Skip to main content
. 2014 Oct 16;8(10):e3251. doi: 10.1371/journal.pntd.0003251

Table 2. Molecular clock model output.

Molecular clock Population model prior AICM++ Mean substitution rate (95% HPD) TMRCA Grenadian strains* Substitute rate Grenadian branch
Strict Bayesian skyline 28430 3.26×10−4 (2.49–4.04×10−4) 16.4 yrs ago (8.6–25.1) 3.26×10−4 (2.49–4.04×10−4)
Strict Constant 28440 2.78×10−4 (1.96–3.57×10−4) 23.4 yrs ago (14.6–34.8) 2.78×10−4 (1.96–3.57×10−4)
Uncorrelated lognormal Bayesian skyline 28360 3.99×10−4 (2.88–5.15×10−4) 11.6 yrs ago (5.5–23.0) 5.09×10−4 (2.99–7.40×10−4)
Uncorrelated lognormal Constant 28370 3.17×10−4 (2.01–4.33×10−4) 22.7 yrs ago (12.6–37.8) 4.36×10−4 (2.38–6.94×10−4)

* years since most recent strains (2013),

++

AICM is an analogue of Akaike's information criterion estimated through Markov Chain Monte Carlo simulation using the method-of-moments estimator [33]. Lower values indicate better model fit.