RelTime relaxes the strict molecular clock throughout the phylogeny

Abstract

The RelTime method estimates divergence times when evolutionary rates vary among lineages. It has been shown to perform well in the analysis of empirical and computer simulated datasets. In addition, theoretical analyses show that RelTime relaxes the strict molecular clock throughout a molecular phylogeny. However Lozano-Fernandez et al. (2017) suggested that RelTime imposes a strict molecular clock for deep divergences in an analysis of one dataset, whereas Bayesian methods do not. We show that their observation is caused primarily by the lack of equivalence between analysis conditions in the application of RelTime and Bayesian methods. We also found comparable differences even among Bayesian estimates when different root priors were used. Overall, less restrictive prior selections produced Bayesian results concordant with RelTime. Therefore, our results suggest that Bayesian estimates obtained under a range of priors need to be considered and presented in any discourse about molecular dating and comparisons with RelTime.