Sweet Corn Did Not Contain GE Material
Monday July 15, 2002
Sweet Corn Did Not Contain GE Material
There have been a number of widely circulated emails and press releases describing my involvement in this issue which, I believe, contain serious misunderstandings. I would like to bring my response to your notice. If you wish to mention any part please do but I am primarily concerned to stop the dissemination of misinformation. I will restrict myself to two of these messages.
Dr. Robert Mann has widely circulated the following email today:
“This is the man who performed some of the assays which measured GM contamination in the Novartis seed-corn in Nov 2000. If the analogy he now offers is correct, why did he ever agree to attempt any such assay? How much money did he accept for purporting to do what he now says was impossible? Within this past few days, this same operative has also:- 1 asserted passionately that there isn't any science in N Hager's bk on the corn cover-up 2 stated that some of the seed he assayed had talcum powder on it, which he takes as proof that it has been in a mechanical seed-planter. This talcum powder, he suggests, implies contamination by bacteria from soil. He then further asserts that the DNA fragment he assayed for is rife in such soil contamination and was likely the whole explanation of the (avg. 0.04%) GM assay readings. On Friday morning Dr Poulter presented himself on Radio NZ as 'U of Otago', not mentioning that he did those assays for Crop&Food®, Conner's GM-fanatical commercial outfit. Why does his 'soil bacteria contamination' story only surface now?”
In response to Dr.Mann’s questions:
I was asked by Heinz-Wattie to assist them in the analysis of these highly technical results, this was done under a contract between the University of Otago and Heinz-Wattie. Neither I nor Heinz-Wattie have ever suggested I performed the tests done by Crop&Food Lincoln, Genescan Melbourne or the independent externally accredited USA laboratories.
On the TV1News today I was indicating that detecting values below 0.02% contamination was beyond the theoretical resolving power of the system. Less than 0.02% corresponds to less than 7 target molecules in a 200ng DNA sample.
I am still of the opinion that there is no science in Mr. Hager’s book.
As stated above I did not analyse any of the seed. The presence of the talc was described by Crop&Food, Lincoln, in a press release on Thursday 11th July. The press release was the first I knew of the observation. Crop&Food drew the reasonable conclusion that the farmer who had bought the seed and had opened it and emptied it out, mixed it with talc to facilitate drilling. The surplus seed had then been rebagged and returned. This indicates the bag is evidentially worthless and suggests a likely source of the soil contamination.
The ‘average value of 0.04% derived by Dr. Hannah (described in the released papers) was substantially due to the two samples from this open bag. If these are set aside as without significance the average value would be 0 since there are no other significant positive values. There were other technical reasons for setting this result aside. For example although there was detection of the nos terminator signal there was no detection of the 35s promotor signal (see Gordon Campbell in the latest issue of the Listener, Mr. Campbell has failed to understand this point, as well as several others). As has been explained in the media this nos result supports the suggestion of contamination with the abundant soil bacterium Agrobacterium, the natural source of the nos sequence. Soil will not usually contain any 35s sequence since this is derived from the cauliflower mosaic virus. 35s is accepted by international authorities as more easily detected than nos as noted in the IUPAC publications below (this contrasts with the anecdotal information of Dr. Wills).
Dr.Wills has produced a press release and has appeared on TVOne today. He refers to the ‘GenScan’ results described in a letter from GenScan to Novartis dated December 6th (released papers).
Apart from the open bag, the only tests that caused any concern was the GeneScan duplicates of samples 4a, 4d and 4e, samples which had been declared free of GM contamination by Crop & Food Research. Reproducibility is a key criterion of scientific proof. The positive signals obtained from the GeneScan tests (4a, 4b and 4e), as reported to ERMA, were described to us as preliminary and were too weak to have any evidential value whatsoever according to the internationally recognized standards for PCR testing for GM contamination. The report in the released papers stresses that the result is ‘not yet final and conclusive’, it apologises for equipment failures which prevented it providing the requested results, it indicates that the level of contamination suggested by the preliminary results was ‘lower than 1/5000 [lower than 0.02%]. The report reiterates ‘that until our testing is finished these figures are not yet conclusive’. No further report on these samples has ever been received by Heinz-Wattie or Cedenco (after some 19 months).
The internationally accepted standards are described in the published scientific literature. For example ‘IUPAC collaborative trial study of a method to detect genetically modified soy beans and maize’ (Lipp et al, Journal of AOAC International 1999, 82 923-928) or ‘Validation of PCR methods for quantitation of genetically modified plants in food.’ (Hubner et al J AOAC Int 2001 84 1855-64). These standards state that “samples containing 2% GM will consistently be detected for maize. Samples containing lower amounts of GM (0.1 to 0.5%) will also be detected a high percentage of the time” (Lipp et al). The value of 0.1% to 0.5% is the empirically derived practical limit, it varies depending on the exact sequence being amplified. This range of values (0.1-0.5%) owes nothing to any ‘vacillation’ on my part described in the recently released documents.
0.1% is given as the theoretical limit of quantitation corresponding to 36 target molecules in a 200ng DNA sample (see Table 2 of Hubner et al). The signal levels found in the GeneScan preliminary ‘positives’ of December 6th was reported as less than 0.02%.This signal corresponds to less than 7 target molecules in a 200ng DNA sample. This is below the accepted theoretical detection sensitivity of the method. Sporadic weak signals of this type have no evidential value.
These points were made clear by me at the December Wellington meeting. Although my analysis is described in the released papers as ‘cavalier’, I continue to hold my views unchanged.
In response to various claims I would like to make clear that I am not a member of any political party or pseudo-political grouping such as Greenpeace. I am not a member of the Life Sciences Network, I did not make any submissions to the Royal Commission. I am not in any way engaged in opposing the Green Party or Ms Jeanette Fitzsimons. I have had occasion to discuss issues relating to GE with Jeanette. She is in my opinion an intelligent and open minded woman for whom I have great respect.