Figure 10

Comparison of ortholog assignments made by Ensembl, MSOAR and MSOAR 2.0 for the two segments of human chromosome 2 (178,123,219 bp-178,685,428 bp) and mouse chromosome 2 (75,773,906 bp-76,192,000 bp). Among the 7 pairs of genes illustrated in the figure, only (TTC30B, Ttc30b) and (PDE11A, Pde11a) are known one-to-one ortholog pairs according to gene symbols, as indicated by solid lines. Since the Ensembl ortholog database includes many-to-many relationship, it outputs 7 ortholog pairs, i.e., (TTC30B, Ttc30b), (TTC30B, Ttc30a2), (TTC30B, Ttc30a1), (TTC30A, Ttc30b), (TTC30A, Ttc30a2), (TTC30A, Ttc30a1), and (PDE11A, Pde11a), introducing 5 false ortholog pairs, as indicated by dashed lines. MSOAR assigns three one-to-one ortholog pairs as indicated by the arrows in the figure, i.e., (TTC30B, Ttc30b), (TTC30A, Ttc30a1), and (PDE11A, Pde11a), including one false one-to-one ortholog pair. MSOAR 2.0, however, identifies TTC30A as an inparalog of TTC30B on the human genome and Ttc30a2 and Ttc30a1 as inparalogs of Ttc30b on the mouse genome during the phylogenetic analysis of TAGs, and removes them before invoking MSOAR. Thus, MSOAR 2.0 only outputs two one-to-one ortholog pairs, i.e., (TTC30B, Ttc30b) and (PDE11A, Pde11a), both of which are true positives.