A new method bridging graph theory and residue co-evolutionary networks for specificity determinant positions detection

• DOI: 10.1093/bioinformatics/bty846
• Publication Date: 2018-10-04T20:36:49Z
• Authors: Néli José da Fonseca; Marcelo Querino L...; Lucas Carrijo de ...; Lucas Bleicher
• Keywords: Protein superfamily; Multiple sequence alignment

Computational studies of molecular evolution are usually performed from a multiple alignment homologous sequences, on which sequences resulting common ancestor aligned so that equivalent residues placed in the same position. Residues frequency patterns full or subset its can be highly useful for suggesting positions under selection. Most methods mapping co-evolving specificity determinant sites focused positions, however, they do not consider case determinants one subclass, but variable others. In addition, many impractical very large alignments, such as those obtained Pfam, require priori information subclasses to analyzed.In this paper we apply complex networks theory, widely used analyze co-affiliation systems social and ecological contexts, map groups functional related residues. This methodology was initially evaluated simulated environments then applied four different protein families datasets, several sets motifs were successfully detected.The algorithms datasets development project available http://www.biocomp.icb.ufmg.br/biocomp/software-and-databases/networkstats/.Supplementary data at Bioinformatics online.