One hundred single locus data sets were generated for each of the following sequence lengths: 250, 500, 1000, 5000, and 10,000 bp for each of the 4 trees.
The programs generally performed well on shallow phylogenies, but posterior mean node ages were biased and 95% posterior intervals included true ages in fewer than 95% of trees in some analyses.
Branch lengths were calculated for each tree by multiplying the time duration () by the rate, and then, 10 tip sequences were generated using EVOLVER from the PAML package (version 4, Yang 2007).
This was done using the F84 model (Hasegawa and Kishino 1989) with base frequencies 0.30 [T], 0.25 [C], 0.30 [A], 0.15 [G]; a transition to transversion ratio κ = 5; and a gamma shape parameter α = 0.5 in order to simulate typical mt DNA sequences.
2008), have made the Bayesian MCMC approach increasingly popular in phylogenetic studies.
To date, the suitability of Bayesian relaxed clock analyses of divergence times in shallow phylogenies has not been rigorously assessed. Most previous investigations of relaxed clock Bayesian dating have used longer more divergent sequences in which the likelihood has a greater influence on the posterior.
Bayesian methods are increasingly being used to estimate divergence times without the restrictive assumption of a global clock.