findMotifRegion: Use Hocomoco motif and homer to identify motifs in a given...

In ELMER: Inferring Regulatory Element Landscapes and Transcription Factor Networks Using Cancer Methylomes

Description

To find for each probe the know motif we will use HOMER software (http://homer.salk.edu/homer/). Homer and genome should be installed before this function is executed Step: 1 - get DNA methylation probes annotation with the regions 2 - Make a bed file from it 3 - Execute section: Finding Instance of Specific Motifs from http://homer.salk.edu/homer/ngs/peakMotifs.html to the HOCOMOCO TF motifs Also, As HOMER is using more RAM than the available we will split the files in to 100k probes. Obs: for each probe we create a winddow of 500 bp (-size 500) around it. This might lead to false positives, but will not have false negatives. The false posives will be removed latter with some statistical tests.

Usage

findMotifRegion(
  regions,
  output.filename = "mapped_motifs_regions.txt",
  region.size = NULL,
  genome = "hg38",
  nstep = 10000,
  cores = 1
)

Arguments

regions	A GRanges object. Names will be used as the identifier.
output.filename	Final file name
region.size	If NULL the motif will be mapped to the region. If set a window around its center will be considered. For example if region.size is 500, then +-250bp round it will be searched.
genome	Homer genome (hg38, hg19)
nstep	Number of regions to evaluate in homer, the bigger, more memory it will use at each step.
cores	A interger which defines the number of cores to be used in parallel process. Default is 1: no parallel process.

Examples

## Not run: 
 # use the center of the region and +-250bp around it
 gr0 <- GRanges(Rle(c("chr2", "chr2", "chr1", "chr3"), 
                    c(1, 3, 2, 4)
                    ), 
               IRanges(1:10, width=10:1)
               )
 names(gr0) <- paste0("ID",c(1:10))
 findMotifRegion(regions = gr0, region.size = 500, genome = "hg38", cores = 1)
 
 # use the region size itself
 gr1 <- GRanges(Rle(c("chr2", "chr2", "chr1", "chr3"), c(1, 3, 2, 4)), 
                IRanges(1:10, width=sample(200:1000,10)))
 names(gr1) <- paste0("ID",c(1:10))
 findMotifRegion(regions = gr0, genome = "hg38", cores = 1)

## End(Not run)

ELMER documentation built on Nov. 8, 2020, 4:59 p.m.