relic: REgression on Logarithm of Internal Control probes (RELIC)

In ENmix: Quality control and analysis tools for Illumina DNA methylation BeadChip

Description

REgression on Logarithm of Internal Control probes (RELIC) correct for dye bias on whole array by utilizing the intensity values of paired internal control probes that monitor the two color channels.

Usage

relic (mdat,at_red,cg_grn)

Arguments

mdat	An object of class methDataSet or MethylSet.
at_red	an intensity matrix for Illumina control probes "NORM_A" and "NORM_T"
cg_grn	an intensity matrix for Illumina control probes "NORM_C" and "NORM_G"

Details

The Illumina MethylationEPIC BeadChip contains 85 pairs of internal normalization control probes (name with prefix NORM_A, NORM_T, NORM_G or NORM_C), while its predecessor, Illumina HumanMethyl-ation450 BeadChip contains 93 pairs. RELIC first performs a regression on the logarithms of the intensity values of the normalization control probes to derive a quantitative relationship between red and green channels, and then uses the relationship to correct for dye-bias on intensity values for whole array.

Value

An object of class methDataSet or MethylSet depends on input class.

Author(s)

Zongli Xu and Liang Niu

References

Zongli Xu, Sabine A. S. Langie, Patrick De Boever, Jack A. Taylor and Liang Niu, RELIC: a novel dye-bias correction method for Illumina Methylation BeadChip, BMC Genomics. 2017

See Also

Package preprocessENmix

Examples

if (require(minfiData)) {

##background correction and dye bias correction
#rgDataSet as input
path <- file.path(find.package("minfiData"),"extdata")
rgSet <- readidat(path = path,recursive = TRUE)
mdat <- preprocessENmix(rgSet,bgParaEst="oob",nCores=6,dyeCorr ="RELIC")

#RGChannelSet as input
mdat=preprocessENmix(RGsetEx,bgParaEst="oob",nCores=6,dyeCorr ="RELIC")

##dye bias correction only
#methDataSet as input
path <- file.path(find.package("minfiData"),"extdata")
rgSet <- readidat(path = path,recursive = TRUE)
ctrls <- getCGinfo(rgSet,type="ctrl")
ctrls <- ctrls[ctrls$Address %in% rownames(rgSet),]

ctrl_r <- assays(rgSet)$Red[ctrls$Address,]
ctrl_g <- assays(rgSet)$Green[ctrls$Address,]
CG.controls <- ctrls$Type %in% c("NORM_C", "NORM_G")
AT.controls <- ctrls$Type %in% c("NORM_A", "NORM_T")
cg_grn=ctrl_g[CG.controls,]
at_red=ctrl_r[AT.controls,]
rownames(cg_grn) = ctrls$ExtendedType[CG.controls]
rownames(at_red) = ctrls$ExtendedType[AT.controls]
mdat=getmeth(rgSet)
mdat <- relic(mdat,at_red,cg_grn)

#MethylSet as input
ctrls <- getProbeInfo(RGsetEx,type="Control")
ctrls <- ctrls[ctrls$Address %in% featureNames(RGsetEx),]
ctrl_r <- getRed(RGsetEx)[ctrls$Address,]
ctrl_g <- getGreen(RGsetEx)[ctrls$Address,]
CG.controls <- ctrls$Type %in% c("NORM_C","NORM_G")
AT.controls <- ctrls$Type %in% c("NORM_A","NORM_T")
cg_grn <- ctrl_g[CG.controls,]
at_red <- ctrl_r[AT.controls,]
rownames(cg_grn) = ctrls$ExtendedType[CG.controls]
rownames(at_red) = ctrls$ExtendedType[AT.controls]
mdat <- preprocessRaw(RGsetEx)
mdat <- relic(mdat,at_red,cg_grn)

}

ENmix documentation built on April 2, 2021, 6 p.m.