Accepted_test

COGcollator 2.0: an improved web server for analysis of distant relationships between homologous protein families

by Daria V. Dibrova | Stanislav Y. Rykov | Belozersky Institute of Physico-Chemical Biology, Lomonosov Moscow State University, 119234 Moscow, Russia; School of Bioengineering and Bioinformatics, Lomonosov Moscow State University, 119234 Moscow, Russia | Institute of Philosophy, Russian Academy of Sciences, 109240, Moscow, Russia

Abstract ID: 37

Event: BGRS-abstracts

Sections: [Sym 1] Section “Structural-functional organization of genomes and transcriptomics”

Functional analysis of a protein sequence starts with a search for similar proteins, however not always the function could be transferred from one homologous protein to another. Distinguishing between orthologs and paralogs is important for the adequate protein annotation and interpretation of phylogenetic trees. An approach of constructing Clusters of Orthologous Groups (COGs) and assigning proteins to them proved fruitful even when a dozen of genomes was sequenced, and it is widely applied until now. Here, we provide a major update to our tool which allows to investigate relationships between different COGs and access the data in the COG database. COGcollator 2.0 has an improved web interface with new features, it covers more COGs and matches the last release of the COG database. The web server for COGcollator 2.0 is freely available at: http://boabio.belozersky.msu.ru/en/COGcollator. All constructed profile HMMs for 4870 COGs are united in the database and are available for download. We also provide another option to access our data: DomainAnalyser, a highly versatile sequence search tool which allows to visualize a domain structure of a protein set in both interactive and downloadable way. This tool is freely available at: http://boabio.belozersky.msu.ru/en/DomainAnalyser.