GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2005-03 > 1110907560


From: (Raymond Whritenour)
Subject: Accuracy of the DNAPrint BGA test (was Kim TallBear, et al.)
Date: Tue, 15 Mar 2005 12:26:00 -0500
In-Reply-To: "Malcolm Dodd" <maldodd@sapo.pt>'s message of Tue, 15 Mar 2005 11:14:06 -0000


Malcolm:

By coming up with your own formula for interpreting the results, what
you're saying is that DNAPrint genomics, Inc. continues to misrepresent,
to its customers, what their BGA test is capable of doing.

What's more, not even what you say it is capable of doing is born out by
the company's own statements. In the early User Manual, which I have,
it was stated as a fact that Tony Frudakis had one full-blood Cherokee
great-grandmother. That would put him into a double digit "NA"
percentile, according to your remarks. But, he now only tests at 7%
"NA." And, given the possibility of statistical noise, plus his Greek
heritage, where the company claims about a constant 5% "NA" reading, on
average, just what are we supposed to believe about the accuracy of this
test? Or, what are we to believe about the probability of Dr. Frudakis
having a full-blood Cherokee great-grandmother?

There are so many contradictory statements made about the accuracy of
this test, over the past few years, that it is impossible to know what
the truth is.

Ray Whritenour


This thread: