expression_plot_symbol: Plots the expression profile of a gene by levels of a factor
In kevinrue/GOexpress: Visualise microarray and RNAseq data using gene ontology annotations

Description Usage Arguments Value Warning Author(s) References See Also Examples

This function will plot the expression profile of a gene across a valid X-axis variable from the AnnotatedDataFrame while representing the mean and confidence interval of groups of samples defined by levels of a valid grouping factor from the AnnotatedDataFrame.

In the case of a gene name represented by multiple gene or probeset identifiers in the dataset, a lattice of plots will be produced. Each of the plots in this lattice can subsequently be plotted separately using its associated index.

expression_plot_symbol(
    gene_symbol, result, eSet, x_var, f=result$factor, subset=NULL,
    index=0, xlab=x_var, ylab="log2(cpm)", ylim=range(exprs(eSet)), 
    col.palette="Accent",
    col=brewer.pal(n=length(levels(pData(eSet)[,f])), name=col.palette),
    level=0.95, titles=c(), title.size=2, axis.title.size=20,
    axis.text.size=15, axis.text.angle=0,
    legend.title.size=20, legend.text.size=20, legend.key.size=30)

`gene_symbol`	A gene name present in `rownames(AlvMac_results$genes)`.
`result`	An output of the `GO_analyse()` or `subset_scores()` function.
`eSet`	`ExpressionSet` of the `Biobase` package including a gene-by-sample expression matrix in the `AssayData` slot, and a phenotypic information data-frame in the `phenodate` slot. In the expression matrix, row names are Ensembl gene identifiers or probeset identifiers, and column names are sample identifiers. In the phentypic data-frame, row names are sample idenfifiers, column names are grouping factors and phenotypic traits usable for the one-way ANOVA.
`x_var`	A column name in `phenodata` to plot on the X-axis. If representing time on the X-axis, users should store the time-points as numeric values in the `AnnotatedDataFrame` to adequately space the time-points.
`f`	A column name in `phenodata` to group the samples when representing mean and confidence interval. The factor specified in the initial `GO_analyse()` call is used by default. Unexpected grouping factors of samples can reveal interesting trends (e.g. "Animal", "Tissue", "CellType", ...).
`subset`	A named list to subset `eSet`. Names must be column names existing in colnames(pData(eSet)). Values must be vectors of values existing in the corresponding column of pData(eSet).
`index`	In the case where multiple gene or probeset identifiers are associated with the gene name, `index=2` will plot the expression profile of the second feature identifier alone, for instance.
`xlab`	Title of the X-axis. Default is tha value of `x_var`.
`ylab`	Title of the Y-axis. Default is "log2(cpm)".
`ylim`	Numeric vector of length 2 specifying the lower and upper bounds of the Y axis. Default is scaled to the full range of expression values in the expression dataset, to ease comparison of different genes. If set to NULL, the axis will be scaled to fit the plotted data only.
`col.palette`	A valid `RColorBrewer` palette name to fetch the colormap from.
`col`	A vector of color names or codes. The number of colors provided must match the number of levels of the grouping factor. Default to a palette with an adequate set of colors. If specified, overrides argument `col.palette`.
`level`	The confidence interval level to visualise around the mean of each group. Default is 0.95.
`titles`	Character vector providing as many titles as there are plots to replace the default titles. Default is a combination of the gene id and the associated gene.
`title.size`	Changes the font size of the title. Default is 2.
`axis.title.size`	Changes the font size of the axes title. Defalt is 20.
`axis.text.size`	Changes the font size of the axes text labels. Default is 15.
`axis.text.angle`	Changes the angle of the X axis text labels. Default is 0 (horizontal).
`legend.title.size`	Changes the font size of the legend title. Default is 20.
`legend.text.size`	Changes the font size of the legend text labels. Default is 15.
`legend.key.size`	Changes the size of the legend keys (in points). Default is 30.

The ggplot object, or the vector of feature identifiers if multiple features are annotated to the gene symbol.

Common issues:

It may not be possible to produce plots where the combination of X-axis variable and grouping factor leaves too few replicates to compute a confidence interval for each X value. This is a limitation imposed by the ggplot2 package to produce proper statistics and confidence intervals. In such cases, it may be preferrable to use the expression_profiles_symbol() method.

Kevin Rue-Albrecht

ggplot2 package.

Package Biobase, methods expression_plot, GO_analyse and ggplot.

# load the sample output data
data(AlvMac_results)

# Expression by gene identifier (TNIP3)
expression_plot_symbol(
    gene_symbol="PIK3AP1",
    result=AlvMac_results, eSet=AlvMac, x_var="Timepoint"
    )

# Same gene, plotted by animal and grouped by treatment (merging time points)
expression_plot_symbol(
    gene_symbol="PIK3AP1",
    result=AlvMac_results, eSet=AlvMac, x_var="Animal",
    f="Treatment"
    )

# Same gene, plotted by animal and grouped by time-point (merging treatments)
expression_plot_symbol(
    gene_symbol="PIK3AP1",
    result=AlvMac_results, eSet=AlvMac, x_var="Animal",
    f="Time")