I'm creating a barplot with error bars using ggplotly. The transition into plotly changes the width and thickness of the error bars, and I would like to format them to my preference. I saw that I'm supposed to do that using the style function, but can't quite figure it out. I can see that the trace I need is called error_y and I can change the attributes of the bars, but can't figure out how to change any attribute of the error bars...

library(dplyr)
library(ggplot2)
library(plotly)

df <- data.frame(y = c(5, 10), group = c("a", "b"))
p <- ggplot(df) +
        geom_bar(aes(x = group, y = y), stat = "identity") +
        geom_errorbar(aes(x = group, ymin = y - 1, ymax = y + 1), linewidth = 0.1, width = 0.7) 
        
ggp <- ggplotly(p)
str(ggp) # I can see the markers (the bars) and the error_y (the error bars in the structure of the plotly object

# can modify attributes of the bars successfully
style(ggp, traces = 1, marker = list(color = "red"))
# fail to modify attributes of the errorbars
style(ggp, error_y = list(thickness = 2.5))

I'm creating a barplot with error bars using ggplotly. The transition into plotly changes the width and thickness of the error bars, and I would like to format them to my preference. I saw that I'm supposed to do that using the style function, but can't quite figure it out. I can see that the trace I need is called error_y and I can change the attributes of the bars, but can't figure out how to change any attribute of the error bars...

library(dplyr)
library(ggplot2)
library(plotly)

df <- data.frame(y = c(5, 10), group = c("a", "b"))
p <- ggplot(df) +
        geom_bar(aes(x = group, y = y), stat = "identity") +
        geom_errorbar(aes(x = group, ymin = y - 1, ymax = y + 1), linewidth = 0.1, width = 0.7) 
        
ggp <- ggplotly(p)
str(ggp) # I can see the markers (the bars) and the error_y (the error bars in the structure of the plotly object

# can modify attributes of the bars successfully
style(ggp, traces = 1, marker = list(color = "red"))
# fail to modify attributes of the errorbars
style(ggp, error_y = list(thickness = 2.5))

• r
• ggplot2
• plotly
• ggplotly

1 Answer 1

It is not very intuitive, but you need to set error_y.VARIABLE to your desired value.

library(ggplot2)
library(plotly)

df <- data.frame(y = c(5, 10), group = c("a", "b"))

p <- ggplot(df) +
  geom_bar(aes(x = group, y = y), stat = "identity") +
  geom_errorbar(aes(x = group, ymin = y - 1, ymax = y + 1), 
                linewidth = 0.1, width = 0.7) 

ggp <- ggplotly(p)

## Original plot
ggp
## Updated with style()
style(ggp,  error_y.thickness = 5, error_y.color = "red", error_y.width = 40)

Original plot:

Updated with style():

Looking at plotly structure could be useful in these cases:

require(listviewer)
plotly_json(ggp)