We have reviewed both this article and an earlier one by Leonard Foster
that argue that Army miscalculated or misinterpreted the proteomics data. We
do not agree with either groups of authors, and we continue to have
confidence in our results.
We have been working with the U.S. Army on a point by point rebuttal to
both articles. We had just about finished our response to the first, when
the second appeared.
To properly debate the issues takes time. The Army just sent us their
part of the rebuttal - it always takes time for them to get clearance for
publication. We do intend to publish our full, joint response.
We're glad to debate this based on the science - that, after all, is the
process. We do wish that the authors had done a bit more due diligence, it
would have saved all of us extra work.
As per both of these papers, Randy is absolutely correct. Their arguments
do not" take into account the simple fact that if the Army team was
misreading Apis mellifera peptides, then they should have misread them in all
samples, which didn't occur." Keep in mind, all of our samples were bee
samples. They all had lots of bee peptides. Army analysis found bee peptides -
lots of them.
Foster speculates that Army miscalculated, confused bee peptides with
pathogen peptides. He ignored the caveat that we made when we provided him
with a list of peptides. We clearly stated that we were only providing a
list of peptides considered in our statistical analysis for the paper, NOT a
list of all (bee, plant, plant pathogen, other) peptides that we found.
The Protein database paper (which I will refer to as the Chalkley paper),
on the other hand, says they looked at the data output from three samples
that they said were representative of the samples from our study. That is
definitely not the case.
Army provided data, for three samples, as blind samples. Two of these
were blanks - bees from healthy/strong colonies with little or no detectable
IIV or Nosema. So, for two of the three samples, they are correct - no
Nosema, no IIV. In essence, they CONFIRMED our results in that they agree
with us TWO out of THREE times.
The paper then can only claim that they have found a different result
ONCE. So, as one of my co-authors, who understands statistics better than I
puts this:
"If any of us reviewed a paper that duplicated results in two cases, and
failed to duplicate results in one case, and then claimed that this means
the entire original data analysis was incorrect, I think your response would
be the same as mine: "you need to look at more cases before you can draw
that conclusion".
We looked at 100 samples. We reported at 95% confidence level, which means
we are stating clearly that in 5 instances (or less) we believe our
results will be incorrect. Put otherwise, when you only look at 3 samples
Chalkley could be completely correct, and still completely disagree with our
results. Given Chalkley agrees with us in 2 out of the 3 cases, based on the
current level of work by Chalkley, he is not able to say we are wrong. All he
should be saying is "I did different protemics work, and agree 2 out 3
times with the prior results and disagreed once. I think this means I should
study more samples."
So, at this time, the Chalkley paper should not be able to claim anything
about our results of finding a nosema / IIV interaction being correlated
with CCD. Basically, he and his team based their conclusions on a Sample of
ONE.
I want to be clear that none of this is a comment on any one's protemics
work. I feel that there is simply too small a sample to be able to do that
at this time."
We've lots more to say in our rebuttal, and our rebuttal addresses specific
points and methods and why we disagree. Here, I've simply presented a
brief overview of our perspective. The point by point debate will appear in
formal rebuttal that we intend to submit for publication.
Jerry
***********************************************
The BEE-L mailing list is powered by L-Soft's renowned
LISTSERV(R) list management software. For more information, go to:
http://www.lsoft.com/LISTSERV-powered.html
Guidelines for posting to BEE-L can be found at:
http://honeybeeworld.com/bee-l/guidelines.htm
|