callPeaks: Summarize peak calls and optionally create a BED 6+3 file in...

In plbaldoni/epigrahmm: Epigenomic R-based analysis with hidden Markov models

Summarize peak calls and optionally create a BED 6+3 file in broadPeak format for visualization

Description

This function imports the output from 'epigraHMM' and outputs a set of peaks (consensus or differential) for a given FDR control threshold or Viterbi sequence.

Usage

callPeaks(
  object,
  hdf5 = metadata(object)$output,
  method = "viterbi",
  saveToFile = FALSE,
  control = NULL
)

Arguments

object	an epigraHMMDataSet
hdf5	a character with the location of the epigraHMM HDF5 output file
method	either 'viterbi' or a numeric FDR control threshold (e.g. 0.05). Default is 'viterbi'.
saveToFile	a logical indicating whether or not to save the results to file. Output files are always saved with peaks of interest defined on the region level. Default is FALSE.
control	list of control arguments from controlEM(). This is an optional parameter and it is only required when 'saveToFile = TRUE' so that the output directory can be obtained. Default is NULL.

Value

A GRanges object with differential peak calls in BED 6+3 format

Author(s)

Pedro L. Baldoni, [e-mail]

References

https://github.com/plbaldoni/epigraHMM

Examples

# Creating dummy object
countData <- rbind(matrix(rnbinom(1e3,mu = 2,size = 10),ncol = 1),
                   matrix(rnbinom(2e3,mu = 7.5,size = 5),ncol = 1),
                   matrix(rnbinom(1e3,mu = 2,size = 10),ncol = 1))

colData <- data.frame(condition = 'A', replicate = 1)

rowRanges <- GenomicRanges::GRanges('chrA',
IRanges::IRanges(start = seq(from = 1, length.out = 4e3,by = 250),width = 250))

object <- epigraHMMDataSetFromMatrix(countData,colData,rowRanges)

# Initializing
object <- initializer(object,controlEM())

# Running epigraHMM
object <- epigraHMM(object,controlEM(),type = 'consensus',dist = 'nb')

# Calling peaks
peaks <- callPeaks(object = object,
                   hdf5 = S4Vectors::metadata(object)$output,
                   method = 'viterbi')

plbaldoni/epigrahmm documentation built on Oct. 14, 2023, 5:13 a.m.