Even as the Genetic Engineering Appraisal Committee (GEAC) decided to constitute a sub-committee to review the toxicology data generated by two applicants for genetically modified brinjal, biologist Dr. Pushpa M. Bhargava has questioned the guidelines of the Department of Biotechnology (DBT) on transgenic crops.
ADVERTISEMENT
Dr. Bhargava and others had asked for the raw data on toxicity studies on rats using transgenic brinjal, which were carried out by Dr. Sesikeran, former director of National Institute of Nutrition at Hyderabad.
ADVERTISEMENT
He found statistically quite significant differences between rats fed on Bt Brinjal and those fed on a normal meal in respect of several important parameters, said Dr. Bhargava.
However, Dr. Sesikeran had said that as all the values (both of the control and of the experimental animals) fell within the normal range of variation, the differences were not significant, and that there was no need to repeat the experiment.
“Our point was that if on repetition the same differences are found again, they are bound to be significant,” Dr. Bhargava pointed out. Further, he used only 20 animals (10 female and 10 male) in both experimental and the control groups, which is the minimum number for such tests. Dr. Sesikeran must explain why only a minimum number was used, he said.
In a letter to Dr. Ranjini Warrier, member-secretary, GEAC, on July 23, Dr. Bhargava, who was responding to the two emails of July 20 from Dr. Sesikaran to all the members of GEAC, said, “According to Dr. Sesikeran, DBT guidelines of 2008 say the following in regard to “Interpretation of results of safety studies”: “The design and analysis of the study should be kept as simple as possible, avoiding unnecessarily complex, sophisticated statistical techniques. If the design is simple, the statistics are likely to give straightforward results. Non-statistical knowledge must be applied in study design and proper interpretation of the biological significance of the results. Just because two treatments are statistically significantly different does not mean that the difference is large enough to have any biological importance or any practical significance.”
Dr. Bhargava said he would like to know which international body endorsed this as scientifically; it does not make any sense. In fact, every sentence in this statement is flawed. For example, what is meant by the statement, “If the design is simple, the statistics are likely to give straight forward results”. Then, what is “Non-statistical knowledge” that “must be applied in study design and proper interpretation of the biological significance of the results”.
Further, the last sentence is biased and misleading, he said.
Surely, if the same statistically significant difference between the control and the experimental group in regard to a particular parameter is observed in repeated experiments, it is bound to have biological importance irrespective of the values falling or not falling between the normal range of variation. This can be verified by referring to any independent biology-oriented statistician of stature in India or abroad, Dr. Bhargava pointed out.
He asked the GEAC to respond to some questions he has raised in the next meeting.
“Twenty is the least number of animals to be used. Does not it stand to reason that if the range of normal variation is large, say between 10 and 20 units, the number of animals that would need to be used to obtain reliable results would also be larger?
Therefore, what was the range of normal variation in the case of each parameter tested? Was just one dose used in the experimental group? If so why? And what was the dose used?” he said.
He also pointed out that at the GEAC meeting on July 18, members said, “Biology should take precedence over statistics”. He said he didn’t understand what that means and sought a clarification.
The GEAC meets next in August.