plotFeatureClasses: Plot Density, Scatterplot, Parallel Plot or Bar Chart for...

In ClassifyR: A framework for cross-validated classification problems, with applications to differential variability and differential distribution testing

Description

Allows the visualisation of measurements in the data set. If targets is of type Pairs, then a parallel plot is automatically drawn. If it's a single categorical variable, then a bar chart is automatically drawn.

Usage

  ## S4 method for signature 'matrix'
plotFeatureClasses(measurements, classes, targets, ...)
  ## S4 method for signature 'DataFrame'
plotFeatureClasses(measurements, classes, targets, groupBy = NULL,
                        groupingName = NULL, whichNumericFeaturePlots = c("both", "density", "stripchart"),
                    measurementLimits = NULL, lineWidth = 1, dotBinWidth = 1,
                    xAxisLabel = NULL, yAxisLabels = c("Density", "Classes"),
                   showXtickLabels = TRUE, showYtickLabels = TRUE,
                   xLabelPositions = "auto", yLabelPositions = "auto",
                   fontSizes = c(24, 16, 12, 12, 12),
                   colours = c("#3F48CC", "#880015"), showDatasetName = TRUE, plot = TRUE)
  ## S4 method for signature 'MultiAssayExperiment'
plotFeatureClasses(measurements, targets, groupBy = NULL, groupingName = NULL,
                   showDatasetName = TRUE, ...)

Arguments

measurements	A matrix, DataFrame or a MultiAssayExperiment object containing the data. For a matrix, the rows are for features and the columns are for samples. A column with name "class" must be present in the DataFrame stored in the colData slot.
classes	Either a vector of class labels of class factor or if the measurements are of class DataFrame a character vector of length 1 containing the column name in measurement is also permitted. Not used if measurements is a MultiAssayExperiment object.
targets	If measurements is a matrix or DataFrame, then a vector of numeric or character indicies or the feature identifiers corresponding to the feature(s) to be plotted. If measurements is a MultiAssayExperiment, then a DataFrame of 2 columns must be specified. The first column contains the names of the tables and the second contains the names of the variables, thus each row unambiguously specifies a variable to be plotted.
groupBy	If measurements is a DataFrame, then a character vector of length 1, which contains the name of a categorical feature, may be specified. If measurements is a MultiAssayExperiment, then a character vector of length 2, which contains the name of a data table as the first element and the name of a categorical feature as the second element, may be specified. Additionally, the value "clinical" may be used to refer to the column annotation stored in the colData slot of the of the MultiAssayExperiment object. A density plot will have additional lines of different line types for each category. A strip chart plot will have a separate strip chart created for each category and the charts will be drawn in a single column on the graphics device. A parallel plot and bar chart plot will similarly be laid out.
groupingName	A label for the grouping variable to be used in plots.
...	Unused variables by the three top-level methods passed to the internal method which generates the plot(s).
whichNumericFeaturePlots	If the feature is a single feature and has numeric measurements, this option specifies which types of plot(s) to draw. The default value is "both", which draws a density plot and also a stip chart below the density plot. Other options are "density" for drawing only a density plot and "stripchart" for drawing only a strip chart.
measurementLimits	The minimum and maximum expression values to plot. Default: NULL. By default, the limits are automatically computed from the data values.
lineWidth	Numeric value that alters the line thickness for density plots. Default: 1.
dotBinWidth	Numeric value that alters the diameter of dots in the strip chart. Default: 1.
xAxisLabel	The axis label for the plot's horizontal axis. Default: NULL.
yAxisLabels	A character vector of length 1 or 2. If the feature's measurements are numeric an whichNumericFeaturePlots has the value "both", the first value is the y-axis label for the density plot and the second value is the y-axis label for the strip chart. Otherwise, if the feature's measurements are numeric and only one plot is drawn, then a character vector of length 1 specifies the y-axis label for that particular plot. Ignored if the feature's measurements are categorical.
showXtickLabels	Logical. Default: TRUE. If set to FALSE, the x-axis labels are hidden.
showYtickLabels	Logical. Default: TRUE. If set to FALSE, the y-axis labels are hidden.
xLabelPositions	Either "auto" or a vector of values. The positions of labels on the x-axis. If "auto", the placement of labels is automatically calculated.
yLabelPositions	Either "auto" or a vector of values. The positions of labels on the y-axis. If "auto", the placement of labels is automatically calculated.
fontSizes	A vector of length 5. The first number is the size of the title. The second number is the size of the axes titles. The third number is the size of the axes values. The fourth number is the size of the legends' titles. The fifth number is the font size of the legend labels.
colours	The colours to plot data of each class in. The length of this vector must be as long as the distinct number of classes in the data set.
showDatasetName	Logical. Default: TRUE. If TRUE and the data is in a MultiAssayExperiment object, the the name of the table in which the feature is stored in is added to the plot title.
plot	Logical. Default: TRUE. If TRUE, a plot is produced on the current graphics device.

Value

Plots are created on the current graphics device and a list of plot objects is invisibly returned. The classes of the plot object are determined based on the type of data plotted and the number of plots per feature generated. If the plotted variable is discrete or if the variable is numeric and one plot type was specified, the list element is an object of class ggplot. Otherwise, if the variable is numeric and both the density and stripchart plot types were made, the list element is an object of class TableGrob.

Settling lineWidth and dotBinWidth to the same value doesn't result in the density plot and the strip chart having elements of the same size. Some manual experimentation is required to get similarly sized plot elements.

Author(s)

Dario Strbenac

Examples

  # First 25 samples and first 5 genes are mixtures of two normals. Last 25 samples are
  # one normal.
  genesMatrix <- sapply(1:15, function(geneColumn) c(rnorm(5, 5, 1)))
  genesMatrix <- cbind(genesMatrix, sapply(1:10, function(geneColumn) c(rnorm(5, 15, 1))))
  genesMatrix <- cbind(genesMatrix, sapply(1:25, function(geneColumn) c(rnorm(5, 9, 2))))
  genesMatrix <- rbind(genesMatrix, sapply(1:50, function(geneColumn) rnorm(95, 9, 3)))
  rownames(genesMatrix) <- paste("Gene", 1:100)
  colnames(genesMatrix) <- paste("Sample", 1:50)
  classes <- factor(rep(c("Poor", "Good"), each = 25), levels = c("Good", "Poor"))
  plotFeatureClasses(genesMatrix, classes, targets = "Gene 4",
                     xAxisLabel = bquote(log[2]*'(expression)'), dotBinWidth = 0.5)
                     
                     
  
  infectionResults <- c(rep(c("No", "Yes"), c(20, 5)), rep(c("No", "Yes"), c(5, 20)))
  genders <- factor(rep(c("Male", "Female"), each = 10, length.out = 50))
  clinicalData <- DataFrame(Gender = genders, Sugar = runif(50, 4, 10),
                              Infection = factor(infectionResults, levels = c("No", "Yes")),
                            row.names = colnames(genesMatrix))
  plotFeatureClasses(clinicalData, classes, targets = "Infection")
  plotFeatureClasses(clinicalData, classes, targets = "Infection", groupBy = "Gender")
  
  dataContainer <- MultiAssayExperiment(list(RNA = genesMatrix),
                                        colData = cbind(clinicalData, class = classes))
  targetFeatures <- DataFrame(table = "RNA", feature = "Gene 50")                                     
  plotFeatureClasses(dataContainer, targets = targetFeatures,
                     groupBy = c("clinical", "Gender"),
                     xAxisLabel = bquote(log[2]*'(expression)'))

ClassifyR documentation built on Nov. 8, 2020, 6:53 p.m.