viterbi {VanillaICE}R Documentation

viterbi algorithm

Description

The Viterbi algorithm for computing the most likely state sequence given a model

Usage

viterbi(beta, pi, tau, states, arm, tau.scale)

Arguments

beta emission probabilities (log scale)
pi initial state probabilities (log scale)
tau transition probabilities (log scale)
states character string of hidden states
arm character string of chromosomal arm
tau.scale scales the probability of transitioning between an altered state to a normal state relative to the probability of transitioning between two altered states.

Details

The Viterbi algorithm is fit independently to each chromosomal arm.

Argument tau.scale is a matrix that scales the probability of transitioning from an altered state to a normal state to the probability of transitioning between two altered states.

When initializing an object of class HmmParameter, the default is to make the probability of transitioning from an altered state to a normal state twice the probability of transitioning between two altered states. Further flexibility could be obtained by replacing the tau.scale matrix in an instance of the HmmParameter class with your own matrix. The dimension of this matrix must be S x S, where S is the number of hidden states.

Value

vector of integers

Author(s)

R. Scharpf


[Package VanillaICE version 1.2.0 Index]