Identification and In silico analysis of proline-glutamate/proline-proline-glutamate proteins of Mycobacterium tuberculosis complex: A comparison of computational web-based tools

Abstract

Background: Understanding the protein's subcellular localization and secretory nature can greatly improve the target identification for diagnostic assays and drug discovery, although their identification in laboratory experiments is a time-consuming and labor-intensive process. In order to identify proteins that could be targeted for therapeutic intervention or the development of diagnostic assays, we used a variety of computational tools to predict the subcellular localization or secretory nature of mycobacterial proline-glutamate/proline-proline-glutamate (PE/PPE) proteins. Methods: PSORTb version 3.0.3, TBpred, and Gpos-mPLoc analyses were performed on 30 selected PE/PPE protein sequences, while, SignalP 6.0, SignalP 5.0, Phobius, PSORTb version 3.0.3 and TBpred were used for signal sequence predictions. Results: Gpos-mPLoc and TBpred had the highest concordance for extracellular prediction, while PSORTb and TBpred had the highest concordance for prediction of membrane localization. The tools for predicting the secretory nature of proteins had little agreement. Conclusion: Multiple computational tools must be considered to provide an indication of the subcellular localization of PE/PPE proteins. Laboratory experiments should be used to confirm the findings of the tools.

Keywords

• proline-glutamate/proline-proline-glutamate proteins
• secretory proteins
• subcellular localization
• web-based computational tools