LAUSR

Abstract Piscirickettsia salmonis and Caligus rogercresseyi are highly transmissible pathogens that cause high mortality in farmed salmonids. Detection of them, in most cases are well characterized for being time consuming and expensive. In this way, new techniques based on mass spectrometry and machine learning were applied and combined in an automatized platform in order to classify and predict these pathogens, in a faster and effective way. MALDI-MS was used to analyze serum samples from salmonid fishes (healthy and diseased) coupled to support vector machines analysis in order to obtain a specific and sensitive pattern ( m/z ) for every pathogen with high reproducibility. The results probed that combining these two techniques a powerful tool in the correct detection of these pathogens is obtained. Accuracy, sensitivity and specificity were equal to or > 92%, implying the good performance of our platform as a potential diagnostic tool in the salmon farming industry.