Identification of mycobacterial membrane proteins and their types using over-represented tripeptide compositions.

Mycobacterium can cause many serious diseases, such as tuberculosis and leprosy. Its membrane proteins play a critical role for multidrug-resistance and its tenacious survival ability. Knowing the types of membrane proteins will provide novel insights into understanding their functions and facilitate drug target discovery. In this study, a novel method was developed for predicting mycobacterial membrane protein and their types by using over-represented tripeptides. A total of 295 non-membrane proteins and 274 membrane proteins were collected to evaluate the performance of proposed method. The results of jackknife cross-validation test show that our method achieves an overall accuracy of 93.0% in discriminating between mycobacterial membrane proteins and mycobacterial non-membrane proteins and an overall accuracy of 93.1% in classifying mycobacterial membrane protein types. By comparing with other methods, the proposed method showed excellent predictive performance. Based on the proposed method, we built a predictor, called MycoMemSVM, which is freely available at http://lin.uestc.edu.cn/server/MycoMemSVM. It is anticipated that MycoMemSVM will become a useful tool for the annotation of mycobacterial membrane proteins and the development of anti-mycobacterium drug design.