Operon Prediction by Decision Tree Classifier Based on VPRSM

The prediction of operons is critical to reconstruction of regulatory networks at the whole genome level. In this paper, a novel approach based on variable precision rough set model (VPRSM) is presented to prediction of operon. We use five effective features: max distance, min distance, gene strand direction information, scores of COG, and scores of gene order conservation. The proposed method is examined on Escherichia coli K12 and an accuracy of 89.4% is obtained. We also compare this method with C4.5 and BP. The results indicate that VPRSM based decision tree classifier is an effective classifier for predicting operon.