contHMM-class

initialize,contHMM

Class for HMMs with continuous emission distributions.

classes

models

Methods and classes to build HMMs
that are suitable for the analysis of ChIP-chip data. The
provided parameter estimation methods include the Baum-Welch
algorithm and Viterbi training as well as a combination of
both.

Peter Humburg

tileHMM

Hidden Markov Models for ChIP-on-Chip Analysis

contHMM-class function

Objects can be created by calls of the form <code>new("contHMM", transition, emission, init)</code>.
For the special case of <em>t</em> distributions objects of this class can be created more
conveniently by a call to <code>getHMM</code>.  The function <code>hmm.setup</code> provides facilities to
create <code>"contHMM"</code> objects with initial parameter estimates obtained from data.

Objects from the Class

<dl>
    <dt><code>transition.matrix</code>:</dt><dd>Object of class <code>"matrix"</code>, storing the transition probabilities of the Markov chain.</dd> <dt><code>emission</code>:</dt><dd>Object of class <code>"list"</code> containing objects of class <code>"contDist"</code> to represent emission distributions for each state.</dd> <dt><code>init</code>:</dt><dd>Object of class <code>"numeric"</code>. The initial state distribution of the Markov chain.</dd> </dl> 

Slots

Extends

<dl>
   <dt>sampleSeq</dt><dd><code>signature(hmm = "contHMM", size = "numeric")</code>: Sample observation sequence of length <code>size</code> from model </dd> </dl> 

contHMM-class: Class "contHMM"

Description

Arguments

Objects from the Class

Slots

Extends

Methods

See Also

Examples