The script builds nodes and links with tfl_sankey_data() and draws them with
tfl_plot_sankey().
Usage
tfl_fig_sankey(
adam = NULL,
style = c("grey_links", "colored_links", "subgroups"),
data = "ADLOT",
id = "USUBJID",
stage = "LINE",
category = "TRT",
by = NULL,
pop = "FASFL",
palette = "treatment",
title = NULL,
width = 9,
height = 5,
dpi = 300,
units = "in",
file = NULL,
plot_id = "sankey"
)Arguments
- adam
Optional ADaM data (
tfl_read_adam()); values found in the data (groups, colours) are written literally into the code.- style
grey_links(links in light grey),colored_links(links in the colour of their source node) orsubgroups(one panel per value ofby, on a shared scale).- data
Dataset with one row per subject and line.
- id, stage, category
Subject, line and category (e.g. treatment class) variables.
- by
Subgroup variable for
style = "subgroups".- pop
Population flag (
== "Y");NULLfor none.- palette
Palette preset for the categories.
- title
Figure title.
- width, height, dpi, units
Figure size.
- file
Write the script to this file.
- plot_id
Used in the output file name.