Sunday, May 19, 2013

Putting My Money Where Your Mouth Is!

The Stunning Corn Comparison promoted by Zen Honeycutt at Moms Across America was certainly stunning.  While Zen, farmer Howard Vlieger and Profit Pro stand by these data as authentic, a codified scientific community sees them as either poor quality, a mistake, (or at worst) fraud.

If the results are real and GMO corn is stripped of almost all carbon, loaded with (carbon-based) formaldeyde and glyphosate, plus a Brix of 1%, it would be a remarkable story that would shake the foundation of modern production agriculture. Such findings, if found transparently and in independent, replicated trials, would likely grace the covers of major science weeklys like Science or Nature.

It would be huge news, and as a scientist, I am thrilled to test the hypothesis that the previous data are authentic.


So excited!  I love to see experimental data replicated! 
More numbers, locations, etc, the better I sleep at night!

I have agreed to personally finance the analytical portion of a replicate of the experiment. Experimental design is pending based on the corn genotype, culture conditions and analytical tests performed to obtain previous data. I am waiting to hear back from those that obtained the last numbers.

It is expected that those that produced the original data participate, and I'm working on that.  I think they'd be glad to see their stunning results replicated in statistically meaningful independent trials. Since they stand by the data so rabidly, it is expected that they participate.

At this point, here are the conditions (please submit suggestions in Comments)

Experiment:
  • The experimental design will contain corn from at least two locations.
  • Each location will contain the "GMO" and "non-GMO" lines (hybrids?) used to obtain the original numbers. 
  • Each measurement will come from three replicates of each genotype. 
  • Statistical methods will be determined up front and be adhered to.
  • At least two analytical facilities will be used. 
  • All samples will be coded in the field by independent personnel (County Agents, grower cooperators) and coding will not be revealed until final data are complete.
  • Exact analysis will be based on previous data reported, including %Brix, carbohydrates, amino acids, minerals, organic compounds (formaldeyde, glyphosate) and any other tests suggested.  I'm not sure how to test "anerobic biology", so I'm waiting for information from Dr. Mae-Wan Ho and Howard Vlieger to tell me how to do this or where we get it done.  Or what it is.

Dissemination:
  • The data will be reported in a peer-reviewed journal consistent with impact of results. Publication will be pursued regardless of results obtained. 
  • All participants will be co-authors.  Offers have been extended to Dr. Ho.  I would also like to include Huber, Vlieger and Zen Honeycutt, as they stand by these data and should be included in their publication.
  • Results will be prominently featured, potentially after publication, on websites and blogs, regardless of outcome.  An announcement will be made to obtain agreement to report results on various websites. This will be a written agreement and publicly accessible.  It is expected that Moms Across America will stand by their rabid defense of the original data and agree to post these findings. 
A proposal will be circulated early this week and it will be adjustable based on public input and the scientific team that vetted/promoted the original data.   We want an airtight, valid, mutually-agreeable experimental design here so that we can have the biggest impact with the results! 

Here goes!  I'm so excited to see this get off the ground.  I'm grateful for Dr. Ho's agreement to participate and also for all of the experimental advice from Anastasia Bodnar and Karl Haro von Mogel. 



4 comments:

WyoWeeds said...

I would suggest increasing to at least 4 replicates (variety trials are typically 5 or 6 reps). For sampling, I suggest pulling all ears from 10 consecutive plants from each plot at maturity. Shell the corn from all 10 ears, and use a bulk sample for analysis.

Bill Price said...

I'd concur with Wyoweeds, assuming there is enough material. Get some power behind the tests, especially since they are potentially multivariate. Also, your protocol above should read something like "measurements will be made separately on each replicate".

Mary M said...

Heh. So glad to see the stats guys weigh in.

We don't want this to happen: http://blog.openhelix.eu/?p=5355

Kevin M. Folta said...

Thanks! Yes, material should not be an issue. We need to have excess material from each rep for DNA fingerprinting and other valdiations too.

PLEASE Wyo and Bill, check back and help shape the final proposal. Really important.