Scientific sessions, PRBB CRG Conferences
Eugene V. Koonin
National Center for Biotechnology, Information National Library of Medicine, NIH. Bethesda, USA
Eugene Koonin, from the National Centre for Biotechnology Information National Library of Medicine, NIH, Bethesda, US, is interested in understanding the evolution of life, and uses existing and new methods of computational biology to perform research in several major areas, including empirical comparative and evolutionary genomics of prokaryotic and eukaryotic genomes or the origin and evolution of viruses. He has been invited by Toni Gabaldón (CRG).
Gene orthology arguably is the central concept of evolutionary genomics. Orthologs are of homologous genes that are related by speciation, i.e. form follow a vertical pattern of descent along substantial evolutionary spans. As captured in the so-called “orthology conjecture”, orthologous genes are generally assumed to perform equivalent functions in different organisms and to share other key properties. Attempts to test the orthology conjecture on genome-wide data led to considerable controversy but at the end do not seem to falsify the predictions. However, the character and tractability of orthologous relationships critically depend, on the one hand, on the evolutionary distance between the compared genomes, and on the other hand, on the functional class of analysed genes. At short evolutionary distances, orthology between the majority of the genes can be easily established. At large distances, such as those between different phyla, simple orthologous relationships are detectable only for a small subset of the most highly conserved genes. Part of the problem is extreme sequence divergence that in many cases can be overcome by using sensitive methods for sequence and structure comparison. More commonly, however, orthology relationships between genes from distant organisms are obscured by the complex history of evolutionary events distinct from vertical descent, such as horizontal gene transfer, lineage-specific gene loss and gene duplication. Perhaps the most severe problem for the traditional concept of orthology is the evolutionary diversification of the domain architectures of protein-coding genes. The results of a recent phylogenomic analysis show that, when it comes to comparison of eukaryotic genomes from distant taxa, changes in domain architecture are so pervasive that gene orthology becomes effectively irrelevant and has to be replaced with the notion of orthology between evolutionary domains.