Figure 2.

Amino acid sequences of present-day and ancestral ECP and EDN proteins. Amino acids are presented by single-letter codes, and dots show the same amino acids as those of the sequence at node a. The arginine (R) residues of ECP and EDN are shown in bold type when they are not identical with those of the ancestral protein at node a. Ambiguous amino acid sites of ancestral sequences at nodes a, b, and c are underlined, where the posterior probability of the most likely amino acid is lower than twice the probability of the second most likely amino acid. The alternative amino acids at the underlined sites were K, R, M, T, Q, N, and T for sites 28, 72, 87, 88, 104, 111, and 113, respectively. These ambiguities occurred almost always at the sites where the orangutan ECP and EDN have the same amino acids but different ones from those of the other orthologous sequences. The overall accuracies (posterior probabilities) of the ancestral proteins were 0.96, 0.97, and 0.98, for nodes a, b, and c, respectively.