A new approach for species discrimination of different processed animal proteins based on fat characteristics

A new method for discriminating processed animal proteins (PAP) from different species based on their fat characteristics was established for the first time in this study. A total of 53 reliable PAP samples containing fish meal (FM), porcine, poultry, bovine, and ovine meat and bone meal (MBM) were collected, and 37 distinct fatty acids were statistically analyzed via gas chromatography. Principal component analysis and the partial least squares discriminant analysis algorithm were performed on the dataset. The results revealed that PAPs from different species could be clearly classified based on the presence or absence of fatty acid compounds with high sensitivity and specificity (>0.95), except for the discrimination of bovine MBM in the mixed dataset. It was possible to distinguish FM samples in the mixed dataset, ruminant MBM from non‐ruminant MBM and FM, and porcine MBM from poultry MBM samples based on the concentrations of 34 fatty acid compounds. Further discrimination of PAP samples from different species using fewer selected fatty acids was comparable to or better than discrimination based on the presence or absence and concentration of 34 fatty acid compounds. Practical applications: A new method for discriminating between processed animal proteins (PAP) from different species based on their fat characteristics was established for the first time in this study. Results suggested that the analytical methodology established in this study has great potential for species‐specific identification of PAP feed with respect to the global “mad cow” crisis. PCA visualisation based on the presence or absence of fatty acid, MBM, meat and bone meal, FM, fish meal. PCA visualisation based on the presence or absence of fatty acid, MBM, meat and bone meal, FM, fish meal.