Figure 1

Reconstruction of ADAR gene and domain evolution. The table (right) lists the number of ADAR family members identified in each species. ADARs are classified based on their domain architecture, as shown by the ‘ball-and-stick’ protein models above each ADAR name. The Z-DNA/RNA binding (ZB) and double-stranded RNA binding (dsRB) domains of the ADAR1 model are marked with an ‘n’ to indicate that multiple copies of these domains may be present in different species. The domain architectures of all ADAR1-like proteins are depicted on the far right. The ADAR gene counts were used to reconstruct ADAT/ADAR evolution, as mapped to the phylogenetic tree as coloured squares (left). Searches for adenosine deaminase (AD), dsRB and ZB domains were performed to determine the phylogenetic positions of whole-genome domain origin and loss events, regardless of ADAT/ADAR complement; these events are also mapped to the tree as coloured shapes. Green boxes separate the tree into the main phylogenetic groupings: Bilateria (B), Eumetazoa (E), Metazoa (M), Holozoa (H) and Opisthokonta (O). For clarity, we present the sponge and ctenophore lineages on equal footing, and depict all three ADARs as present in the metazoan stem. The loss and gain of the ADAR1-like gene is marked with a question mark to illustrate the uncertainty in reconstructing these evolutionary events, which are elaborated upon further in Figure 3 and Additional file 2.