consistent with results identified with the DNA Tribes algorithms.

DNA Tribes results identify a primarily Northwest European affiliation for your DNA profile, consistent with your known ancestry primarily from the British Isles.

[tomcat]

... I wish the extended report was organized in separate parts, so I can just read the part that is an extension of "Part B" of the original. I wish the diaspora groups and the native groups weren't all lumped together like this....

I was able to copy-and-paste from my Extended report into a word processing page and then, through much cut-and-pasting, cluster matches by geographic region, align diverse matches by score, and segregate diaspora/admixed populations.

[Jim Honeychuck]

I was able to copy-and-paste from my Extended report into a word processing page and then, through much cut-and-pasting, cluster matches by geographic region, align diverse matches by score, and segregate diaspora/admixed populations.

I have also found that to be an excellent way to try to make sense of DNA Tribe scores.

For example, the following set of Italian scores would seem to indicate that the individual (me) is not really very Italian at all, and is more likely to share a pre-Germanic ancestry with the Lombards:

6 Italy (0.7) 13.13
21 Emilia-Romagna, Italy (0.47) 6.91
24 Lombardia, Italy (0.45) 6.57
48 Veneto, Italy (0.35) 4.25
53 Italy (0.38) 4.00
57 Lazio, Italy (0.41) 3.81
67 Piemonte, Italy (0.33) 3.33
74 Liguria, Italy (0.44) 3.12
91 Sicily, Italy (0.37) 2.69
103 Marche, Italy (0.28) 2.45
130 Piedmont, Italy (0.27) 1.74
136 Toscana, Italy (0.16) 1.54
153 Toscana, Italy (0.13) 1.36
162 Calabria, Italy (0.14) 1.16
170 Campania, Italy (0.13) 1.05
195 Puglia, Italy (0.24) 0.82
196 Umbria, Italy (0.21) 0.78
199 Sicilia, Italy (0.08) 0.75
228 Catanzaro, Italy (0.15) 0.45
290 Allia, Sicilia, Italy (0.22) 0.13
374 Genoa, Italy (0.03) 0.03
444 Milan, Italy (0.03) 0.01
462 Rome, Italy (0.03) 0.00
508 Naples, Italy (0.21) 0.00
566 Venice, Italy (0) 0.00


Jim
no known Italian ancestry

[tomcat]

...cluster matches by geographic region, align diverse matches by score, and segregate diaspora/admixed populations.

My aim was to see if a rearrangement would produce any insight. I chose a simple cross axis scheme where matches of similar geography are ranked in a column by score and aligned horizontally by score with matches of other geography in other columns. There may be other arrangements that could be more useful.

[Jambalaia32]

Blood Types inherited from Mother????

Meant to say that it was an admixture test, giving percentages!
Maria

Are blood types inherited from the mother? Seems so. Yours is and mine too. Or maybe it's just chance.