Heatmap Colormaps!

Normal Incrementing Data

Nine times out of ten, you probably need just one of the {viridis} colormaps for values that are increasing in one direction. For example, correlation values that are all positive, or abundances from RNA-Seq, etc. If you want to use them with {ComplexHeatmap}, you do so like this:

# correlation example for 0 - 1
# imagine cor_vals is a set of correlation values
library(ComplexHeatmap)
cor_start = 0
cor_end = 1
n_break = 20
viridis_map = circlize::colorRamp2(seq(cor_start, 
                                       cor_end, 
                                       length.out = n_break),
                                   viridis::viridis(n_break))

Heatmap(cor_vals, 
        col = viridis_map, 
        ...)

Divergent Data

In this case, you have values where a central value is a natural point that you want to highlight things above and below that value. Examples include correlations that include negative and positive correlations, log-fold-changes above and below zero. In this case, I prefer the {scico} package with the vik colormap.

cor_start = -1
cor_end = 1
n_break = 20

scico_map = circlize::colorRamp2(seq(cor_start
                                     cor_end,
                                     length.out = n_break),
                                 scico::scico(n_break, 
                                              palette = "vik"))

Heatmap(cor_vals, 
        col = scico_map,
        ...)

Other Colormaps?

Do you have other suggestions for colormaps, both incremental and divergent? Please let me know in the comments!

Reuse

https://creativecommons.org/licenses/by/4.0/

Citation

BibTeX citation:

@online{mflight2022,
  author = {Robert M Flight},
  title = {Heatmap {Colormaps!}},
  date = {2022-01-31},
  url = {https://rmflight.github.io/posts/2022-01-31-heatmap-colormaps},
  langid = {en}
}

For attribution, please cite this work as:

Robert M Flight. 2022. “Heatmap Colormaps!” January 31, 2022. https://rmflight.github.io/posts/2022-01-31-heatmap-colormaps.