First of all, thank you for the really nice piece of work! At the moment, I’m trying to draw a Venn diagram for a research article and I wonder whether it is possible to display the percentages based on the total number of observations of the data (instead of observations included in the diagram)?

For example, let say I have data of 1000 observations and I create a Venn diagram of its logical variables A, B, C and D. However, values of 100 observations are false regarding all of those four variables and thus they are not included in the diagram. The intersection of ABCD consists of 90 observations and when displaying the percentages, it shows 10% (90 out of 900 observations). However, I would like to show the proportion of ABCD in relation to the total data, which in this case would be 9% (90/1000). Is that somehow possible?

I would really appreciate if someone could help me out with this.

When arranging with cowplot::plot_grid, two venn plot will change the ratio for each plot. Could the ggvenn fix the ratio of width and height for each plots?
Thanks!
Below is what I got.

Hello
thank you for this nice package.

it is exactly what I was looking for with one limitation.. I have a dataset made of 6 lists of overlapping genes across 6 tissues
can we increase the number of sets above 4? I found other packages like 'venn' doing it but they do not offer all your nice options

thank you!

Hey, @yanlinlin82!

I like this package a lot. It's easy to use.

Have you considered adding a pkdown website/documentation as well as some contributing docs?

It's pretty straightforward to set up, and I think it'd help the package's visibility and make it a little easier to look at the api.

Dear @yanlinlin82,

Thank you for the package!

I have a minor question regarding the plot size. Namely, if you try to align couple of plots using i.e. patchwork it seems like the venn diagrams are not occupying the full plot area.

Following code:

library(ggvenn)
library(patchwork)

a <- list(`Set 1` = c(1, 3, 5, 7, 9),
          `Set 2` = c(1, 5, 9, 13),
          `Set 3` = c(1, 2, 8, 9),
          `Set 4` = c(6, 7, 10, 12))

p1 <- ggvenn(a, c("Set 1", "Set 2"))

p2 <- iris %>% ggplot(aes(x = Sepal.Length, y = Sepal.Width)) + geom_point()

p1 / p2

produces this figure from which it seems like the venn diagrams are not in the full plot area:

I would expect the plots to occupy full ploting area, like in the example bellow:

p2 / p2

I tried following:

p1 <- ggvenn(a, c("Set 1", "Set 2")) + theme(plot.margin = unit(c(0,0,0,0), "cm"))

p1/p2

but seems like it has no effect.

Do you have any suggestion how to expend ggvens to occupy full plot area?

Thank you in advance
Amel

I used the function:
ggvenn(venn_list, fill_color = c("#1F77B4FF", "#FF7F0EFF"), fill_alpha = .7, stroke_size = 0.3, stroke_alpha = .5, show_percentage = FALSE, text_size = 8, set_name_size = 6) + labs(title = 'TNK') + theme(plot.title = element_text(size=18, face="bold", hjust=0.5))
The output is like this:

I want to adjust the distance between the setnames (such as HCC and ICC) and the circle, or do something which making the title (TNK) will not cover the setnames.

Hi,
Thanks for the beautiful package.
Any ideas on how can I adjust the position of the set names? for longer set names it doesn't show them fully no matter the plot size. (e.g test1test2test3test4 in this example)
Thanks

Hi,

Is it possible to give a different stroke colour to each circle just as the fill colour?

regards

I get various errors when trying to use the package by itself wit ggplot2 loaded, for example:

• Error in data %>% select_if(is.logical) %>% names: could not find function "%>%"
• could not find function "as_tibble"

They go away if I explicitly load entire tidyverse (library(tidyverse)) which is a thing I would prefer to avoid. Wold you please consider declaring the imports that you use, e.g. with @importFrom or manual NAMESPACE curation?

Also, would you consider submitting the package for CRAN? I think it has a great potential because it has the cleanest API from several R packages implementing venn diagrams I looked at.

I was wondering how to display large numbers separated by a comma in ggvenn? Any help will be appreciated?

