Classifying Genes Encode Transmembrane Proteins Using Neighboring Genes

Abstract

    Like bio-diversity, there are too many proteins to experimentally annotate. Thus, methods for predicting the functions of proteins become necessary. In this article, we proposed a method that uses biological data to classify membrane transporters according to transported substrates. Motivated by the concept of Operons, our method used expression data and GO terms of neighboring genes to create input data for support vector machine. To rapidly obtain the result, we integrated LIBSVM in our tool then used this tool to train and test our classifiers. With this tool, users can classify membrane transporters and other kinds of proteins. This tool also allows users to add their desired organisms beside our tested ones.