Splice sites prediction of Human genome using AdaBoost

Abstract

With the rapid growth of huge amounts of DNA sequence, gene prediction has become a challenging problem in bioinformatics. Splice sites prediction plays a key role in identification of genes. Hence, development of new methods to improve the accuracy of the splice sites prediction has great significance. This paper introduces a new method for splice sites prediction by combining AdaBoost classifier with a modified nucleotide encoding method, namely DM2. This method has been applied to the (HSD)-D-3 dataset with repeated 10-fold cross validation. Experimental results show that this method improves accuracy of the splice sites prediction and performs better than the MM1-SVM, Reduced MM1-SVM, SVM-B, LVMM2 and DM-SVM.