Computational Prediction and Experimental Assessment of Secreted/Surface Proteins from Mycobacterium tuberculosis H37Rv

• DOI: 10.1371/journal.pcbi.1000824
• Publication Date: 2010-06-24T20:18:58Z
• Authors: Carolina Vizcaíno; Daniel Restrepo-M...; Diana Rodríguez; Luis F. Niño; Marisol Ocampo; Magnolia Vanegas; María T. Reguero; Nora L. Martínez; Manuel E. Patarroyo; Manuel A. Patarroyo
• Keywords: Secretory protein; Cell fractionation

The mycobacterial cell envelope has been implicated in the pathogenicity of tuberculosis and therefore a prime target for identification characterization surface proteins with potential application drug vaccine development. In this study, genome Mycobacterium H37Rv was screened using Machine Learning tools that included feature-based predictors, general localizers transmembrane topology predictors to identify are potentially secreted M. tuberculosis, or extracellular milieu through different secretory pathways. subcellular localization set 8 hypothetically secreted/surface candidate experimentally assessed by cellular fractionation immunoelectron microscopy (IEM) determine reliability computational methodology proposed here, 4 experimental confirmation as positive controls 2 cytoplasmic negative controls. Subcellular IEM studies provided evidence Rv0403c, Rv3630, Rv1022, Rv0835, Rv0361 Rv0178 either milieu. Surface also confirmed controls, whereas were located on cytoplasm. Based statistical learning methods, we obtained predictions allowed us construct protocol support new candidates.