README.md
In ggcorrheatmap: Make Flexible 'ggplot2' Correlation Heatmaps

ggcorrheatmap

ggcorrheatmap is a convenient package for generating correlation heatmaps made with ggplot2, with support for triangular layouts, clustering and annotation. As the output is a ggplot2 object you can further customise the appearance using familiar ggplot2 functions. Besides correlation heatmaps, there is also support for making general heatmaps.

You can install ggcorrheatmap from CRAN using:

install.packages("ggcorrheatmap")

Or you can install the development version from GitHub with:

# install.packages("devtools")
devtools::install_github("leod123/ggcorrheatmap")

Below is an example of how to generate a correlation heatmap with clustered rows and columns and row annotation, using a triangular layout that excludes redundant cells.

library(ggcorrheatmap)

set.seed(123)
# Make a correlation heatmap with a triangular layout, annotations and clustering
row_annot <- data.frame(.names = colnames(mtcars),
                        annot1 = sample(letters[1:3], ncol(mtcars), TRUE),
                        annot2 = rnorm(ncol(mtcars)))
ggcorrhm(mtcars, layout = "bottomright",
         cluster_rows = TRUE, cluster_cols = TRUE,
         show_dend_rows = FALSE, annot_rows_df = row_annot)

Or a mixed layout that displays different things in the different triangles.

# With correlation values and p-values
ggcorrhm(mtcars, layout = c("topright", "bottomleft"),
         cell_labels = c(FALSE, TRUE), p_values = c(FALSE, TRUE))

A correlation heatmap of mtcars columns. Half of the plot, above the diagonal, is a normal heatmap with coloured cells and the other half has white cells with coloured text inside. The diagonal has coloured cells with names column names written inside. The coloured text in the bottom triangle shows correlation values with asterisks added indicating p-value thresholds.

It is also possible to make a normal heatmap, for a more flexible output.

gghm(scale(mtcars), cluster_rows = TRUE, cluster_cols = TRUE)

A heatmap of the mtcars data (scaled). Cells are coloured with a scale going from dark blue to light blue. Both rows and columns have been clustered, with dendrograms to the right and below the heatmap.