The animal neurotoxin proteins are one of the most common functional groups among the animal toxin proteins. Because
of the high affinity of these animal neurotoxin proteins in some special target site, they were usually used as pharmacological tools
and therapeutic agents in medicine to gain deep insights into the function of the nervous system. Thus, it was very important to
characterize and predict the animal neurotoxin proteins. In this study, the differences between the animal neurotoxin proteins and nontoxin proteins were analyzed. Significant differences were found between them. In addition, the support vector machine was proposed
to predict the animal neurotoxin proteins. The predictive results of our classifier achieved the overall accuracy of 96.46%.
Furthermore, the random forest and k-nearest neighbors were applied to predict the animal neurotoxin proteins. The compared results
indicated that the predictive performances of our classifier were better than other two algorithms.
Keywords: neurotoxin protein, reduced amino acid alphabet, biological property, support vector machine.
Rights & PermissionsPrintExport