iTerm-PseKNC is a webserver for the identification of bacterial transcriptional terminators based on machine learning method. In the predictor, 5-tuple nucletide frequency and physicochemical property were extracted to formulate samples. The binomial distribution technique was proposed to rank 1024 5-tuple nucleotides. Then the incremental feature selection (IFS) was used to determine the optimal features which could produce the maximum accuracy. The support vector machine (SVM) was utilized to perform prediction. Five-fold cross-validated results showed that 86.78% terminators and 99.99% non-terminators can be correctly recognized, respectively, suggesting that our proposed model is very powerful. This study provides a new strategy to identify terminators. We hope the webserver could provide convenience for Biochemistry scholars.
