ENmix (version 1.8.0)

QCfilter: Sample or CpG probe filter.

Description

Filter low quality samples or CpGs, outlier samples or user specified samples or CpGs.

Usage

QCfilter(mdat,qcinfo=NULL,detPthre=0.000001,nbthre=3,samplethre=0.05,CpGthre=0.05, bisulthre=NULL,outlier=FALSE,outid=NULL, outCpG=NULL,plot=FALSE)

Arguments

mdat

An object of class MethylSet or beta value matrix.

qcinfo

An object outputed from function QCinfo

detPthre

Detection P value threshold to identify low quality data point

nbthre

Number of bead threshold to identify low quality data point

samplethre

Threshold to identify low quality samples, the percentage of low quality methylation data points across probes for each sample

CpGthre

Threshold to identify low quality probes, percentage of low quality methylation data points across samples for each probe

bisulthre

Threshold of bisulfite intensity for identification of low quality samples. By default, Mean - 3 x SD of sample bisufite control intensities will be used as the threshold.

outlier

If TRUE, outlier samples will be excluded.

outid

A list of user specified samples to be excluded.

outCpG

A list of user specified CpGs to be excluded.

plot

TRUE or FALSE, whether to produce quality checking plots.

Value

An same type object as input object after excluding low quality samples and CpGsFigure "qc_sample.jpg": scatter plot for Percent of low quality data per sample and Average bisulfite conversion intensityFigure "qc_CpG.jpg": histogram for Percent of low quality data per CpG.Figure "freqpolygon_beta_beforeQC.jpg": distribution plot before filtering.Figure "freqpolygon_beta_afterQC.jpg": distribution plot after filtering.

References

Zongli Xu, Liang Niu, Leping Li and Jack A. Taylor, ENmix: a novel background correction method for Illumina HumanMethylation450 BeadChip. Nucleic Acids Research 2015.

Examples

if(FALSE){
if (require(minfiData)) {
sheet <- read.450k.sheet(file.path(find.package("minfiData"),"extdata"), pattern = "csv$")
rgSet <- read.450k.exp(targets = sheet,extended = TRUE)
qcscore<-QCinfo(rgSet)
rgSet=QCfilter(rgSet,qcinfo=qcscore,outlier=TRUE)
}}