maxGapminRun: Max-gap-min-run algorithm for 2 states segmentation

In biomvRCNS: Copy Number study and Segmentation for multivariate biological data

Description

A custom Max-gap-min-run implementation using physical position for gap and run length calculation.

Usage

maxGapminRun(x, xPos = NULL, xRange = NULL, cutoff = NULL, q = 0.9, high=TRUE, minrun = 5, maxgap = 2, splitLen = Inf, na.rm=TRUE)

Arguments

x	a numeric vector for the input signal
xPos	a numeric vector, same length as x, carrying positional information for each element of x
xRange	an IRanges object, same length as x, carrying range information for each element of x
cutoff	numeric value used as cut-off, optional if q is specified
q	numeric value used to derive cut-off of x, as the q quantile of x , optional if cutoff is specified
high	TRUE if the cutoff or q here is the lower bound and values greater than the threshold are considered
minrun	minimum run length for the resulting segments
maxgap	maximum genomic distance below which two adjacent qualified tiles can be joined
splitLen	numeric value, maximum length of segments, split if too long
na.rm	TRUE if NA value should be ignored

Details

A custom Max-gap-min-run implementation using physical position for gap and run length calculation.

Value

a list of segment starts and ends indices

IS	the start index for each segment
IE	the end index for each segment
CUTOFF	the cutoff value used in the run
MG	the parameter value for maxgap
MR	the parameter value for minrun
SPL	the parameter value for splitLen

Author(s)

Yang Du

See Also

biomvRhsmm biomvRseg biomvRmgmr

Examples

	x<-rpois(50, 10)
	xpos<-rnorm(50, 300, 100)
	xpos<-xpos[order(xpos)]
	maxGapminRun(x, xpos, cutoff=9.5, maxgap=30, minrun=100)

biomvRCNS documentation built on Nov. 8, 2020, 6:49 p.m.