Skip to main content

View full-text article in PMC

. 2017 May 1;8:737. doi: 10.3389/fmicb.2017.00737

Table 1.

Mean values and ESSs of key BEAST run statistics computed using GTR substitution model, strict molecular clock, and constant population size models.

Clock model	Population model	Substitution model^a	Tree		Posterior		Clock rate		Tree model root height		Constant population size^c
			Mean likelihood	ESS^b	Mean	ESS^b	Mean	ESS^b	Mean	ESS^b	Mean	ESS^b
Lognormal relaxed^d	Constant	GTR	-6.532260E+06	53862	-6.532942E+06	5010	7.21E-07	3272	35	4390	66	5300
Lognormal relaxed	Bayesian skyline	GTR	-6.532256E+06	10179	-6.535220E+06	577	8.73E-07	561	23	746	na	na
Strict	Constant	GTR	-6.532319E+06	15931	-6.533004E+06	3676	7.28E-07	2616	31	2963	68	4747
Strict	Bayesian skyline	GTR	-6.532318E+06	7905	-6.535308E+06	3050	7.46E-07	2503	29	2958	na	na

^aGTR, general time-reversible substitution model.

^bESS, effective sample size.

^cna, not applicable.

^dThe parameters shown are a result of four combined BEAST runs initiated from four different random seeds (Supplementary Material file log.sh).