GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2003-09 > 1062513210


From:
Subject: From Tony Frudakis RE: [DNA] Cecelia's Mother's DNAPrint 2.0 Results
Date: Tue, 2 Sep 2003 10:33:30 EDT


Tony Frudakis sent a message to Cecilia with a copy to GENEALOGY-DNA, but
he's not currently a subscriber so it didn't show up. I will paste his message
below, and add some comments about this thread in a separate message later.


In a message dated 09/02/03 7:09:56 AM Pacific Daylight Time,
writes:

> Subj: {not a subscriber} RE: [DNA] Cecelia's Mother's DNAPrint 2.0 Results
> - a big surprise
> Date: 09/02/03 7:09:56 AM Pacific Daylight Time
> From: <A HREF="mailto:"></A>
> To: <A HREF="mailto:"></A>, <A HREF="mailto:"></A>
> Sent from the Internet
>
>
>
> Cecelia,
>
> The results we obtain for a person are remarkably similar from test to test
> and across various algorithm types. Mark Shrivers teen level percent
> African is still teen level African whether a 30 marker test is used, a
> completely non-overlapping 71 marker test or our new 184 marker super test.
> Same with my Native American. With the 30 marker test and the 71 marker
> test I am 8% even though these are almost entirely overlapping (maybe a few
> markers in common only). With the new 184 marker super test I register with
> 6% Native American, the drop most likely due to the greater chromosomal
> coverage afforded by the new, more expensive and more sensitive test.
> Comparing a 30 marker test result with a 71 marker test for another person
> is just as reliable (or unreliable, depending on how you look at it) as
> comparing 71 marker test results for two separate people - both the 30 and
> 71 marker tests involve a few percentage points of error on average if the
> MLE is used as an absolute result (which it really shouldnt be - the 2 fold
> space is a better answer).
>
> When using 3-way versus 4-way algorithms, most of our customers show the
> same exact result, and virtually all of the differences we have seen have
> been in the percentage of African. Since there is no African for your
> profile, I would venture to guess that your results would be exactly the
> same with the 4-way algorithm. The software modification will soon be moot
> however with the 2.5 version of Ancestrybydna.
>
> The change in algorithm was made because in certain highly admixed
> individuals (such as 33% Euro, 33% EAS, 10% NAM, 23% AFR), the African was
> being under reported as a result of the fact that for the African-Euro,
> African-NAM and AFR-EAS pairs, a large number of completely private alleles
> exist and the genetic distance is the greatest, so the cumulative delta
> value of markers we use is weighted more strongly for these distinctions
> which causes the algorithm we were using problems in highly admixed
> individuals with African ancestry. That is the only reason we modified the
> software, and it seems to have fixed the problem. By the way, the single
> greatest discrepancy we still find is negative NAM ancestry in people who
> believe one of their great grandparents was Amerind (usually they claim full
> blood). We are finding lower percentages in people who look and can prove
> Amerind heritage than expected. All of this leads me to believe that
> Amerinds were admixed with European blood much farther back than most people
> realize so that a 100% Cherokee is actually someone who was 50% European and
> 50% Native American (causing the dilution over generations to be more severe
> than most people think). Rosenbergs paper supports this - showing very
> heavy European Admixture for traditionally isolated Maya, Pima and Colombian
> Native Americans, and there has been speculation that Vikings may have
> contributed to the Amerind gene pool long before Colombian times.
>
> As for how we are testing, right now we are relying on mathematical
> simulations rather than large international collection efforts, because the
> former is much tighter and easier to control variables for (which is crucial
> for a sound study). For example, we can make genotypes for a fictitious set
> of samples from a population based on probability theory and the allele
> frequencies, and determine their proportions to study the incidence of error
> and bias much better than typing people from, say Kenya, which may be an
> eclectic community of high and various admixture levels...
>
>
> Tony.


This thread: