GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2008-02 > 1201887303


From:
Subject: Re: [DNA] DNA Testing Times
Date: Fri, 1 Feb 2008 12:35:03 EST



In a message dated 2/1/2008 10:53:37 A.M. Central Standard Time,
writes:

Giving your dna to the SMGF research project is not the same as paying for a

product from a commercial testing company. You pay no fee; you are told
your dna may or may not be used in the research project. The project
provides a public database from which you can seek your outcome if they
include your dna. The research project has a backlog of thousands of dna
samples from all over the world to process; the research project's
priorities may not match yours.

Not always, but most of the time I have enjoyed faster turn around time on
STR haplotype measurements from the commercial companies who have employed
the Sorenson labs for their measurements.




You're right of course, and I said I suspect I know why their data has not
been completed. Odd though, because it was a 4 generation submission and I
thought they were interested in autosomal and mutation rate research. Still, I
am NOT imputing that their labs are of lower quality, nor would I. Nor do I
disagree that turnaround time is faster from Sorenson. NOT IN DISPUTE!

I maintain only that turnaround is not related to data
quality/accuracy/reliability, but is a function of the resources in terms of manpower and
equipment capacity and the workload in the lab. A staff of x with y pieces of
equipment will turn around test kits, even with double testing, faster than a staff
of x/n with y/q pieces of equipment offering a wider range of products. I
have not toured Sorenson and seen the number of people working there nor how
many pieces of equipment their lab has. Nor have I been to the AZ facility of
FTDNA, nor to EA or OA. However, I have been told the Sorensen lab is much
larger with more equipment and has a staff much larger than the combined FTDNA
AZ and Houston facilities (and I have toured the Houston facility). If my
impression of the resources is incorrect, I will happily adjust my thinking on
that.

I also maintain that the ability to purchase products not offered by
Sorenson is a major factor in lab selection -- as yourself must agree as you have
purchased products from labs other than Sorenson. I rather suspect that the
wider the range of products offered the longer the turnaround time too, as some
processes take longer on the equipment (or so I understand).

My training leads me to suggest that if 60 people process 5,000 kits a month
with 99.9 accuracy after double testing runs for a limited product range
while 20 people process 6,000 kits a month (single test run) with a wider
product range with 99.8% accuracy, the reliability is the SAME at both labs. If
the 20 people process 10,000 kits a month with the wider product range and
99.8% accuracy then they are slightly more accurate and and much more efficient.
We do not have the data on which to evaluate this, of course, as the
companies are proprietary -- this is an illustrative hypothetical example.

If and when I see objective evidence that the error rate adjusted for staff
size is higher at one lab or another -- which frankly I doubt we ever will --
then I will entertain discussions of the relative "lab quality." Until such
time, I think it is intellectually dishonest to make statements which imply
that turnaround time is related to data reliability and that Sorenson (or any
other company) is superior in that area. I resent the confusion of personal
opinion and preferences with scientific evidence of higher data quality.
That's all.

One last time for the benefit who do not yet understand my position. I
think ALL the labs produce reliable data. I think how fast they do it is a
function of staff size, number of pieces of equipment and the workload. I don't
think any of us are qualified to make those judgements based on what we know
and we should not make assertions as IF IT WERE PROVEN FACT.



**************Biggest Grammy Award surprises of all time on AOL Music.
(http://music.aol.com/grammys/pictures/never-won-a-grammy?NCID=aolcmp003000000025
48)

This thread: