Knowledge-based consensus methods for secondary structure prediction of transmembrane beta-barrel proteins
Abstract
Transmembrane β-barrel proteins are an important class of proteins that span the entire membrane and form a β-barrel structure. They play key roles in human body and are targeted by modern medicine. Computational and experimental difficulties in determining their structure have limited the number of resolved structures to a tiny percentage of the Protein Data Bank. As a result, over the years a number of methods based on different assumptions and models have been proposed. In this paper, we discuss consensus methods for combining results from different sources. We show how these methods can be used to improve the prediction power significantly and thus form the most accurate method for secondary structure prediction. We also discuss interpretation of basic social-choice theory axioms and results in the context of secondary structure prediction of proteins.
