Structural signature is a set of characteristics that unequivocally identifies protein folding and the nature of interactions with other proteins or binding compounds. We investigate the use of the geometric linearity of the main chain as a key feature for structural classification. Using polypeptide main chain atoms as structural signature, we showed that this signature is better to preciselly classify than using Cα only. Our results are equivalent in precision to a structural signature built including artificial points between Cα s and hence we believe this improvement in classification precision occurs due to the strengthening of geometric linearity.
CITATION STYLE
Gadelha Campelo, J. A. F., Rodrigues Monteiro, C., da Silveira, C. H., de Azevedo Silveira, S., & Cardoso de Melo-Minardi, R. (2019). Protein Structural Signatures Revisited: Geometric Linearity of Main Chains are More Relevant to Classification Performance than Packing of Residues. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 11465 LNBI, pp. 391–402). Springer Verlag. https://doi.org/10.1007/978-3-030-17938-0_35
Mendeley helps you to discover research relevant for your work.