14.0 Markov Chain Monte Carlo (MCMC)

(Win version)

< 13.0 | 14.0 | 14.1 >

The next step is to create a Chain class that carries out a Markov chain Monte Carlo (MCMC) simulation for the purpose of sampling from a Bayesian posterior distribution. Our initial effort will update only the gamma shape parameter used to model among-site rate heterogeneity. Later we will add the ability to update other substitution model parameters (e.g. nucleotide state frequencies and GTR exchangeabilities) as well as the tree topology and edge lengths.

Step Title Description
Step 14.1 The Updater Class

Create the Updater class that will serve as the base class for classes that update model parameters

Step 14.2 The GammaRateVarUpdater Class

Create a class derived from Updater that can update the rate variance parameter governing the amount of among-site rate heterogeneity

Step 14.3 The Chain Class

Create a class that encapsulates a Markov chain that can carry out MCMC

Step 14.4 Testing the Chain Class

Perform an MCMC analysis to test the new Chain class

< 13.0 | 14.0 | 14.1 >