Successful prediction of the potential allergenicity of a protein may be a key factor in the development of novel, genetically modified foods. The use of the decision tree approach for the prediction of allergenicity is discussed. The methods currently used for identifying allergenic proteins (including use of IgE from patient sera for recognition of proteins) are reviewed. Finally, a specific review of the literature concerning identification of allergens from sesame leads to the conclusion that in the absence of validated animal models, identification of allergenicity (and, consequently, prediction of allergenicity) may be problematic.