DNA encoding for splice site prediction in large DNA sequence

Abstract

Splice site prediction in the pre-mRNA is a very important task for understanding gene structure and its function. To predict splice sites, SVM (support vector machine) based classification technique is frequently used because of its classification accuracy. High classification accuracy of SVM largely depends on DNA encoding method for feature extraction of DNA sequences. However, existing encoding approaches do not reveal the characteristics of DNA sequence very well enough to provide as much information as DNA sequences have. In this paper, we propose new effective DNA encoding method which can give more information of DNA sequence. Our encoding method can provide density information of each nucleotide along with positional information and chemical property. Extensive performance study shows that our method can provide better performance than existing encoding methods based on several performance criteria such as classification accuracy, sensitivity, specificity and area under receiver operating characteristics curve (ROC). © Springer-Verlag 2013.