Venn Diagram by ggplot2, with really easy-to-use API.
install.packages("ggvenn") # install via CRAN
or
if (!require(devtools)) install.packages("devtools")
devtools::install_github("yanlinlin82/ggvenn") # install via GitHub (for latest version)
This package provides two main functions: ggvenn() and
geom_venn(). It supports both list and
data.frame type data as input.
For list data (each element is a set):
library(ggvenn)
a <- list(A = 1:5, B = 4:9, C = 3:7, D = 1:20, E = 15:19)
ggvenn(a, c("A", "B")) # draw two-set venn
ggvenn(a, c("A", "B", "C")) # draw three-set venn
ggvenn(a, c("A", "B", "C", "D")) # draw four-set venn
ggvenn(a) # without set names, all elements in list will be chosen to draw venn
For data.frame data (each logical column is a set):
d <- data.frame(
id = 1:32,
A = 1:32 %% 2 == 1,
B = (1:32 %/% 2) %% 2 == 1,
C = (1:32 %/% 4) %% 2 == 1,
D = (1:32 %/% 8) %% 2 == 1,
E = (1:32 %/% 16) %% 2 == 1
)
ggvenn(d, c("A", "B")) # draw two-set venn
ggvenn(d, c("A", "B", "C")) # draw three-set venn
ggvenn(d, c("A", "B", "C", "D")) # draw four-set venn
ggvenn(d) # without set names, all logical columns in data.frame will be chosen to draw venn
ggvenn(d, element_column = "id", show_elements = TRUE)
ggvenn() for standalone
plots, geom_venn() for ggplot2 grammarFor data.frame data, there is also another way to plot
in ggplot grammar:
# draw two-set venn (use A, B in aes)
ggplot(d, aes(A = `Set 1`, B = `Set 2`)) +
geom_venn() + theme_void() + coord_fixed()
# draw three-set venn (use A, B, C in aes)
ggplot(d, aes(A = `Set 1`, B = `Set 2`, C = `Set 3`)) +
geom_venn() + theme_void() + coord_fixed()
# draw four-set venn (use A, B, C, D in aes)
ggplot(d, aes(A = `Set 1`, B = `Set 2`, C = `Set 3`, D = `Set 4`)) +
geom_venn() + theme_void() + coord_fixed()
There are more options for customizing the venn diagram.
Tune the color and size
For filling:
fill_color - default is c(“blue”, “yellow”, “green”,
“red”)fill_alpha - default is 0.5For stroke:
stroke_color - default is “black”stroke_alpha - default is 1stroke_size - default is 1stroke_linetype - default is “solid”For set name:
set_name_color - default is “black”set_name_size - default is 6For text:
text_color - default is “black”text_size - default is 4All parameters above could be used in both ggvenn() and
geom_venn().
For example:
a <- list(A = 1:4, B = c(1,3,5))
ggvenn(a, stroke_linetype = 2, stroke_size = 0.5,
set_name_color = "red", set_name_size = 15,
fill_color = c("pink", "gold"))Show elements
show_elements - default is FALSElabel_sep - text used to concatenate elements, default
is “,”For example:
a <- list(A = c("apple", "pear", "peach"),
B = c("apple", "lemon"))
ggvenn(a, show_elements = TRUE)
ggvenn(a, show_elements = TRUE, label_sep = "\n") # show elements in lineHide percentage
show_percentage - default is TRUEFor example:
a <- list(A = 1:5, B = 1:2)
ggvenn(a, show_percentage = FALSE)Change digits of percentage
digits - default is 1For example:
a <- list(A = 1:5, B = 1:2)
ggvenn(a, digits = 2)Show/hide statistics
show_stats - control what to display: “cp” (count +
percentage), “c” (count only), “p” (percentage only)show_set_totals - show totals for each set: “cp”, “c”,
“p”, or “none”For example:
a <- list(A = 1:5, B = 1:2)
ggvenn(a, show_stats = "c") # show only counts
ggvenn(a, show_stats = "p") # show only percentages
ggvenn(a, show_set_totals = "cp") # show set totalsControl outside elements
show_outside - show elements not belonging to any set:
“auto”, “none”, “always”For example:
a <- list(A = 1:5, B = 4:8, C = 10:15) # element 10-15 are outside A and B
ggvenn(a, c("A", "B"), show_outside = "always")Auto-scaling (2-set diagrams only)
auto_scale - automatically resize circles based on
element countsFor example:
a <- list(A = 1:100, B = 50:150) # very different sizes
ggvenn(a, auto_scale = TRUE)When creating multiple venn diagrams, you can use
patchwork or gridExtra for layout:
library(ggvenn)
library(patchwork) # or library(gridExtra)
# Create multiple plots
g1 <- ggvenn(list(A = 1:5, B = 4:8))
g2 <- ggvenn(list(A = 1:5, B = 4:8, C = 3:7))
g3 <- ggvenn(list(A = 1:5, B = 4:8, C = 3:7, D = 1:20))
g4 <- ggvenn(list(A = 1:5, B = 4:8, C = 3:7, D = 1:20, E = 15:19))
# Using patchwork (recommended)
(g1 | g2) / (g3 | g4)
# Using gridExtra
# gridExtra::grid.arrange(g1, g2, g3, g4, ncol = 2, nrow = 2)
The ggvenn support two types of input data: list and
data.frame. Two functions (data_frame_to_list() and
list_to_data_frame()) can convert data between the two
types.
a <- list(A = 1:5, B = 4:6)
d <- dplyr::tibble(key = 1:6,
A = c(rep(TRUE, 5), FALSE),
B = rep(c(FALSE, TRUE), each = 3))
identical(a, data_frame_to_list(d)) # TRUE
identical(d, list_to_data_frame(a)) # TRUE