diagplot.noiseq: Diagnostic plots based on the NOISeq package

In pmoulos/metaseqr: An R package for the analysis and result reporting of RNA-Seq data by combining multiple statistical algorithms.

Description

A wrapper around the plotting functions availale in the NOISeq Bioconductor package. For analytical explanation of each plot please see the vignette of the NOISeq package. It is best to use this function through the main plotting function diagplot.metaseqr.

Usage

    diagplot.noiseq(x, sample.list, covars,
        which.plot = c("biodetection", "countsbio", "saturation", "rnacomp", 
               "readnoise", "biodist"),
        output = "x11",
        biodist.opts = list(p = NULL, pcut = NULL, name = NULL),
        path = NULL, is.norm = FALSE, ...)

Arguments

x	the count data matrix.
sample.list	the list containing condition names and the samples under each condition.
covars	a list (whose annotation elements are ideally a subset of an annotation data frame produced by get.annotation) with the following members: data (the data matrix), length (gene length), gc (the gene gc_content), chromosome (a data frame with chromosome name and co-ordinates), factors (a factor with the experimental condition names replicated by the number of samples in each experimental condition) and biotype (each gene's biotype as depicted in Ensembl-like annotations).
which.plot	the NOISeq package plot to generate. It can be one or more of "biodetection", "countsbio", "saturation", "rnacomp", "readnoise" or "biodist". Please refer to the documentation of the EDASeq package for details on the use of these plots. The which.plot="saturation" case is modified to be more informative by producing two kinds of plots. See diagplot.noiseq.saturation.
biodist.opts	a list with the following members: p (a vector of p-values, e.g. the p-values of a contrast), pcut (a unique number depicting a p-value cutoff, required for the "biodist" case), name (a name for the "biodist" plot, e.g. the name of the contrast.
output	one or more R plotting device to direct the plot result to. Supported mechanisms: "x11" (default), "png", "jpg", "bmp", "pdf" or "ps".
path	the path to create output files.
is.norm	a logical indicating whether object contains raw or normalized data. It is not essential and it serves only plot annotation purposes.
...	further arguments to be passed to plot devices, such as parameter from par.

Value

The filenames of the plots produced in a named list with names the which.plot argument. If output="x11", no output filenames are produced.

Note

Please note that in case of "biodist" plots, the behavior of the function is unstable, mostly due to the very specific inputs this plotting function accepts in the NOISeq package. We have tried to predict unstable behavior and avoid exceptions through the use of tryCatch but it's still possible that you might run onto an error.

Author(s)

Panagiotis Moulos

Examples

require(DESeq)
data.matrix <- counts(makeExampleCountDataSet())
sample.list <- list(A=c("A1","A2"),B=c("B1","B2","B3"))
lengths <- round(1000*runif(nrow(data.matrix)))
starts <- round(1000*runif(nrow(data.matrix)))
ends <- starts + lengths
covars <- list(
    data=data.matrix,
    length=lengths,
    gc=runif(nrow(data.matrix)),
    chromosome=data.frame(
        chromosome=c(rep("chr1",nrow(data.matrix)/2),
            rep("chr2",nrow(data.matrix)/2)),
        start=starts,
        end=ends
    ),
    factors=data.frame(class=as.class.vector(sample.list)),
    biotype=c(rep("protein_coding",nrow(data.matrix)/2),rep("ncRNA",
            nrow(data.matrix)/2))
)
p <- runif(nrow(data.matrix))
diagplot.noiseq(data.matrix,sample.list,covars=covars,
        biodist.opts=list(p=p,pcut=0.1,name="A_vs_B"))

pmoulos/metaseqr documentation built on Dec. 29, 2020, 5:56 a.m.