1Department of Cell Research and Immunology, Tel Aviv University, Tel Aviv, Israel 2Department of Molecular Genetics, Weizmann Institute of Science, Rehovot
Comparative genomics studies of prokaryote genomes revealed a huge variance in gene content, and uncovered the ubiquity and importance of macro evolutionary events in which genes are gained and lost. We developed a novel methodology to reconstruct the co-evolutionary network among bacterial genes based on correlated events of gains and losses. By analyzing thousands of genes evolving across hundreds of genomes we found that nearly half of all gene families evolve under strong co-evolutionary constraints with their genomic partners. Importantly, we found that these partnerships elucidate complex functional associations among genes. Aiming to extract meaningful groups of genes we clustered the co-evolutionary network and found hundreds of significant clusters. Focusing on the biggest clusters (with at least 5 members) we found that over 90% of the clusters group together genes with similar biological function. These include cases of genes that are members of the same metabolic pathway (e.g., B12 synthesis), genes that contribute to the same phenotype (e.g., motility), genes which are subunits of proteins complex (e.g., NADH:ubiquinone oxidoreductase complex), or genes that are annotated with the same functional category (e.g., Cell wall/membrane biogenesis). Our findings suggest that accurate prediction of co-evolutionary interactions is a crucial step towards revealing function-specific coalitions of genes and for better understanding functional modularity within bacterial genomes. Our results further suggest that such interactions are valuable for functional annotation of genes and for the discovery of novel gene functions.
