GMSAFOOD coordinator talks to the EFSA GMO Risk Assessment Network
May 3, 4 2012, Parma, Italy
Michelle Epstein presented a lecture entitled Using mouse models and machine learning as tools for GMO safety for the GMO risk assessment network in Parma, Italy at EFSA on May 4th, 2012.
She began by summarizing the GMSAFOOD project objectives and concepts.The rest of her talk focused on two issues, 1. Work done with mouse models and the alpha amylase peas generated by TJ Higgins at CSIRO in Australia and 2. A Machine learning strategy for biomarker search for animals and humans ingesting GMOs.
The important take home messages were:
The GMSAFOOD consortium group from the Medical University of Vienna were unable to reproduce the results from the Prescott paper (Transgenic Expression of Bean r-Amylase Inhibitor in Peas Results in Altered Structure and Immunogenicity Agric. Food Chem. 2005, 53, 9023-9030)
- They found that there is a cross reactive protein in peas with the amylase inhibitor
- Machine learning is a novel approach to search for useful biomarkers of GMO
- Machine learning can be used for the analysis of mulitple disparate datasets between species
- A public repository with datasets from the pre- and post market periods would be useful for assessing the health risks associated with GMO ingestion