4.2. Conserved Sequence and Phylogenetic Analysis

For conserved domain analysis, the sequences were extracted from the full-length of CO-like protein sequences according to the domain positions retrieved from Interproscan 63.0, and ClustalX 2.0 was used for multiple-sequence alignments with the default parameters. For phylogenetic analysis, we used the ClustalW program (build-in MEGA 7.0) [39] to perform the multiple sequences alignment using the full-length sequences from rice, sorghum, maize, cacao, Arabidopsis, G. arboreum, G. raimondii, and G. hirsutum. Thereafter, we prepared a phylogenetic tree using the neighbour-joining (NJ) method with 1000 bootstrap replicates. Substitution was evaluated by the Poisson model using the default parameters. To validate the NJ tree, we used the Maximum likelihood method to construct an ML tree. Substitution was evaluated by the Poisson model, and bootstrap method (100 replications) was used to test the tree. A discrete Gamma distribution was used to model evolutionary rate differences among sites (5 categories (+G, parameter = 0.8524)). All positions with less than 95% site coverage were eliminated.