Models for the Prediction of Antimicrobial Peptides Activity
Antimicrobial peptides AMP are small proteins produced by the innate immune system in multicellular microorganisms. The mechanism of action of AMP on target membranes can be divided in two main categories: pore forming and non-pore forming mechanisms. We applied a computational approach to design novel linear peptides having high specificity and low toxicity against common pathogens. We built up QSAR models using the data present in a database of antimicrobial peptides. Here, we present new models of activities obtained by the use of evolutionary methods and the relative statistical validation.
KeywordsGenetic Algorithm Artificial Neural Network Antimicrobial Peptide Artificial Neural Network Model Methicillin Resistant Staphylococcus Aureus
- 9.Hancock, R.E., Chapple, D.S.: Peptide antibiotics. Antimicrob. Agents Chemother. 43(6), 1317–1323 (1999)Google Scholar
- 11.Holland, J.H.: Adaptation in Natural and Artificial Systems: an Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence. MIT Press, Cambridge (1992)Google Scholar
- 13.Accelrys, Accelrys Materials Studio. Accelrys Inc., San Diego, California (2014)Google Scholar
- 14.MATLAB, R.: Version 8.1. 0.604 (R2013a). The MathWorks Inc., Natrick, Massachusetts (2013)Google Scholar
- 15.Chen, L., et al.: How the antimicrobial peptides kill bacteria: computational physics insights. Commun. Comput. Phys. 11(3), 709 (2012)Google Scholar