In silico Peptide Analysis

Two different antimicrobial peptide predictor tools were used; the Antimicrobial Peptide Database with APD3 algorithm: Antimicrobial Peptide Calculator and Predictor http://aps.unmc.edu/AP/ (Wang et al., 2016) and the AMP predictor tool of the Collection of Anti-Microbial Peptides (CAMP) (Thomas et al., 2009). The latter operates with four different prediction models taking into account the sequence composition, physico-chemical properties, and structural characteristics of amino acids; Support Vector Machine (SVM), Random Forest (RF), Artificial Neural Network (ANN), and Discriminant Analysis (DA) (Waghu et al., 2014). As a result, SVM, RF, and DA models give a probability score (between 0 and 1) (Waghu et al., 2016). Higher score means greater possibility for the peptide to exert antimicrobial activity. AMP: the sequence predicted to be antimicrobial. NAMP: the sequence predicted to be not antimicrobial.