2.6 Positive selection analyses
We indicated positive selection by the ratio of the rate of nonsynonymous to synonymous substitutions (dN/dS or ω) per gene among lineages. Using our phylogenetic tree topology as the guide tree, we ran branch-site model (Yang, 2007) (Parameters: null hypothesis: model=2, NSsites=2, fix_omega=1, omega=1; alternative hypothesis: model=2, NSsites-2, fix_omega=0, omega=1) in the codeml module of PAML4 to detect positive selection for the one-to-one orthologous genes of five species within the macrotisgroup, respectively. For each analysis, we selected only one species as foreground branch, and regarded all other species as the background branches. A likelihood rate test (LRT) was performed to detect positive selection on the foreground branch, and the Bayes Empirical Bayes (BEB) analysis (Yang et al., 2005) was implemented to calculate posterior probabilities and to deduce positively selected sites. P -values of all positively selected genes (PSGs) were also normalized by FDR using Benjamini-Hochberg approach(Benjamini and Hochberg, 1995) andP value less than 0.05 was considered significant.