specify that zero values get neutral color in diverging data

[kieran-mace]

Depending on the distribution of the data, sometimes I get zero values as red, or blue. could there be an option that forces zero to white?

[kieran-mace]

similar to balanceColor=TRUE in heatmap3

[talgalili]

Hi @KMace
The d3heatmap is currently not under active development. Another project that you might consider is the heatmaply package. For what you ask for, please look at the last example here:
https://cran.r-project.org/web/packages/heatmaply/vignettes/heatmaply.html#changing-color-palettes

heatmaply(mtcars, margins = c(40, 130), 
          scale_fill_gradient_fun = ggplot2::scale_fill_gradient2(low = "blue", high = "red", midpoint = 200, limits = c(0, 500)))

With this output (which will be interactive)

[klaukh]

@KMace
Over on my fork - https://github.com/klaukh/d3heatmap - I've been slowly working many of the community requests, including the one you've mentioned and a more modern API. You can either do d3heatmap(..., symbreaks = TRUE) to match the heatmap.2 and heatmap.3 APIs, or do d3heatmap(...) %>% hmColors(..., symmetrical = TRUE) to achieve what you're looking for.