ABDNA Test Results...

[Maria_W]

Title for a Sci Fi movie...

WOW! 1/1024th, thats even more invisible! Just call me the Invisible women!

Maria

[DKF]

My ABDNA results were so off for my whole family. Unless I find another my recent line my ancestry goes back to around 1640. About 370 years or 10 literal generations or my 8th great grandfather. If I kept dividing it to 10 generations
(100,4/4)
(50,1/2)
(25,1/4)
(12.5,1/8)
(6.25,1/16)
(3.125,1/32)
(1.56,1/164)
(0.780,1/112)
(0.390,1/224)
(0.195, 1/512)

That is 1/512th! Now that I see it like this it is totally invisible! I call it ancestry...
Some Tribes in the US require any where from 1/2 to1/16. The Choctaw of Mississippi require 1/2, the Cherokee of North Carolina require at least 1/16.
Some tribes require any degree decent from a tribal member, such as the Choctaw of Oklahoma, the Chickasaw, etc.
That pretty much settles that. Well I still have the Potowomeke ancesrty and I am very proud of it....

Maria

Is there really an average age of 37 years between generations? That would be somewhat unusual. I did tha math for a 10 th great grandparent and I get 1 / 1024 (of course match is not my strong suit). No matter how you slice it, using any sort of DNA test to locate any genomic bits and pieces that may have come from one single ancestry from 1640 is challenging. The only hope is that during the crossover process, via linkage disequilibrium, by chance you happened to inherit a number of intact haplotype blocks from the NA ancestor. It is a long shot but possible. This is why I have tested family members who are closer in generation time to my NA ancestry. The probability of these haploblocks showing up in them is higher than it is for me. You half the chances with each succeeding generation.

You can see why some of the figures that DNAP generated for your family members are simply not realistic in light of the reality of an 8th great grandparent. Unless you can rule out Mediterranean and west Asian (e.g., Pakistani) heritage then some of the results may be coming from this source (both of which show substantial NA on the DNAP test).

Of the million or so SNPs tested by decodeme, only a small subset are going to be in any way ancestrally informative. Virtually all marker variants are found in all populations, but in some groups the percentage is considerably higher (e.g., 70% of that group has a T allele at such and such a marker). Alas, say 15% of those in a group in another continent also have the T allele so the only thing one can do is use an algorithm assigning weights and probabilities. The Duffy null variant is the only marker exclusive to one population that I know of (Africans). The D9S1120 STR (not SNP) 9 repeat variant is exclusive to Native Americans (however only about a third), but this is not at present being used by any company in a biogeographical test.

For a biogeographical test to work as well as we hope for you would need SNPs, STRs, and indels (insertions - deletions). Lets hope that some company uses the data out there to construct a test with all three components. It is only then that one might be able to speak with confidence about small percentage ancestry from whatever source.

[Maria_W]

Cool

I know some people say 20 years for a generation but I literally mean there is only 10 generations between me and my 8th great grandfather...

Cool about DeCodeMe!

Maria

[Maria_W]

Totally invisible!

My ABDNA results were so off for my whole family. Unless I find another my recent line my ancestry goes back to around 1640. About 370 years or 10 literal generations or my 8th great grandfather. If I kept dividing it to 10 generations
(100,4/4)
(50,1/2)
(25,1/4)
(12.5,1/8)
(6.25,1/16)
(3.125,1/32)
(1.56,1/164)
(0.780,1/112)
(0.390,1/224)
(0.195, 1/512)

That is 1/512th! Now that I see it like this it is totally invisible! I call it ancestry...
Some Tribes in the US require any where from 1/2 to1/16. The Choctaw of Mississippi require 1/2, the Cherokee of North Carolina require at least 1/16.
Some tribes require any degree decent from a tribal member, such as the Choctaw of Oklahoma, the Chickasaw, etc.
That pretty much settles that. Well I still have the Potowomeke ancesrty and I am very proud of it....

Maria

[DKF]

Didn't realize that DeCodeMe's main focus was medical. Although I am still glad I did it....

Maria

Yes, so am I. I note on their FAQ that they are in the process of developing an autosomal ancestry test so perhaps my criticism was a bit hasty.

[Maria_W]

Didn't know...

Didn't realize that DeCodeMe's main focus was medical. Although I am still glad I did it....

Maria

[Maria_W]

Not a total waste

Well at least I have of copy of my families genotypes from ABDNA on disc if I ever want to compare with each other.

Maria

[DKF]

What about DeCodeMe results....I am 87-92 European, 6-9 Asian, and 2-4 African. Is that even correct?

Maria

Every person of European descent (those born in Europe) has some percentage of East Asian and African on decodeme. They have a long way to go to "get it right" - their focus is on disease and ancestry comes in a very poor second (or rather 5th).

[GregKiroKHR1bL1]

I have been reading a few papers about how Asian and Native American DNA are similar.

