CoeViz: a web-based tool for coevolution analysis of protein residues

BMC Bioinformatics

Software

Aleksey Porollo¹  Frazier N. Baker² 

[1] Center for Autoimmune Genomics and Etiology, Cincinnati Children’s Hospital Medical Center, 3333 Burnet Avenue, 45229, Cincinnati, OH, USA;Division of Biomedical Informatics, Cincinnati Children’s Hospital Medical Center, 3333 Burnet Avenue, 45229, Cincinnati, OH, USA;Department of Electrical Engineering and Computing Systems, University of Cincinnati, 2901 Woodside Drive, 45221, Cincinnati, OH, USA;Center for Autoimmune Genomics and Etiology, Cincinnati Children’s Hospital Medical Center, 3333 Burnet Avenue, 45229, Cincinnati, OH, USA;

关键词: Coevolution;    Coevolution analysis;    Coevolving residues;    Co-occurring residues;    Covariation of residues;    Protein structure;    Protein function;    Protein annotation;    Web-server;   

DOI  :  10.1186/s12859-016-0975-z

 received in 2015-10-23, accepted in 2016-03-01,  发布年份 2016

来源: Springer

【 摘 要 】

BackgroundProteins generally perform their function in a folded state. Residues forming an active site, whether it is a catalytic center or interaction interface, are frequently distant in a protein sequence. Hence, traditional sequence-based prediction methods focusing on a single residue (or a short window of residues) at a time may have difficulties in identifying and clustering the residues constituting a functional site, especially when a protein has multiple functions. Evolutionary information encoded in multiple sequence alignments is known to greatly improve sequence-based predictions. Identification of coevolving residues further advances the protein structure and function annotation by revealing cooperative pairs and higher order groupings of residues.ResultsWe present a new web-based tool (CoeViz) that provides a versatile analysis and visualization of pairwise coevolution of amino acid residues. The tool computes three covariance metrics: mutual information, chi-square statistic, Pearson correlation, and one conservation metric: joint Shannon entropy. Implemented adjustments of covariance scores include phylogeny correction, corrections for sequence dissimilarity and alignment gaps, and the average product correction. Visualization of residue relationships is enhanced by hierarchical cluster trees, heat maps, circular diagrams, and the residue highlighting in protein sequence and 3D structure. Unlike other existing tools, CoeViz is not limited to analyzing conserved domains or protein families and can process long, unstructured and multi-domain proteins thousands of residues long. Two examples are provided to illustrate the use of the tool for identification of residues (1) involved in enzymatic function, (2) forming short linear functional motifs, and (3) constituting a structural domain.ConclusionsCoeViz represents a practical resource for a quick sequence-based protein annotation for molecular biologists, e.g., for identifying putative functional clusters of residues and structural domains. CoeViz also can serve computational biologists as a resource of coevolution matrices, e.g., for developing machine learning-based prediction models. The presented tool is integrated in the POLYVIEW-2D server (http://polyview.cchmc.org/) and available from resulting pages of POLYVIEW-2D.

【 授权许可】

CC BY   
© Baker and Porollo. 2016

【 图 表 】

12864_2017_3604_Article_IEq5.gif

12864_2015_2118_Article_IEq2.gif

12864_2016_2789_Article_IEq52.gif