GGplot2 coord flip problem

Question

[garciajessica0311]

Hi, I have this dataframe from RNAseq enrichment pathways : 4 columns (Module, Pathway, OR, adjusted p=value) and 83 rows (enriched pathways)

Module  Pathway OR  Adjusted p-value
Gene_atlas  WholeBlood  9.035869654 0.005206627
Gene_atlas  CD14+_Monocytes 3.610643472 0.036066265
GO_Biological Process   GO:0045321_leukocyte activation 4.316600829 0.003847901

I have plot this data with ggplot2. This is my R code :

gplot(my_data_C2C1, aes(x = reorder(Pathway, Adjusted.p.value), y = -log10(Adjusted.p.value), fill = Module)) + geom_bar(stat = "identity")

But I do no see correctly my pathway names. So, I would like to flip my data :

ggplot(my_data_C2C1, aes(x = reorder(Pathway, Adjusted.p.value), y = -log10(Adjusted.p.value), fill = Module)) + geom_bar(stat = "identity") + coord_flip()

I see better the pathway names but no totally and the bars have disapeared.

How can I reduce the police size of my pathway names and display the bars ?

Thans a lot for your help,

Best,

Jessica

Answer 1

[benformatics]

Does converting the underscores to new-lines help?

my_data_C2C1$Pathway <- gsub("_","\n",my_data_C2C1$Pathway)

[benformatics]

The example dataset in your post works fine for both scripts - if you can't see the bars you probably need to make the display window larger.

[garciajessica0311]

Yes, the windows was too small ... Sorry ! Many thanks for your help ! It's perfect

[GenoMax]

If an answer was helpful, you should upvote it; if the answer resolved your question, you should mark it as accepted. You can accept more than one if they work.

Answer 2

[wm]

We can use the str_wrap( ) function from stringr package to process the text, here is a example on scale_x_discrete(). we can also apply this function to other text content.

# R code
library(ggplot2) # version 3.3.0
library(stringr) # version 1.4.0
library(patchwork) # version 1.0.0

### demo data
df <- data.frame(name = "wrap the text by fix width", score = 1)

### wrap the text
p1 <- ggplot(df, aes(name, score)) +
  geom_bar(stat = "identity")

p2 <- ggplot(df, aes(name, score)) +
  geom_bar(stat = "identity") +
  scale_x_discrete(labels = function(x) str_wrap(x, width = 15))

p1 + p2

Another option, to switch the coordinates

a new feature added in ggplot2 version 3.3.0 , Bi-directional geoms, so we can achieve the same goal as coord_flip() by specify aes(x = score, y = name), here are the code:

See ggplot2 new features link

### switch the coordinates
p1 <- ggplot(df, aes(score, name)) +
  geom_bar(stat = "identity")

p2 <- ggplot(df, aes(score, name)) +
  geom_bar(stat = "identity") +
  scale_y_discrete(labels = function(x) str_wrap(x, width = 15))
p1 + p2

[garciajessica0311]

Thanks for your help ! I keep in mind this technic

Answer 3

[Friederike]

In addition to the line breaks, you can also change the angle of the labels (e.g. to 90 degrees) and the size of the labels as explained in the R cookbook.