ClassifyR (version 1.2.4)

rankPlot: Plot Pair-wise Overlap of Ranked Features

Description

The average pair-wise overlap is computed for every pair of cross-validations. The overlap is converted to a percentage and plotted as lineplots.

Usage

"rankPlot"(results, topRanked = seq(10, 100, 10), comparison = c("within", "classificationName", "validation", "datasetName"), lineColourVariable = c("validation", "datasetName", "classificationName", "None"), lineColours = NULL, lineWidth = 1, pointTypeVariable = c("datasetName", "classificationName", "validation", "None"), pointSize = 2, legendLinesPointsSize = 1, rowVariable = c("None", "datasetName", "classificationName", "validation"), columnVariable = c("classificationName", "datasetName", "validation", "None"), yMax = 100, fontSizes = c(24, 16, 12, 12, 12, 16), title = "Feature Ranking Stability", xLabelPositions = seq(10, 100, 10), yLabel = "Average Pairwise Common Features (%)", plot = TRUE, parallelParams = bpparam())

Arguments

results

A list of ClassifyResult objects.

topRanked

A sequence of thresholds of number of the best features to use for overlapping.

comparison

The aspect of the experimental design to compare. See Details section for a detailed description.

lineColourVariable

The slot name that different levels of are plotted as different line colours.

lineColours

A vector of colours for different levels of the line colouring parameter. If NULL, a default palette is used.

lineWidth

A single number controlling the thickness of lines drawn.

pointTypeVariable

The slot name that different levels of are plotted as different point shapes on the lines.

pointSize

A single number diameter of points drawn.

legendLinesPointsSize

A single number specifying the size of the lines and points in the legend, if a legend is drawn.

rowVariable

The slot name that different levels of are plotted as separate rows of lineplots.

columnVariable

The slot name that different levels of are plotted as separate columns of lineplots.

yMax

The maximum value of the percentage to plot.

fontSizes

A vector of length 6. 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. The sixth number is the font size of the titles of grouped plots, if any are produced. In other words, when rowVariable or columnVariable are not NULL.

title

An overall title for the plot.

xLabelPositions

Locations where to put labels on the x-axis.

yLabel

Label to be used for the y-axis of overlap percentages.

plot

Logical. IF TRUE, a plot is produced on the current graphics device.

parallelParams

An object of class MulticoreParam or SnowParam.

Value

An object of class ggplot and a plot on the current graphics device, if plot is TRUE.

Details

Possible values for slot names are "datasetName", "classificationName", and "validation". If "None", then that graphic element is not used. If comparison is "within", then the feature rankings are compared within a particular analysis. The result will inform how stable the feature rankings are between different iterations of a particular analysis. If comparison is "classificationName", then the feature rankings are compared across different classification algorithm types, for each level of "datasetName" and "validation". The result will inform how stable the feature rankings are between different classification algorithms, for every cross-validation scheme and every dataset. If comparison is "validation", then the feature rankings are compared across different cross-validation schemes, for each level of "classificationName" and "datasetName". The result will inform how stable the feature rankings are between different cross-validation schemes, for every classification algorithm and every dataset. If comparison is "datasetName", then the feature rankings are compared across different datasets, for each level of "classificationName" and "validation". The result will inform how stable the feature rankings are between different datasets, for every classification algorithm and every dataset. This could be used to consider if different studies have a highly overlapping feature ranking pattern. Calculating all pair-wise set overlaps can be time-consuming. This stage can be done on multiple CPUs by providing the relevant options to parallelParams.

Examples

  predicted <- data.frame(sample = sample(10, 100, replace = TRUE),
                          label = rep(c("Healthy", "Cancer"), each = 50))
  actual <- factor(rep(c("Healthy", "Cancer"), each = 5))
  result1 <- ClassifyResult("Example", "Differential Expression", LETTERS[1:10], LETTERS[10:1], list(1:100, c(1:9, 11:101)), list(sample(10, 10), sample(10, 10)),
                            list(predicted), actual, list("fold", 100, 5))
  predicted[, "label"] <- sample(predicted[, "label"])
  result2 <- ClassifyResult("Example", "Differential Variability", LETTERS[1:10], LETTERS[10:1], list(1:100, c(1:5, 11:105)), list(sample(10, 10), sample(10, 10)),
                            list(predicted), actual, validation = list("leave", 1))
  # rankPlot(list(result1, result2), pointTypeVariable = "classificationName") # Wait for namespace problems to be fixed.