Residue-covariation networks cluster similar functional domains

CRISTINA ESTER MARINO; CRISTINA MARINO-BUSLJE; FRANCO L SIMONETTI

Congreso; International Society for computational biology/Asoc. Arg de Bioinformática y Biol. Computacional; 2016

ISCB/A2B2C

Coevolution networks carry information about structure, function, interactions, phylogeny and stochasticcomponents. Most works point out the usefulness of estimating coevolution to infer protein residue contacts orprotein-protein interactions. Despite some structural constraints in a given protein are critical for performing aspecific function, they are not the only constraints that made up coevolutionary information. networks clustersimilar functional domains. We hypothesize that protein families with similar function and a common evolutionaryhistory should have similar coevolution networks.We examined a set of 180 domain family hierarchies from the Conserved Domain Database. A domain familyhierarchy is a set of related Conserved Domains (CDs) that share a common ancestor, a common set of conservedresidues, and a common general function, but differ from each other in their specific phylogeny, specificfunctions, and sets of conserved residues (Figure 1A and 1B). Hierarchies have a generic "parent" and morespecific "children". In the hierarchy tree structure, the leaf domains represent highly conserved and oftenorthologous protein subgroups. Parent nodes contain a span of conserved residues that are also present in each ofthe children. Each of the child nodes can have additional conserved residues that extend beyond that span andhelp to further characterize the members of the child node.