d3heatmap is not actively developed, but I will accept PR. You might consider using heatmaply, which is based on plotly (it comes with more features, but is not based on d3)
This is an R package that implements a heatmap htmlwidget. It has the following features:
- Highlight rows/columns by clicking axis labels
- Click and drag over colormap to zoom in (click on colormap to zoom out)
- Optional clustering and dendrograms, courtesy of
base::heatmap
To install:
if (!require("devtools")) install.packages("devtools")
devtools::install_github("talgalili/d3heatmap")
Like any htmlwidget, you can visualize a d3 heatmap directly from the R console:
library(d3heatmap)
d3heatmap(mtcars, scale = "column", colors = "Spectral")
You can also include them in R Markdown chunks, or use them in Shiny applications with the d3heatmapOutput
and renderD3heatmap
functions.
See ?d3heatmap
for options.
Inspired by dygraphs and able to leverage magrittr, the new API provides a second method for invoking d3heatmap and integrating with Shiny apps, modules, and gadgets! Examples:
library(d3heatmap)
library(magrittr)
d3heatmap(mtcars, dendrogram = 'none', key = TRUE, col = 'RdYlGn',
scale = 'column', key.title = "Legend", print.values = T,
notecol = 'white') %>%
hmAxis("x", title = "test", location = 'bottom') %>%
hmAxis("y", title = "test", location = 'left') %>%
hmCells(font.size = 8, color = 'blue') %>%
hmLegend(show = T, title = "Title", location = "tl")
library(d3heatmap)
library(magrittr)
rsc<-matrix(rep_len(c('good', 'bad', 'ugly'), length.out = 64), ncol = 2)
rsccols<-c('red', 'white', 'blue')
rscnames <- c('Row 1', 'Row 2')
csc<-matrix(rep_len(c('first', 'second', 'third', 'fourth', 'fifth'), length.out = 33), nrow = 3)
csccols<-c('orange', 'blue', 'grey', 'green', 'red')
cscnames <- c('Column 1', 'Column 2', 'Column 3')
library(d3heatmap)
d3heatmap(mtcars,
key = TRUE, scale = 'column',
key.title = "Legend",
col = 'RdYlGn',
srtCol = 30,
breaks = 8,
xlab = 'test',
ylab = 'TEST',
print.values = T,
density.info = 'histogram',
denscol = 'grey',
sideCol = 3,
sideRow = 4,
RowSideColors = rsc,
ColSideColors = csc,
RowColorsPalette = rsccols,
ColColorsPalette = csccols,
RowColorsNames = rscnames,
ColColorsNames = cscnames)
## using the modern API:
d3heatmap(mtcars, key = TRUE, scale = 'none') %>%
hmSideColors(axis = 'column', side.colors = csc,
palette = csccols, names = cscnames) %>%
hmSideColors(axis = 'y', side.colors = rsc,
palette = rsccols, names = rscnames)
An shiny gadget coupled with an S4
class that provides print()
and
save()
methods. Gadget takes normal d3heatmap
inputs and allows for
interactive adjustment of the heatmap. Gadget allows for filtering rows
and columns, and also a dynamic filter to interatively subset the entire
underlying data set. Saving the gadget to an object generates the S4
class
that contains the heatmap, data, filter, and settings. Passing the gadget
back into the function d3heatmapGadget(gadget)
starts the user at the last
state of the gadget.
gadget <- d3heatmapGadget(mtcars, col = 'blues')
print(gadget)
save(gadget, file = "heatmap.html")
gadget <- d3heatmapGadget(gadget)
Based on example contributions and forks from several people, the side colors components of heatmap.2 and heatmap.3 have been added! Functionality includes color labels, axis labels for the color sections, and hover info. Further alignment of the old API parameters to heatmap.2 and heatmap.3, plus heatmap.2/3 and the modern API were implemented for side colors. the Readme, package news and examples were also updated.
The master branch now includes a newer, modern API, motivated by the main d3heatmap fork's desire for a new API and inspired by the API of the dygraphs package produced by RStudio. The new API takes advantage of magrittr piping and offers smaller functions to modify selected portions of the heatmap. I have conducted good, but by no means exhaustive, testing... so feel free to poke around, find bugs, and open up issues or PRs for them.