TY - JOUR T1 - Computational Prediction and Experimental Assessment of Secreted/Surface Proteins from Mycobacterium tuberculosis H37Rv A1 - Vizcaíno, Carolina A1 - Restrepo-Montoya, Daniel A1 - Rodríguez, Diana A1 - Niño, Luis F. A1 - Ocampo, Marisol A1 - Vanegas, Magnolia A1 - Reguero, María T. A1 - Martínez, Nora L. A1 - Patarroyo, Manuel E. A1 - Patarroyo, Manuel A. Y1 - 2010/06/24 N2 - Author Summary Since the publication of the Mycobacterium tuberculosis genome in 1998, great expectations have emerged regarding speeding up the process of developing vaccines against tuberculosis. Our group has been focused on identifying molecules localized on the mycobacterial surface that could act as ligands facilitating this pathogen's entry into host cells. Immune responses exerted against these proteins might block receptor-ligand interactions, thus hampering mycobacterial invasion. Since protein fragments involved in these interactions might serve as vaccine candidates and, taking into account that a relatively small number of mycobacterial surface proteins have been experimentally identified to date due to the inherent difficulty of proteomics studies for characterizing surface proteins, in this study, we used Machine Learning-based tools available on the World Wide Web to obtain accurate predictions of surface and secreted proteins from this pathogen and found experimental support of such predictions for a group of candidate proteins selected based on novel criteria. JF - PLOS Computational Biology JA - PLOS Computational Biology VL - 6 IS - 6 UR - https://doi.org/10.1371/journal.pcbi.1000824 SP - e1000824 EP - PB - Public Library of Science M3 - doi:10.1371/journal.pcbi.1000824 ER -