Generating Funky Heatmaps for Data Frames.

{funkyheatmap}: Visualising data frames with mixed data types

{funkyheatmap} allows generating heatmap-like visualisations for benchmark data frames. Funky heatmaps can be fine-tuned by providing annotations of the columns and rows, which allows assigning multiple palettes or geometries or grouping rows and columns together in categories.

Installation

You can install funkyheatmap like so:

install.packages("funkyheatmap")

You can also download {funkyheatmap} as a standalone executable or a Nextflow pipeline.

Example

Let’s use the mtcars dataset as an example of what a funky heatmap looks like.

library(funkyheatmap)
library(dplyr, warn.conflicts = FALSE)
library(tibble, warn.conflicts = FALSE)

data("mtcars")

You can visualise the dataset as follows.

funky_heatmap(mtcars)

Customising the plot

However, this plot can look so much better if you provide additional metadata for the rows and columns. See more information on how you can customise.

data <- mtcars %>%
  rownames_to_column("id") %>%
  arrange(desc(mpg))

column_info <- tribble(
  ~id,     ~group,         ~name,                      ~geom,        ~palette,    ~options,
  "id",    "",             "",                         "text",       NA,          list(hjust = 0, width = 6),
  "mpg",   "overall",      "Miles / gallon",           "bar",        "palette1",  list(width = 4, legend = FALSE),
  "cyl",   "overall",      "Number of cylinders",      "bar",        "palette2",  list(width = 4, legend = FALSE),
  "disp",  "group1",       "Displacement (cu.in.)",    "funkyrect",  "palette1",  lst(),
  "hp",    "group1",       "Gross horsepower",         "funkyrect",  "palette1",  lst(),
  "drat",  "group1",       "Rear axle ratio",          "funkyrect",  "palette1",  lst(),
  "wt",    "group1",       "Weight (1000 lbs)",        "funkyrect",  "palette1",  lst(),
  "qsec",  "group2",       "1/4 mile time",            "circle",     "palette2",  lst(),
  "vs",    "group2",       "Engine",                   "circle",     "palette2",  lst(),
  "am",    "group2",       "Transmission",             "circle",     "palette2",  lst(),
  "gear",  "group2",       "# Forward gears",          "circle",     "palette2",  lst(),
  "carb",  "group2",       "# Carburetors",            "circle",     "palette2",  lst()
)

funky_heatmap(data, column_info = column_info, expand = list(xmax = 4))

More information

• The reference documentation on funky_heatmap() details the exact formats of each annotation object that you can pass to it.

• Check out the vignette vignette("mtcars", "funkyheatmap") for more information on how to customize this visualisation.

• In vignette("dynbenchmark", "funkyheatmap") we use funkyheatmap to regenerate the figures from Saelens et al. (2019) doi:10.1038/s41587-019-0071-9.

• We used Viash to wrap the funkyheatmap::funky_heatmap() function as a standalone executable and Nextflow module.