Athena Review Vol. 3. No. 2 (2002)

Annuals of Human Genetics (2003) 67, 391 - 411

Achilli A, Perego UA, Bravi CM, Coble MD, Kong Q-P, et al (2008) The Phylogeny of the Four Pan-American MtDNA Haplogroups: Implications for
Evolutionary and Disease Studies. PLoS ONE 3(3): e1764. doi:10.1371/journal.pone.0001764

The mtDNA HVR-I region just is not detailed enough, even though it is an interesting beginning. The full scans are much more intriguing. And then there is not a good way of interpreting all the values. It is even more impossible to understand with autosomes. I am glad I did some work. However, it is going to take years to understand all of this data as related to populations.

What about DeCodeMe results....I am 87-92 European, 6-9 Asian, and 2-4 African. Is that even correct?

Maria

[DKF]

DKF,
It is strange that my mother has 7%, me 10%, daughter 0%, daughter 14% and grandaughter 9%. They are all over the place arent they..Maybe this is an expensive experiment! Don't know if I want to put anymore money into testing my other family members.

Maria

In my case the NA ancestry is closer to the present so I have found distant cousins closest to (fewest generations removed) the NA ancestor and tested them. I have tested family members from as many different lines as possible (so far 3 of the four children). In this way I maximize the likelihood of the NA showing up in some famiily lineage. Of course the danger of false positives (that the 20% is "no real") or false negatives (that the zero percent is incorrect). To date everyone has at least a small amount (down to me with 3 percent), if not NA then East Asian. Testing your children and grandchildren is not, in my opinion, going to give you useful information, only false positives or true positives related to ancestry from the other sides of their family.

For ancestors "way back" it would be very very unlikely to find genetic evidence of one individual from the 1600s, but with the decodeme test using 18000 ancestral informative markers it becomes at least within the realm of possibility. However even here they are only using SNPs and we are infering (guessing) about haploblocks (pieces of DNA inherited as a unit of say 3 million bases). I am afraid that only a full genome scan is going to work in your case (and for many of us). Once this becomes routine for less than $1000 and the scientists begin finding haploblocks that relate to one geographical region only (e.g., the Americas) the location of any bits and pieces of DNA from an ancestor from the 1600s enters the realm of possibility.

With present tests - I would save my money for what is coming down the pike. For now lets hope that decodeme takes an interest in the concept of haploblocks. However they are only using SNPs so I am not sure how well their output of one million SNPs is going to work. We really need to see the STRs (short tandem repeats) and indels (insertions and deletions) included in the mix and that is not going to happen until a full genome scan becomes available.

[Maria_W]

I will disreard ABDNA results...

Since I already have the proven 1600's decent and hence my whole family does, I don't need the flawed test results from ABDNA!!!

Maria

[Maria_W]

ABDNA totally flawed....?

I have proven Amerindian ancestry (Potowomele)back to VA in the 1600's. I have known that for a long time before the ABDNA test was taken....So this isn't what is being questioned...So when the ABDNA test came out I thought that it might back it up... Well I soon found out that you either love it or hate it with very few in betweens. When I first took the test and I got 10% I was so excited and really I didn't have negative comments untill I tested my parents and my mother came up with 7% and my father received 0%. According to some people the way the test works it should keep dividing by 50%, if I my father had 0, my mother would need 20% for me to have 10%. What about my children, should they both recieve 5%, then my grandaughter 2.5%, what about her child 1.25% till the 6th generation has nothing.... So how many people have tested their parents and did they have less. Have you tested your children and futher down the line. I am trying to figure out wether the ABDNA test is wholly flawed and the majority of people are reciving invalid test resutls....Now I am also talking about test results period no matter what ethnic group...I just got tired of the negative comments and removed the results for now.... Do any of you question your results....I quess in a way I am starting to have issues with autosomal DNA becasue of the problem it cause when one of my child recieves 14% and the other child receives 0. I feel bad about it....In a way I wish I hadn't taken the test... I still have Amerindian decent no matter what the back up test say...

Maria

[juan carlos]

I think those two companies, DecodeMe and 23andMe, should devote more time to the ancestry part and less to the disease part. Or give people the option of choosing the ancestry part only, at a lower price, and if they want the whole thing, including their probabilities of getting this or that illness, then pay the higher price. Since I have hypochodriac tendencies, I do not want to know what they say I am going to catch or die from. I already know the color of my eyes and hair, as well as my height and face and head shape.

[Maria_W]

Why?

Why would 23and Me be any different? Why would it not show Asian and African? So since they don't have a Native American category, would't it just be European? Wouldn't it still be the same issue. My ancestry is too far back to show up right unless i have a more recent line which i haven't found prove of...

Maria

[Maria_W]

Expensive experiment..

DKF,
It is strange that my mother has 7%, me 10%, daughter 0%, daughter 14% and grandaughter 9%. They are all over the place arent they..Maybe this is an expensive experiment! Don't know if I want to put anymore money into testing my other family members.

Maria