initHMM

The observations. A list of one or more entries containing the observation matrix (<code>numeric</code>) for the samples (e.g. chromosomes).

nStates

Emission distribution of the model. One out of c("NegativeBinomial", "PoissonLogNormal", "NegativeMultinomial", "ZINegativeBinomial", "Poisson", "Bernoulli", "Gaussian", "IndependentGaussian")

method

Library size factors for Emissions PoissonLogNormal or NegativeBinomial as a length(obs) x ncol(obs[[1]]) matrix.

sizeFactors

If TRUE, (co-)variance of (Independent)Gaussian is shared over states. Only applicable to 'Gaussian' or 'IndependentGaussian' emissions. Default: FALSE.

sharedCov

Genome segmentation with hidden Markov models has become a useful tool to annotate genomic elements, such as promoters and enhancers. STAN (genomic STate ANnotation) implements (bidirectional) hidden Markov models (HMMs) using a variety of different probability distributions, which can model a wide range of current genomic data (e.g. continuous, discrete, binary). STAN de novo learns and annotates the genome into a given number of 'genomic states'. The 'genomic states' may for instance reflect distinct genome-associated protein complexes (e.g. 'transcription states') or describe recurring patterns of chromatin features (referred to as 'chromatin states'). Unlike other tools, STAN also allows for the integration of strand-specific (e.g. RNA)  and non-strand-specific data (e.g. ChIP).

initHMM: Initialization of hidden Markov models

Description

Usage

Arguments

Value

Examples