analysis.bgx: Analyse BGX output.
In bgx: Bayesian Gene eXpression

Description Usage Arguments Details Value Author(s) See Also

Functions for plotting expression densities, differential expression densities, histogram of proportion of differentially expressed genes, etc.

  plotExpressionDensity(bgxOutput, gene=NULL, normalize=c("none","mean","loess"),...)
  plotDEDensity(bgxOutput, gene=NULL, conditions=c(1,2), normalize=c("none","mean","loess"), normgenes=c(1:length(bgxOutput[["geneNames"]])), ...)
  plotDEHistogram(bgxOutput, conditions=c(1,2), normalize=c("none","mean","loess"), normgenes=c(1:length(bgxOutput[["geneNames"]])), df=floor(1.8 * log10(length(bgxOutput[["geneNames"]]))))
  rankByDE(bgxOutput, conditions=c(1,2),normalize=c("none", "mean", "loess"), normgenes=c(1:length(bgxOutput[["geneNames"]])), absolute=TRUE)
  plotDiffRank(bgxOutput, conditions=c(1,2),normalize=c("none", "mean", "loess"), normgenes=c(1:length(bgxOutput[["geneNames"]])), ymax=NULL)

`bgxOutput`	A list obtained from running `readOutput.bgx` on a BGX output directory.
`gene`	Which gene to analyse. This can either be an index or a name.
`conditions`	Indices of conditions to compare.
`normalize`	"none": do not normalise posterior distributions of mu. "mean": normalise by scaling posterior distributions of mu for conditions > 1 to have the same mean as the posterior distribution of mu for condition 1. "loess": same as "mean" but use loess normalisation.
`normgenes`	Which genes to use for loess normalisation. By default, use all genes.
`df`	Residual degrees of freedom. Decrease to 6 if the histogram fit goes haywire.
`absolute`	Rank genes by absolute differential expression.
`ymax`	Specify upper limit of y axis.
`...`	Parameters to pass to density function (where applicable).

plotExpressionDensity plots gene expression distributions under various conditions for the specified gene.

plotDEDensity plots the differential expression distribution between two conditions for a given gene.

plotDEHistogram plots a histogram of differential expression between two conditions and estimates the number of up and down regulated differentially expressed genes.

rankByDE ranks genes by differential expression and returns ordering and corresponding DE values in a matrix.

plotDiffRank plots 2.5-97.5% confidence intervals for ranked differential expression estimates.

None, except plotDERank, which returns a matrix of genes ranked by differential expression.

Ernest Turro

bgx, standalone.bgx, readOutput.bgx, plotExpressionDensity, plotDEDensity, plotDEHistogram