ggheat2 | R Documentation |
Function to plot a heat map using ggplot
.
ggheat2(
data,
corr = cor(data, use = "pairwise.complete"),
cluster = TRUE,
nbreaks = NULL,
palette = if (is.null(nbreaks)) c("blue", "white", "red") else 1,
legend_name = expression(rho),
pch,
cex = c(2, 6),
label = FALSE,
label_alpha = FALSE,
label_color = "black",
label_digits = 2,
midpoint = 0,
clim = c(-1, 1),
...
)
data |
a data frame or matrix (observations x variables) of numeric values |
corr |
a correlation matrix |
cluster |
logical or function; if |
nbreaks |
number of breaks to categorize the correlations (default is
|
palette |
for a continuous scale, a vector of length three giving the
low, mid, and high colors of the gradient (default is
|
legend_name |
the legend name; see |
pch |
(optional) plotting character; if |
cex |
size of |
label |
logical; if |
label_alpha |
logical, if |
label_color |
color of correlations (default is |
label_digits |
number of digits in correlation labels |
midpoint |
the midpoint value for continuous scaling of correlations
(default is |
clim |
vector of length two giving the limits of correlation
coefficients (default is |
... |
additional arguments passed to |
Default cluster method is stats::hclust(dist(x), method = 'average')
which will return a list containing a named vector, "order", which is used
to reorder the variables.
In order to pass a custom clustering function to cluster
, the
function must take a single input (a correlation matrix) and return either
a vector or a list with a named vector, "order"
.
cor
, ggheat
, icorr
,
corrplot
, https://github.com/briatte/ggcorr
library('ggplot2')
ggheat2(mtcars)
ggheat2(mtcars, label = TRUE, label_alpha = TRUE, cluster = FALSE,
## additional args passed to diagonal labels
colour = 'red', angle = 45, size = 7)
ggheat2(mtcars, pch = 19, nbreaks = 6, cex = c(2,10),
palette = 'PuOr', ## colorblind palette
size = 5, hjust = 0.75) + ## passed to diag text
labs(title = 'Correlation Matrix')
## custom clustering function
ggheat2(data = NULL, corr = cor(mtcars, use = 'pairwise'),
nbreaks = 5, palette = 'Blues',
cluster = function(...) sample(ncol(mtcars)))
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.