In silico prediction of anticancer peptides by TRAINER tool

Abstract

A B S T R A C T Cancer is one of the causes of death in the world. Several treatment methods exist against cancer cells such as radiotherapy and chemotherapy. Since traditional methods have side effects on normal cells and are expensive, identification and developing a new method to cancer therapy is very important. Antimicrobial peptides, present in a wide variety of organisms, such as plants, amphibians and mammals, are newly discovered agents. These peptides have various structures, sizes and molecular compositions; hence developing a computational method to predict these anticancer peptides is useful. In the present study, first, 2 databases with 138 and 206 anticancer and non-anticancer peptides were introduced, classified by TRAINER. TRAINER (http://www.baskent.edu.tr/~hogul/ TRAINER/) is a new online tool designed for classification of any alphabet of sequences. TRAINER allows users to select from among several feature representation schemes and supervised machine learning methods with relevant parameters. In this study, Naive Bayes and radial basis were used in a support vector machine. The accuracy and specificity in combination of features by Naive Bayes were 83% and by radial basis 87% and 92% respectively. The results demonstrate that two methods are useful for classification of these peptides; however, the accuracy of Radial Basis is higher than Naive Bayes.