ENmix (version 1.8.0)

rcp: Regression on Correlated Probes(RCP)

Description

Probe design type bias correction using Regression on Correlated Probes (RCP) method

Usage

rcp(mdat, dist=25, quantile.grid=seq(0.001,0.999,by=0.001), qcscore = NULL, nbthre=3, detPthre=0.000001)

Arguments

mdat

An object of class MethylSet.

dist

Maximum distance in base pair between type I and type II probe pairs for regression calibration

quantile.grid

Quantile grid used in linear regression

qcscore

If the data quality infomation (the output from function QCinfo) is provied, low quality data points as defined by detection p value threshold (detPthre=0.000001) or number of bead threshold (nbthre=3) will be set to missing.

detPthre

Detection P value threshold to define low qualitye data points, detPthre=0.000001 in default.

nbthre

Number of beads threshold to define low qualitye data points, nbthre=3 in default.

Value

A beta value matrix

Details

The function will first identify type I and type II probe pairs within specified distance, and then perform linear regression between the probe types to estimate regression coefficients. With the estimates the function will then calibrates type II data using type I data as references.

References

Liang Niu, Zongli Xu and Jack A. Taylor RCP: a novel probe design bias correction method for Illumina Methylation BeadChip, Bioinformatics 2016

Examples

if(FALSE){
if (require(minfiData)) {
mdat=preprocessENmix(RGsetEx,bgParaEst="oob",nCores=6)
mdatq1=normalize.quantile(mdat,method="quantile1")
beta=rcp(mdatq1)
}}