Cascading classifier application for topology prediction of TMB proteins

This paper is concerned with the use of a cascading classifier for trans-membrane beta-barrel topology prediction analysis. Most of novel drug design requires the use of membrane proteins. Trans-membrane proteins have key roles such as active transport across the membrane and signal transduction among other functions. Given their key roles, understanding their structures mechanisms and regulation at the level of molecules with the use of computational modeling is essential. In the field of bioinformatics, many years have been spent on the trans-membrane protein structure prediction focusing on the alpha-helix membrane proteins. Technological developments have been increasingly utilized in order to understand in more details membrane protein function and structure. Various methodologies have been developed for the prediction of TMB (trans-membrane beta-barrel) proteins topology however the use of cascading classifier has not been fully explored. This research presents a novel approach for TMB topology prediction. The MATLAB computer simulation results show that the proposed methodology predicts trans-membrane topologies with high accuracy for randomly selected proteins.