CS-AMPPred: an updated SVM model for antimicrobial activity prediction in cysteine-stabilized peptides

Abstract

The antimicrobial peptides (AMP) have been proposed as an alternative to control resistant pathogens. However, due to multifunctional properties of several AMP classes, until now there has been no way to perform efficient AMP identification, except through in vitro and in vivo tests. Nevertheless, an indication of activity can be provided by prediction methods. In order to contribute to the AMP prediction field, the CS-AMPPred (Cysteine-Stabilized Antimicrobial Peptides Predictor) is presented here, consisting of an updated version of the Support Vector Machine (SVM) model for antimicrobial activity prediction in cysteine-stabilized peptides. The CS-AMPPred is based on five sequence descriptors: indexes of (i) α-helix and (ii) loop formation; and averages of (iii) net charge, (iv) hydrophobicity and (v) flexibility. CS-AMPPred was based on 310 cysteine-stabilized AMPs and 310 sequences extracted from PDB. The polynomial kernel achieves the best accuracy on 5-fold cross validation (85.81%), while the radial and linear kernels achieve 84.19%. Testing in a blind data set, the polynomial and radial kernels achieve an accuracy of 90.00%, while the linear model achieves 89.33%. The three models reach higher accuracies than previously described methods. A standalone version of CS-AMPPred is available for download at <http://sourceforge.net/projects/csamppred/> and runs on any Linux machine.

Figure 1. Principal component analysis of sequence descriptors for cysteine-stabilized peptides.

The components are indicated by arrows: as larger the arrow is, major is the component contribution to the set’s variance. (A) The disposition of the nine sequence descriptors in the peptide space; (B) the final ensemble of descriptors, the descriptors hydrophobic moment, index of β-sheet formation, rate between charged and hydrophobic residues and α-helix propensity were ruled out.

Figure 2. Distribution of sequence descriptor values.

The left box in each panel corresponds to the AMPs. All descriptors have statistical differences when compared to the non-antimicrobial data set, with a critical value of 0.05. The observed p-values are as follows: charge (<2.2e-16), hydrophobicity (2.169e-06), flexibility (<2.2e-16), index of α-helix formation (<2.2e-16) and index of loop formation (2.908e-10).

Figure 3. ROC curves for the CS-AMPPred models against the blind data set (BS1).