Discovering New Orthologous Groups with P-COGs.

Deciphering the evolutionary history of genes is foundational for biomedical research, enabling the identification of compensatory mutations in disease-associated genes and the selection of evolutionarily relevant model organisms. However, challenges of multi-isoform handling, the presence of incomplete genomes/proteomes initial datasets, and problems of evolutionary history representation are still present. To address these issues, we introduce a novel framework based on the Clusters of Orthologous Groups (COG) method, primarily designed for improving phylogenetic tree clustering in studying evolutionary history of eukaryotes, P-COGs (Pavlov's COGs). It encourages the use of a single sequence dataset for both constructing phylogenetic trees and inferring COGs. We demonstrate the tool's utility through an evolutionary study of the voltage-dependent chloride channel genes family (CLCN). Moreover, P-COGs allowed us to observe multiple clusters of orthologous genes in CLCN that were not identified by other COG-based tools. Additionally, we resolved the evolutionary history of hundreds of cancer-associated genes with P-COGs to support accurate evolution-based variant effect prediction. The resulting COG graphs are accessible via our interactive web application ( https://epicenter.1spbgmu.ru/shiny/pavlovscogs/ ). P-COGs is openly available at https://github.com/bugds/Pavlovs_COGs .