fit.model: Fitting a heterogeneous HMM to the log2 ratios on a particular chromosome.

Description

This function fits five homogeneous HMMs to the log2 ratios on a particular chromosome. It then uses either the AIC or BIC to determine which of the five models is optimal before using a scaled version of the Viterbi algorithm to assign clones to states with the same underlying copy number.

Usage

fit.model(sample, chrom, dat, datainfo = clones.info, useCloneDists = TRUE, covariates,
aic = TRUE, bic = FALSE, delta = 1, var.fixed=FALSE, epsilon = 1e-06,
numiter = 30000)

Arguments

sample

If there are multiple samples, the number of the sample to be segmented

chrom

The chromosome on which the segmentation is to be carried out on

dat

The log2 ratios obtained from the clones located on that chromosome

datainfo

A dataframe containing information about the clones on that chromosome (name, chromosome and location (in Mbs))

useCloneDists

Boolean stating whether the distance between clones should be incorportated into the HMM. If false then the HMM become homogeneous.

covariates

A matrix containing the covariate information for the clones located on the chromosome to be segmented. It should have length one less than the number of clones as covariate information is not used when segmenting the first clone on the chromosome.

aic

Set to true if you want to use the aic. This is the default. Only one of aic and bic should be set to true.

bic

Set to true if you want to use the bic.

delta

A parameter to be set if you want to use the BIC

var.fixed

Logical variable - TRUE if you want to tie the variance to be the same across all states. Defaults to FALSE

epsilon

numiter

Number of iterations to be used in the optimization algorithm.

Value

The output is in the same format as that obtained when the nlm function is applied.