Hi Yan,

I am using your ggvenn for my own data analysis. However, I can't convert my dataset from the list “a" to dataframe "d" as you showed on your GitHub. Could you tell me how to convert this list to the tibble?

Thanks for making this package.

I was getting the following error for any input, including the example data:

Error in count(., A, B, C, D) : Argument 'x' must be a vector: list

The issue was from another loaded package (DESeq2) that masked dplyr::count. It works if I copy the ggvenn code and replace "count" with "dplyr::count" or if I detach DESeq2 before calling ggvenn.

Not sure that's worth an update, but someone may find this if they run into the same issue.

Thank you for a very proimising package!
I was wondering if you could add an option to e.g. remove the percentage display to only show counts. It would improve readability in small displays.

Is there a way, that values like 0 and NA are not recognized in an intersection in a list? I believe that zero should be recognized as nothing, or null.

Thank Anders for posting this question: https://support.bioconductor.org/p/133731/

library(ggvenn)
d<-data.frame(id=paste(LETTERS[1:9]),
           A=c(TRUE,TRUE,TRUE,FALSE,FALSE,TRUE,TRUE,FALSE,FALSE),
           B=c(FALSE,FALSE,TRUE,TRUE,TRUE,TRUE,TRUE,FALSE,FALSE),
           C=c(FALSE,FALSE,TRUE,TRUE,FALSE,TRUE,TRUE,TRUE,FALSE))

p<-ggplot(d) +
  geom_venn(aes(A=A,B=B,C=C,label=id)) +
  coord_fixed()
  theme_void()
p

In the above code, the label of each intersection is collapsed by comma (,). It would be better to allow customize it.

It would be good to be able to show values.

Something like this (not sure if I am representing this correctly):

a <- c(1, 3) # A, AB
b <- c(1, 2) # AB, B

venn <- c(a[1], # A
          a[2] + b[1], # AB
          b[2]) # B

> venn
[1] 1 4 2

Hi,

I am trying to run ggvenn with 3 areas, this is my code:

ggvenn(
  y, 
  fill_color = c("#0073C2FF", "#EFC000FF", "#868686FF"),
  stroke_size = 0.4, set_name_size = 6
)

but I am getting gray areas, and also the following warning:

Warning message:
`guides(<scale> = FALSE)` is deprecated. Please use `guides(<scale> = "none")` instead. 

Any ideas how to fix this?
I do have the latest version, 0.1.9.

Hi, when show_percentage = FALSE the function geom_venn() return a warning. Is there a way to avoid this and still use show_percentage = FALSE. Thanks in advance

Warning message:
Problem while computing text = sprintf("%d", n, 100 * n/sum(n)).
ℹ one argument not used by format '%d'

Hi,
Thank you for creating this. I was wondering if there's an option to make the circle areas represent the percentage of overlap, like the scale parameter in VennDiagram. Thank you!

Nice package.
How can I adjust the colors of circles/elipses and borders with either ggvenn or geom_venn?
Also any way to change transparency, border thickness, etc. Thanks.

Hello

Thank you for your code - it is very nice. I am using the code to look at 3 gene sets with the overlapping and unique genes represented by the Venn diagram. I have used show_elements to look at these genes directly on the Venn diagram but I am wondering whether there is a convenient way to have them in a table?

I am especially interested in a way to view the unique genes in each set (i.e. not overlapping with either of the other 2 sets).

Best wishes,
Jane

Please add ggvenn to conda-forge, it would get a lot more users and people could better integrate into their environments and software dists

Hello,

I use your package but I have a bug when I use the diagramme in 4d, there is an intersection where the number is false.

Take this test file test.txt,

library(data.table)
library(tidyverse)
library(ggvenn)

test = fread(test.txt)

test %>% ggvenn()

nrow(test[A == T & B == T & C == F & D==T])

The problem is on the intersection of A, B and D, your diagramme indicate 2 instead of 13 like the nrow(test[A == T & B == T & C == F & D==T]) indicate. This is the image I obtained with this test : .

Thanks,

I'm Hanbeen Kim.

Thank you for developing helpful tool "ggvenn".

I would like to change the font family from default setting to "Times New Roman".

