Skip to main content
. Author manuscript; available in PMC: 2010 Aug 18.
Published in final edited form as: AIDS. 2009 Nov 27;23(18):2523–2532. doi: 10.1097/QAD.0b013e3283320ef3

Table 3.

Bayes factor comparisons between different evolutionary models for HIV-1C pol in Zimbabwe.

Model comparison log10 Bayes factor Evidence against Ho
Const Strict (H0) vs. relaxed (H1) clock 27.732 Very strong
Expo Strict (H0) vs. relaxed (H1) clock 19.496 Very strong
BSP Strict (H0) vs. relaxed (H1) clock 17.736 Very strong
Const (H0) vs. Expo (H1) relaxed clock 0.86 Positive
Const (H0) vs. BSP (H1) relaxed clock 3.694 Very strong
BSP (H0) vs. Expo (H1) relaxed clock 4.554 Very strong

BF, Bayes factor is the difference (in log space) of the marginal likelihood of the null (H0) and alternative (H1) models. BFs were estimated by comparing the approximate marginal likelihoods of the different models given in Table 3. Const, constant population size; Expo, exponential population growth; BSP, Bayesian skyline plot; Strict, strict molecular clock; Relaxed, relaxed molecular clock.