Archiver > GENEALOGY-DNA > 2008-06 > 1213548889

From: Robert Stafford <>
Subject: Re: [DNA] How could we tell?
Date: Sun, 15 Jun 2008 09:54:49 -0700 (PDT)
In-Reply-To: <>

Not matching a pedigree doesn't prove a mistake. Afterall, that is why we are testing. You need to have them retested elsewhere to be sure.

Bob Stafford

Bill Howard <> wrote:
Just a short note about my experience at SMGF - I sent in 10 samples
in August 2005 and the results were not posted for more than a year.
Some, but not all could later be identified on their web site.
However, they made bad mistakes on two of the samples, due, I think,
to sample switches, and I made many phone calls to people named
Angela, Anna, and Karen. The pedigree did not match the relationships
indicated by the test. My phone calls were taken with courtesy and I
was told to wait. I waited and waited and waited. More phone
calls...... nothing done although I told them exactly what was wrong
and they all promised to look into it. Yes, I could have paid money to
get the results, but if the samples had been switched, it would not
have done any good.
That's a 20% error rate in my book.
- Bye from Bill Howard

On Jun 14, 2008, at 1:10 PM, Robert Stafford wrote:

> It is hard to tell what the lab error rate is, since most errors are
> not reported publically. However, I suspect there are a lot more
> than people think. I have seen about 20 posts about errors here and
> on Genforum. I have also private reports from people about errors.
> One big problem seems to be clerical errors. I am not clear where
> they occur, unless the firms post results to their web sites manually.
> There are probably many actual lab errors on single markers that
> have not been discovered. They would probably not arouse suspicion
> and would be discovered only if the person retested. DYS464 seems to
> be a big problem, because the relative peak heights are used. I
> think it is a good idea to retest at, if there is a
> mutation from the ancestral haplotype. It is a worthwhile project
> anyway.
> One of the biggest problems, according to a person who works at a
> lab, is sample switches, often on just one panel. I have seen
> several cases posted. For people working within a documented
> genealogy, they immediate arouse suspicion. However, a loner might
> get bad results and never know.
> Bob Stafford

"History is the consensus of survivors in authority"
"Reality is the weighted mean of individual perceptions"

To unsubscribe from the list, please send an email to with the word 'unsubscribe' without the quotes in the subject and the body of the message

This thread: