Prediction of Presynaptic and Postsynaptic Neurotoxins Using Hybrid Approach and Pseudo Amino Acid Composition

Neurotoxins can be divided into presynaptic neurotoxins and postsynaptic neurotoxins based on their mechanism of action. The two neurotoxins have important application in basic research and drug design. Presynaptic neurotoxins have been used for treatment of migraine headache and cerebral palsy. Therefore, the successful prediction of these two neurotoxins is becoming very important in recent years. In this article, based on the concept of Chou´s pseudo amino acid compositions, a novel ED-SVM algorithm combined increment of diversity (ED) with support vector machine (SVM) for predicting presynaptic and postsynaptic neurotoxins is proposed. The results of five fold cross validation show that the sentivity are respectively 98.72% for presynaptic neurotoxins and 79.71% for postsynaptic neurotoxins. In order to compare ID-SVM algorithm with other approaches, the ID-SVM algorithm is also used to predict neurotoxins that described in the work of Saha and Raghava, the higher predictive success rates than the previous algorithms are obtained by the five fold cross validation.