SubgrPlots (version 0.1.0)

plot_overlap_alternative: a graphical display for showing relative proportions of pairwise subgroup overlap with two unidirectional arrowed lines around two circle

Description

this function produces a plot for displaying relative proportions of pairwise subgroup overlap, where the relative proportion is defined by |intersect(A, B)|/|A|, for any sets A and B and A is the baseline set. The letters represent subgroups defined by the categories of the selected covariates. Subgroup letters located around the two circles are considered as baselines; each has unidirectional arrowed lines to the other. In addition, there are two modes of display - one is that arrowed lines have different colours, reflecting the level of overlap proportions; the other is that lines have five thickness and types, indicating five levels of overlap proportions.

Usage

plot_overlap_alternative(dat, covari.sel, para = c(0.5, 0.15, 1), mode,
  font.size = c(1.5, 1.5, 0.8), title = NULL)

Arguments

dat

a data set

covari.sel

a vector of indices of covariates

para

a vector with three elements specifying the parameters of plot display; the first element is for specifying the length of the arrowhead; the second is for specifying the width of the arrowheads; the third is for specifying the adjustment of the arrowhead (there are three possible values, 0, 0.5 and 1)

mode

a value specifying the type of display; either 1 or 2.

font.size

a vector specifying the size of labels and text; the first element is for the title; the second is for the covariates labels.

title

a string specifying the main title.

Examples

Run this code
# NOT RUN {
data(prca)
dat <- prca
## 1. Overlap alternative plot -------------------------------------------------
plot_overlap_alternative(dat = dat,
                         covari.sel = c(6,5,4,7),
                         mode = 1,
                         para = c(0, 0.6, 1),
                         font.size = c(1.2, 1.2, 0.8),
                         title = NULL)

# }

Run the code above in your browser using DataLab