Using Bayes Factors via Thermodynamic Integration in Phylobayes

Question

[lorimc]

I've spent all day trying to figure this out but I can't find anything.

In the Phylobayes 3 manual its states its will calculated bayes factors via thermodynamic integration using bf and gives the following example

For example: bf -cov moc10.oes -ln moc10ln

and says

will compute the Bayes factor between the log-normal and the deconstrained model. You can do this for several models, and then choose the model returning the largest Bayes factor.

The first question I have is what files are they comparing, are they the tree files?

Also what is the best way for me to generate a dataset using a "deconstained model"?

I apologise if these questions are quite stupid, I'm not a phylogeneticist and only just get my head around the Bayes Factors!

Answer 1

[Brice Sarver]

Bayes factors can be used to compare models that aren't nested and, therefore, can't be compared using other approaches (like a likelihood ratio test, for example). All of what you said above is can be distilled down to a few main points:

1. PhyloBayes will use thermodynamic integration to calculate the marginal likelihoods. Other approaches that estimate the marginal likelihoods, such as taking the harmonic mean, have been shown to produce inaccurate results.

2. You ought to have estimates of the marginals from your data. In your example, they are testing two models. It is possible to compare other models, too. Once you have your Bayes factors, you have evidence for which model is the best fit to your data.

3. What you want are the likelihoods for your trees given the model. The trees themselves don't provide any information that could be used to select among models absent the data and some info about the model itself.

Hope this clears some things up.