controlPeaks: Control parameters for ZIMHMM and ZIHMM

In plbaldoni/ZIMHMM: Improved Detection of Epigenomic Marks With Mixed Effects HMMs

Description

This function passes controlling parameters for ZIMHMM and ZIHMM. Most of these parameters control the EM algorithm

Usage

controlPeaks(epsilon.em = c(0.001, 0.001, 0.001, 0.001),
  epsilon.inner.em = 0.001, maxit.em = 500, minit.em = 3,
  gap.em = 3, maxcount.em = 3, max.phi = 1000, min.sigma2 = 1e-08,
  max.sigma2 = 10, maxcount.inner.em = 50, criterion = "MULTI",
  min.zero = .Machine$double.xmin, pcut = 0.05, quiet = F)

Arguments

epsilon.em	Either a positive value or a vector of size 4 with the convergence tolerance values for the EM algorithm (see 'criterion' below). Default is c(1e-3,1e-3,1e-3,1e-3)
epsilon.inner.em	positive convergence tolerance for the conditional maximizations used in ZIMHMM
maxit.em	integer giving the maximum number of EM iterations (default 500)
minit.em	integer giving the minimum number of EM iterations to start evaluating the convergence (default 3)
gap.em	integer giving the number of EM iterations apart to compute the convergence criterion (default 3)
maxcount.em	integer giving the number of consecutive EM iterations satisfying the convergence criterion in order to stop the algorithm (default 3)
max.phi	maximum positive value allowed for the dispersion parameters (default 1000)
min.sigma2	minimum positive value allowed for the variance component (default 1e-08)
max.sigma2	maximum positive value allowed for the variance component (default 10)
maxcount.inner.em	integer giving the maxium number of conditional maximizations used in ZIMHMM (default 50)
criterion	convergence criterion: either "MRCPE" (maximum absolute relative change in parameter estimates), "MACPE" (maximum absolute change of parameter estimates), "ARCEL" (absolute relative change of the Q-function), "ACC" (agreement of Viterbi peak calls), or "MULTI" (simultaneously check for MRCPE, MACPE, ARCEL, and ACC). For ACC, it computes the percentage of windows falling in the main diagonal of a 2 by 2 table of Viterbi predictions 'gap.em' iterations apart. Default is "MULTI"
min.zero	minimum positive value allowed in computations to avoid having zeros (default is .Machine$double.xmin)
pcut	cutoff for rejection controlled EM algorithm (default 0.05)
quiet	whether to print messages (default F)

Value

A list with components equal to the arguments

Author(s)

Pedro L. Baldoni, [e-mail]

References

https://github.com/plbaldoni/ZIMHMM

Examples

data(Huvec)
ChIP = SummarizedExperiment::assay(Huvec,'ChIP')
Control = log(SummarizedExperiment::assay(Huvec,'Control')+1)
offset = matrix(0,nrow = nrow(ChIP),ncol = ncol(ChIP))
# Setting maxit.em = 100 (no more than 100 EM iterations)
control = controlPeaks(maxit.em = 100)
## Not run: ZIMHMM(ChIP = ChIP,Control = Control,offset = offset,random = 'intercept',control = control)

plbaldoni/ZIMHMM documentation built on Dec. 5, 2019, 11:43 p.m.