We apply Hidden Markov Model to phylogenetic tree construction; that is, wecompute the probabilities of the branches of the phylogentic tree and decidewhether to accept it or not. The probabilities are the proposal distributions inthe Gibbs sampler method. We will discuss the method in chapter 5.Carrying out simultaneous tree-building and alignment of sequence data is adifficult computational task. Many methods are either limited to a few sequencesor restricted to highly simplified models of alignment and phylogeny. A methodis given in chapter 6 for overcoming these limitations. We introduce a simplemethod in chapter 5 in order to help us to understand the method in chapter6. In chapter 5, we introduce a sampling method. This sampling method do amultiple sequences alignment conditioned on a phylogenetic tree. However wecan do the two things simultaneously in chapter 6.We will first introduce the basic concepts of the Hidden Markov models. Thenwe will discuss the phylogenetic tree and three kinds of the phylogenetic treeconstruction methods most commonly used. And we will use the Gibbs samplingmethod of the Markov chain Monte Carlo to combine the Hidden Markov modelsand sample the phylogenetic tree in a tree Hidden Markov model.
