cnSegments.paired: Identifies regions of differential copy-number alteration...

In chrisamiller/copyCat: Find regions of genomic copy number loss and gain from short reads

Description

uses Circular Binary Segmentation (as implemented in the DNAcopy package) to segment the genome and returns regions that contain CN gains or losses relative to the reference

Usage

  cnSegments.paired(rdo.ref, rdo.test, onlyAlts=FALSE, minWidth=3, alpha=0.01, undoSD=2, rmGaps=TRUE)

Arguments

rdo.ref	a readDepth object created from the reference sample using new("rdObject") and filled with read counts using the getReadDepth() command (optionally corrected first).
rdo.test	same as rdo, but containing the test sample's reads
onlyAlts	whether the function should return just those segments that exceed the threshold values for gain and loss The Default is FALSE, which returns all segments.
minWidth	the minimum number of consecutive aberrant bins required before making a segment call. Default is three.
alpha	the significance level for the CBS to accept change-points. Default 0.01
undoSD	require this many SDs between segment means to keep a split
rmGaps	Controls whether adjacent segments have their boundaries adjusted to remove gaps. If so, the new segment boundary is set to halfway between the existing boundary calls.

Value

returns a dataframe with 5 columns: chr start stop #probes copynumber

See Also

'readDepth'

Examples

  #create a rdObject, set bin params
  #  x <- new("rdObject")

  #fill the bins with reads
  #  x.reads <- readDepth(x)

  #do gc correction on the read counts
  #  x.gcCorr <- rd.gcCorrect(x)

  #return regions of CN gain and loss
  #  segs <- rd.cnSegments(x.gcCorr)

chrisamiller/copyCat documentation built on July 20, 2021, 12:59 a.m.