ErrorRateBoxPlot: Function for boxplot the error rate

In elqumsan/RNAseqMVA: Application of Multivariate Analysis methods to RNA-seq Data

Description

this function takes the experiment List from the error rate that are collected from the experiment.

Usage

ErrorRateBoxPlot(
  experimentList,
  classifier,
  expMisclassificationRate = NULL,
  horizontal = TRUE,
  fig.height = 10,
  fig.width = 4 + 0.05 * length(experimentList),
  cex.axis = 0.8,
  experimentLabels = names(experimentList),
  main = paste0(parameters$short_label, " (", parameters$recountID, ") ",
    parameters$feature, "s", "\n", classifier, "; ", parameters$iterations,
    " iterations"),
  boxplotFile = NULL,
  medianLine = FALSE,
  expColor = "#00BBFF",
  permColor = "grey",
  expLegend = "Train/test",
  permLegend = "Permuted labels",
  legend.place = "topright",
  ....
)

Arguments

experimentList	this is list of experiment where each cell of list error rate of the single experiment with special parameters
classifier	name of the classifier (e.g. svm, knn, rf), used in the default main title
expMisclassificationRate	user-specified value for the expected misclassification rate
main	main title for the boxplot
expColor	is colore each box plot that represent the error rate for each experiment with actual class lables.
permColor	is the coler for the box plot that represent error rate for each experiment with permuted calss lables.
expLegend	is the legend for the real class label train/test experiment.
permLegend	legend for the IterateTrainingTesting with permuted class lables.
legend.place	location for the legend, which is passed to legend()
medianLine=FALSE	if TRUE, draw a line joining the medians

Author(s)

Mustafa AbuElqumsan and Jacques van Helden

Examples

compareExperiments(experimentList = train.test.results)

elqumsan/RNAseqMVA documentation built on March 10, 2021, 8:10 a.m.