COGcollator 2.0: a tool for analyzing distant evolutionary relationships among families of homologous proteins

The analysis of protein evolutionary relationships is of both theoretical and practical significance. Identifying and characterizing evolutionary relationships between the components of protein complexes and metabolic enzymes fosters a deeper understanding of their evolutionary history and molecular evolution. The existence of related enzymes makes the annotation of new sequences complicated, since this process relies heavily on accurately determining the protein family classification of the sequence being analyzed. Clusters of Orthologous Groups (COGs) are widely used for the classification of prokaryotic proteins. COGs are constructed based on the occurrence of the respective protein-coding genes within complete genomes. Previously, we introduced COGcollator, a tool designed to visualize the relatedness between COGs by analyzing the hits of their profile HMMs (Hidden Markov Models). This paper presents an update of the COGcollator web service. It is based on the latest version of the COG database and features a completely new interface and additional functionalities. To demonstrate the capabilities of our tool and the validity of the data, we present the COGcollator results for the subunits of NADH:quinone oxidoreductase type 1 (NDH-1), a homologue of the mitochondrial complex I, as the evolutionary relationships of NDH-1 with other protein complexes have been extensively documented in the literature. The web service is available free of charge without registration at https://boabio.belozersky.msu.ru/en/COGcollator. Through the web service interface, users can access pre-calculated COGcollator results for 4,972 COGs and download their respective profile HMMs.