Utility Functions to Plot and Work with COMPASS Output • UtilsCompassSV

The goal of UtilsCompassSV is to supply useful plotting and utility functions for working with output from COMPASS.

Installation

You can install the latest version of UtilsCompassSV from GitHub with

if (!require("remotes", quietly = TRUE)) install.packages("remotes")
remotes::install_github("SATVILab/UtilsCompassSV")

Example

Plots

Create boxplots of posterior probabilities for individual cytokine combinations, as well as the PFS and FS responses. These are more concise than standard COMPASS heatmaps and automatically force the same cytokine combinations to be present and cytokines in the same order for all “groups” plotted together. Note that the PFS and FS scores plot need not be added.

library(UtilsCompassSV)
data('c_obj_list', package = 'UtilsCompassSV')
plot_compass(
  c_obj = c_obj_list,
  dir_save = here::here('data-raw'),
  type = c('pp', 'scores'),
  return_plot_list = FALSE,
  shift_plot_scores = c(-0.05, 0.05),
  shift_plot_pp_y = -0.075,
  shift_plot_heatmap_x = 0.052
  )
knitr::include_graphics('data-raw/compass_boxplots_grid.png')

Colours of boxplots and tiles in the cytokine grid plot can be customized.

plot_compass(
  c_obj = c_obj_list,
  dir_save = here::here('data-raw'),
  file_grid = "compass_boxplots_grid_custom_colours",
  type = c('pp', 'scores'),
  return_plot_list = FALSE,
  shift_plot_scores = c(-0.05, 0.05),
  shift_plot_pp_y = -0.075,
  shift_plot_heatmap_x = 0.034,
  plot_prob_fill = c("skyblue", "forestgreen", "mediumvioletred", "darkorange"),
  tile_fill = RColorBrewer::brewer.pal(n = 11, name = "Spectral")[2:4] |> rev()
)
knitr::include_graphics('data-raw/compass_boxplots_grid_custom_colours.png')

Utilities

Convert cytokine combination formats between standard “+/-” format and COMPASS “!&” format. Can also label cytokines/markers.

cyt_combn_vec_compass <- c("IFNg&!IL2&TNF&IL6&!IL22",
                           "IFNg&IL2&TNF&IL6&IL22",
                           "!IFNg&!IL2&!TNF&!IL6&!IL22")
cyt_combn_vec_std <- UtilsCompassSV::convert_cyt_combn_format(cyt_combn_vec_compass,
                                                            to = 'std')
pander::pandoc.table(tibble::tibble(
  `COMPASS format` = cyt_combn_vec_compass,
  `Standard format` = cyt_combn_vec_std
  ))

COMPASS format	Standard format
IFNg&!IL2&TNF&IL6&!IL22	IFNg+IL2-TNF+IL6+IL22-
IFNg&IL2&TNF&IL6&IL22	IFNg+IL2+TNF+IL6+IL22+
!IFNg&!IL2&!TNF&!IL6&!IL22	IFNg-IL2-TNF-IL6-IL22-

Calculate the probability of an individual responding to at least one cytokine combination, assuming independence of the probability estimates between cytokine combinations.

response_prob(c_obj = c_obj_list[[4]]) %>%
  head() %>%
  pander::pandoc.table()

sampleid	prob
010673_D0	0.076
010782_D0	0.4405
010782_D720	0.903
010978_D0	1
010993_D0	0.8039
020185_D0	0