About

 Cell wall lytic enzymes, as an important biotechnical tool in drug development, agriculture and food industry, its related research has attracted more concern. Among of them, accurately identification of cell wall lytic enzymes is one of the key and fundamental tasks. In this study, in order to get rid of inefficiency of vitro experiments, a support vector machine-based cell wall lytic enzymes identifying model was constructed by bioinformatic way. This machine learning process includes feature extraction, feature selection, model training and optimization. According to jackknife cross validation test, this model obtained 0.853 in sensitivity, 0.977 in specificity, 0.845 in MCC, 0.915 in average accuracy and 0.915 in AUC. This benchmarking result demonstrates that the proposed model not only outperforms state-of-the-art method but also has powerful cell wall lytic enzymes identifying ability.