Is there any possible way to change?

Sincerely,

Hanbeen Kim

Excellent package! I was wondering if you would consider an option to use text labels instead of counts inside the circles, allowing the package to be used to create "traditional" Venn diagrams, such as the first example from Wikipedia.

Because sometimes one decimal digit precision isn't enough..

Is it possible to include the total count next to the set labels? Thanks.

--Hi,

i use this code below:
ven <- ggplot(d, aes(A = x, B = y, C = z, D = f)) +
geom_venn(
stroke_color="grey",
stroke_size = 0.5,
stat = "identity",
position = "identity",
show_percentage = FALSE,
fill_color=c("#8fd1b2", "#c4b3ab", "#a18073", "#409c9e"),
set_name_color=c("#8fd1b2", "#c4b3ab", "#a18073", "#409c9e"),
set_name_size = 6,
text_size = 6,
text_color = "black",
fill_alpha = 0.5,
stroke_alpha = 0.8,
label_sep = ",",
count_column = NULL,
show_outside = c("auto", "none", "always"),
auto_scale = FALSE) +
theme_void() +
ggtitle("blabla") +
theme(plot.title = element_text(hjust = 0.5, size = rel(1.5))) +
theme(legend.position="none")

and on my plot i have a total count (see on attached picture number 21615),
How to remove this count from picture ?

Thank you --

For example, if there is a set that contains all of the elements, draw the other circles inside that set

Hi,
I get follows error when use this packages to plot a veen figure.

> list(A=1:3,B=3:6) %>%
+     ggvenn(show_elements=TRUE)
Using `n` as weighting variable
i Quiet this message with `wt = n` or count rows with `wt = 1`
Error in prepare_venn_data(data, columns, show_elements, show_percentage) : 
  (d1 %>% count(A, B) %>% with(n)) == 1 are not all TRUE

Hi LinLin!

ggVenn is really a GREAT Pkg,
but I get an error msg
using the GIThub page example code:

# use data.frame as input...
d <- tibble(value = c(1, 2, 3, 5, 6, 7, 8, 9, 10, 12, 13),
Set 1 = c(T, F, T, T, F, T, F, T, F, F, F),
Set 2 = c(T, F, F, T, F, F, F, T, F, F, T),
Set 3 = c(T, T, F, F, F, F, T, T, F, F, F),
Set 4 = c(F, F, F, F, T, T, F, F, T, T, F))

d %>% ggvenn(aes(A = `Set 1`, B = `Set 2`))
Error: Must subset columns with a valid subscript vector.
x Subscript columns[[1]] has the wrong type quosure/formula.
It must be logical, numeric, or character.

Help!
Sfd99
San Francisco
R / Rstudio / Ubuntu Linux 20.04 (all latest versions).

Hello, I was using the ggvenn function without any problems before, but I recently got this error messsage with the same code. Please find an example that give me the same wrning message.

`A=c("blue", "green", 'purple', "orange")
B=c("blue", "yellow", "black", "pink")
C=c("pink", "red", "orange", "green")
D=c("green", "blue", "orange")

ListeVenn=list(A=A, B=B, C=C, D=D)

ggvenn(ListeVenn)

"Warning message:
guides(<scale> = FALSE) is deprecated. Please use guides(<scale> = "none") instead." `

The veen diagramm is made but without any colours and some lines are missing

Below is the plot I am obtaining following the warning message

Apart from the standard venn diagram plot:

• How would I add separate labels within and outside the overlap areas?
• Can there be a possibility of doing this similar to what ggrepel does?
• Are you planning to add a scaling functionality where the size of overlap and circles scales to the number of the elements being represented?
• How do I change the font of the text, how would I do it separately for the set names and within the venn diagram?

Fantastic package btw! This has been missing for a while for ggplot2.

Dear @yanlinlin82,
I am developing some modules for jamovi.org
I think ggvenn will be useful for clinical researchers, so I have added your package to see if it will work inside jamovi.
I will try to add arguments as well.
Please let me know what you think, and please check if the reference I have given is appropriate.

Best wishes