eggNOG: automated construction and annotation of orthologous groups of genes

Abstract

The identification of orthologous genes forms the basis for most comparative genomics studies. Existing approaches either lack functional annotation of the identified orthologous groups, hampering the interpretation of subsequent results, or are manually annotated and thus lag behind the rapid sequencing of new genomes. Here we present the eggNOG database ('evolutionary genealogy of genes: Non-supervised Orthologous Groups'), which contains orthologous groups constructed from Smith-Waterman alignments through identification of reciprocal best matches and triangular linkage clustering. Applying this procedure to 312 bacterial, 26 archaeal and 35 eukaryotic genomes yielded 43 582 course-grained orthologous groups of which 9724 are extended versions of those from the original COG/KOG database. We also constructed more fine-grained groups for selected subsets of organisms, such as the 19 914 mammalian orthologous groups. We automatically annotated our non-supervised orthologous groups with functional descriptions, which were derived by identifying common denominators for the genes based on their individual textual descriptions, annotated functional categories, and predicted protein domains. The orthologous groups in eggNOG contain 1 241 751 genes and provide at least a broad functional description for 77% of them. Users can query the resource for individual genes via a web interface or download the complete set of orthologous groups at http://eggnog.embl.de.

Figure 1.

Statistics on the content of the eggNOG database. The eggNOG assignments for 373 complete genomes [19] were mapped onto the tree of life. The stacked bar charts outside the tree show the proportion of genes from each genome that can be assigned to a functionally annotated orthologous group (green), to an unannotated orthologous group (orange) or to no orthologous group (grey). The length of each bar is proportional to the logarithm of the number of genes in the respective genome. The pie charts inside the tree show the fractions of orthologous groups at each level in the hierarchy that could be annotated with a function description (green for NOGs, light green for extended COGs and KOGs) and that could not be functionally annotated (orange for NOGs, light orange for extended COGs and KOGs). The areas of the pie charts are proportional to the number of orthologous groups at the phylogenetic level in question. This figure was made using iTOL [20].

Figure 2.

Screenshot of the main results page. The eggNOG database was queried for the three G₁-type cyclins in budding yeast, namely Cln1–Cln3. These have been correctly assigned to two fungal orthologous groups. The navigation tree at the top of the page allows the user to change the view to more coarse-grained orthologous groups, for example the eukaryotic orthologous groups in which these cyclins are all grouped together.