intensityOutliersPlot: Plot mean intensity and highlight outliers
In GWASTools: Tools for Genome Wide Association Studies

Description Usage Arguments Details Author(s) See Also Examples

intensityOutliersPlot is a function to plot mean intensity for chromosome i vs mean of intensities for autosomes (excluding i) and highlight outliers

1 2	intensityOutliersPlot(mean.intensities, sex, outliers, sep = FALSE, label, ...)

`mean.intensities`	scan x chromosome matrix of mean intensities
`sex`	vector with values of "M" or "F" corresponding to scans in the rows of `mean.intensities`
`outliers`	list of outliers, each member corresponds to a chromosome (member "X" is itself a list of female and male outliers)
`sep`	plot outliers within a chromosome separately (TRUE) or together (FALSE)
`label`	list of plot labels (to be positioned below X axis) corresponding to list of outliers
`...`	additional arguments to `plot`

Outliers must be determined in advance and stored as a list, with one element per chromosome. The X chromosome must be a list of two elements, "M" and "F". Each element should contain a vector of ids corresponding to the row names of mean.intensities.

If sep=TRUE, labels must also be specified. labels should be a list that corresponds exactly to the elements of outliers.

Cathy Laurie

meanIntensityByScanChrom

# calculate mean intensity
library(GWASdata)
file <- system.file("extdata", "illumina_qxy.gds", package="GWASdata")
gds <- GdsIntensityReader(file)
data(illuminaScanADF)
intenData <- IntensityData(gds, scanAnnot=illuminaScanADF)
meanInten <- meanIntensityByScanChrom(intenData)
intenMatrix <- meanInten$mean.intensity

# find outliers
outliers <- list()
sex <- illuminaScanADF$sex
id <- illuminaScanADF$scanID
allequal(id, rownames(intenMatrix))
for (i in colnames(intenMatrix)) {
  if (i != "X") {
    imean <- intenMatrix[,i]
    imin <- id[imean == min(imean)]
    imax <- id[imean == max(imean)]
    outliers[[i]] <- c(imin, imax)
  } else {
    idf <- id[sex == "F"]
    fmean <- intenMatrix[sex == "F", i]
    fmin <- idf[fmean == min(fmean)]
    fmax <- idf[fmean == max(fmean)]
    outliers[[i]][["F"]] <- c(fmin, fmax)
    idm <- id[sex == "M"]
    mmean <- intenMatrix[sex == "M", i]
    mmin <- idm[mmean == min(mmean)]
    mmax <- idm[mmean == max(mmean)]
    outliers[[i]][["M"]] <- c(mmin, mmax)
  }
}

par(mfrow=c(2,4))
intensityOutliersPlot(intenMatrix, sex, outliers)
close(intenData)