estimateFDR: Estimate false discovery rate of copy number variation...

In hshdndx/new-to-CNV: patternCNV: a versatile tool for detecting copy number changes from exome-seq data

Description

This function estimates false discovery rate (FDR) of CNV log2-ratio values, with given Null distribution model returned by fitNullModel

Usage

estimateFDR(data.name, null.model, 
                   FDR.type = "q.value", is.plot = FALSE)

Arguments

data.name	an object of PatCNVData-class
null.model	a list of Null distribution model returned by fitNullModel
FDR.type	"q.value" or "localFDR"
is.plot	logical, whether intermediate plots are generated

Value

an object of PatCNVData-class, with FDR results computed

Author(s)

Chen Wang

See Also

compute CNV log2-ratio values of multiple samples computeMultiCNV

Examples

#=== load simulated data with pre-computed coverage and CNV
data(sim.cnvData)

#=== fit null distribution of CNV signals and plot distribution
null.model <- fitNullModel(sim.cnvData, "all")
germline.data <- estimateFDR(sim.cnvData, null.model)

#=== plot distribution of FDR of sample "n2"
hist(fdrMatrix(germline.data,"n2"), xlab="FDR", 
     main="FDR distribution of sample n2")

hshdndx/new-to-CNV documentation built on May 17, 2019, 5:55 p.m.