GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2004-01 > 1073181720


From: "Lucy Hubrich" <>
Subject: [DNA] DNAPrint Test Worked for Me
Date: Sat, 3 Jan 2004 20:13:13 -0600


Raymond,
If what you state is true, what is the ethnic group that is dropped off? They test for four ethnic groups and a triangle only has three sides. So, are you saying that Native American is the only ethnic group that is on everyone's results, and only the Indo/European, East Asian, and African varies? In my case, I had no African side.

It is statistically possible for a sister to have 20% and a brother to have 0%. The only thing that would not be statistically possible is for a child to exceed the sum of their parents. I'll ignore the sarcasm, and state that is why more than one sibling should be tested. Obviously, if both test out as having the same parents, and the parents (as in my case) are deceased, then I say--both should start looking (but, then, you knew that would be my response).

Lucy
From: (Raymond Whritenour)
Subject: [DNA] DNAPrint Test Worked for Me
Date: Sat, 3 Jan 2004 16:55:38 -0500 (EST)



Lucy:

Everybody--I mean EVERYBODY--who takes the DNAPrint 2.0 BGA test
receives a triangle chart which indicates "POSSIBLE" Native American
ancestry. Even those who get an MLE of 0% NA still have confidence
intervals which show possible percentages of NA ancestry. SO, why
doesn't EVERYBODY simply look for Native American ancestors? No need to
lay out money for this test.

To understand where I'm coming from you would have to know, as I do,
that my mother has NO American Indian ancestry which could show up on
this test, yet tested 10% NA, with a gray confidence interval which
never drops to 0%, To prove this to you, you would have to see all the
documentation, and all the family photographs that I've researched for
over twenty years. But, I know the truth. And, knowing the truth, I
know that this 10% NA result is bogus. And that means, to me, that this
test is inaccurate. There's something very wrong with it. It can't be
just lab error or statistical error. There has to be great systematic
flaws, in order for such a result to appear.

I can't say what percentage might REALLY indicate NA ancestry. I would
guess that if your yellow confidence interval remains entirely above
0%--no matter what your MLE is--you likely have NA ancestry.

And, keep in mind, that we have seen at least one case, here, where a
sister has over 20% NA, while her full brother had 0% NA! Which one
should NOT have looked for NA ancestors?

Ray Whritenour

This thread: