dyebias.boxplot: Creates boxplots of the reporters with the strongest dye bias

In dyebias: The GASSCO method for correcting for slide-dependent gene-specific dye bias

Description

The aim of this routine is to show the magnitude of the dye bias across the data set, as well as the extent to which the GASSCO method could get rid of it. Typically, two boxplots would be shown, one before, one after dye bias correction. For esthetic reasons, the boxplots are usually ordered by the overal slide bias of the uncorrected data set. See also Margaritis et al. (2009), Fig. 1 and 3.

Usage

dyebias.boxplot(data, iGSDBs, dyebias.percentile=5,
                application.subset=TRUE, order, output=NULL,
                ylim=c(-4,4), ...)

Arguments

data	The marrayNorm object to boxplot.
iGSDBs	A data frame with intrinsic gene-specific dye biases, the same as that used in dyebias.apply.correction, probably returned by dyebias.estimate.iGSDBs; see there for documentation.
dyebias.percentile	The percentile of intrinsic gene specific dye biases (iGSDBs) for which to highlight the reporters.
application.subset	The set of reporters that was eligible for dye bias correction; same argument as for dyebias.apply.correction.
order	If order==FALSE, no ordering of slides prior to boxplotting takes place. If order==NULL, the slides are sorted by increasing slide bias prior to boxplotting. This is typically done for data that is not yet dye bias corrected. This order is also returned as a value. If an order!=NULL, the slides are put this order before boxplotting. This is typically done for a dye bias-corrected data set, using the order of the uncorrected set. (See also Fig. 3 in the paper).
output	Specifies the output. If NULL, the existing output device is used; if output is one of "X11", "windows", "quartz", a new X11 (Unix)/windows (Windows)/quartz (Mac) device is created. If output is a string ending in one of ".pdf", ".png", ".eps", ".ps" is given, a file of that name and type is created and closed afterwards.
ylim	As for boxplot()
...	Other arguments (such as main, etc.) are passed on to boxplot().

Value

The order obtained, for use in a later call to this same function.

Author(s)

Philip Lijnzaad [e-mail]

References

Margaritis, T., Lijnzaad, P., van Leenen, D., Bouwmeester, D., Kemmeren, P., van Hooff, S.R and Holstege, F.C.P. (2009) Adaptable gene-specific dye bias correction for two-channel DNA microarrays. Molecular Systems Biology, 5:266, 2009. doi: 10.1038/msb.2009.21.

See Also

dyebias.estimate.iGSDBs, dyebias.apply.correction, dyebias.rgplot, dyebias.maplot, dyebias.trendplot

Examples

                                       

  ylim <- c(-1, 1)

  layout(matrix(1:2, nrow=1,ncol=2))

  order <- dyebias.boxplot(data=data.norm, 
                        iGSDBs=iGSDBs.estimated,  # from e.g. dyebias.estimate.iGSDBs
                        order=NULL,               # i.e., order by increasing slide bias
                        output=NULL,
                        main="before correction",
                        ylim=ylim)

  order <- dyebias.boxplot(data=correction$data.corrected, # from dyebias.apply.correction
                        iGSDBs=iGSDBs.estimated,
                        order=order,              # order by the original slide bias
                        output=NULL,
                        main="after correction",
                        ylim=ylim
                        )

dyebias documentation built on Nov. 8, 2020, 5:53 p.m.