binaryclass_scatterplot: Plot differences in f1-score or entropy of individual genes

In PatZeis/AssessMe: A cluster assessment tool for preprocessing and clustering optimization

Plot differences in f1-score or entropy of individual genes

Description

This function serves to explore differences in f1-score or entropy of individual genes between different assessed cluster partitions. Genes with large differences between the two assessments are highlighted.

Usage

binaryclass_scatterplot(
  assessment1,
  assessment2,
  maplot = F,
  toplot = "F1",
  signifi = F,
  out_doub_dif = NULL,
  xlabs = NULL,
  ylabs = NULL,
  logmean = NULL,
  size = NULL,
  label.rectangle = F
)

Arguments

assessment1	assessment of a cluster partition for which either f1_score or Entropy computation has been performed based on a derived per gene cutoff.
assessment2	assessment of a second partition derived for the same data.
maplot	logical. If T, plot the F1 or Entropy per gene difference of assessment 1 and assessment 2 against the mean expression of the gene.
toplot	plot either “F1” or “Entropy”. Differences of F1 or Entropy calculations of genes between the assessed cluster partitions. Default = "F1".
signifi	logical. if T, calculate standard deviation of absolute differences between F1/Entropy gene-pairs of two assessments and multiplies this standard deviation by 1.64485361, representing the z-score with significance level based on the normal distribution function. The expected difference of a pair equals 0. Genes with 1.64485361 standard deviations(sds) difference between the two assessments are considered significant and are highlighted. Default = F and all genes with differences of one standard deviation are highlighted.
out_doub_dif	vector of gene names of user specified gene sets to highlight. Default = NULL and genes with differences larger than 1 sd or 1.64485361 sds are highlighted.
xlabs	character string for x-axis label indicating assessment 1. Default = NULL and label is derived from object name of assessment 1.
ylabs	character string for y-axis label indicating assessment 2. Default = NULL and label is derived from object name of assessment 2.
logmean	either “log2”, “log10” or NULL. If “log2” or “log10” performs log transformation of the mean expression per gene. Default = NULL.
size	numeric. Size of the label. Default = NULL and size is set to 2.
label.rectangle	logical. If T, text of labels is highlighted with a rectangle. Default = F.

Examples

binaryclass_scatterplot(assessmentRC$Sres.1, assessmentRC$Sres.5, maplot = T, logmean = "log2", signifi = T)

PatZeis/AssessMe documentation built on Nov. 19, 2022, 6:03 a.m.