SimBindProfiles (version 1.10.0)

compensationRegions: Classify two Binding Profiles - Compensation

Description

This function is used to classify three Binding Profiles into regions which are bound in data set one and two but not three.

Usage

compensationRegions(xSet, sgrset = c(1, 2, 3), bound.cutoff, diff.cutoff,  probes, probe.max.spacing, writeBedFile = TRUE)

Arguments

xSet

object of class ExpressionSet

sgrset

vector of lenght 3; specifying which data sets to compare from the ExpressionSet

bound.cutoff

numeric; threshold above probes are considered “bound”

diff.cutoff

numeric; difference threshold to determine if objects uniquely bound

probes

integer; minimum number of probes in a valid region

probe.max.spacing

integer; maximum number of base pairs in a gap before splitting a region into 2 regions

writeBedFile

logical; should bed file be written

Value

data.frame with the following columns:

name

name(s) of data set to which region belongs

class.group

class group, in this case only 1

chr

chromsome

start

start position of region

end

end position of region

scrore

score of region

nProbes

number of probes in region

Details

Select probes with a signal above the bound.cutoff in data set 1 and 2 and below in set 3, and for which the average signal of set 1 and set2 is above the diff.cutoff compared to set 3. These probes are then filtered into regions using the probes and probe.max.spacing details. The score is calculated as mean (probes in region average(set 1 and set 2) minus set 3). Optional bed file formated result files are written using the choosen options in the file names.

See Also

pairwiseRegions, threewayRegions, increasedBindingRegions

Examples

  dataPath <- system.file("data",package="SimBindProfiles")
  load(file.path(dataPath,"SGR.RData"))
  transcompABC <- compensationRegions(SGR, sgrset=c(1,2,3), bound.cutoff=1.86, 
                  diff.cutoff=1.4, probes=10, probe.max.spacing=200)