Figure 2.
Maximum likelihood estimates of evolutionary rates for partitioned reference HIV-1 group M and SIV gp120 RNA reference. Maximum likelihood estimation of evolutionary rates in substitutions/site/year for HIV-1 group M and SIV reference sequences (LANL HIV Sequence Database) was performed in HYPHY (Pond et al. 2005) for internal branches of a fixed phylogeny. Reference-specific maximum clade credibility trees used as the fixed tree (topology and branch lengths scaled in time) were obtained from the posterior distribution of a Bayesian analysis of the same data sets. Rates were estimated for individual site partitions according to SHAPE reactivity (A), final structure pairing status (B), helical position within the final structure (C) codon positions (1, 2, and 3) within paired (P) and unpaired (U) regions according to the final structure (D), pairing probabilities (Pprob) (E), and the nucleotides identified as concordant between SHAPE reactivity and Pprob (F). Error bars represent Wald 95% confidence intervals. *P value < 0.05, **P value < 0.01, ***P value < 0.001 using profile likelihood analysis. SHAPE reactivity values were obtained from Pollom et al. (2013), whereas Pprob were obtained from Watts et al. (HIV) (2009) and Pollom et al. (SIV) (2013). The final structure referred to in the graph was derived by Pollom et al. (2013) using SHAPE reactivity constraints in RNAstructure (Mathews 2014) thermodynamic folding.