ggplot with ORR in y axis

Good afternoon,

I was wondering if anyone could guide me here: I would like to change my current plot in a way that I get the overall response rate (percentage of Responders) in the y-axis. I am not interested to show the percentage of Non-responder. Then I would also remove the current legend.

enter image description here

Current code:

p<-metadata %>%
    dplyr::count(triplet_douplet, ORR) %>% 
    dplyr::group_by(triplet_douplet) %>%
    dplyr::mutate(prop = n / sum(n)) %>%
  ggplot(aes(x =triplet_douplet, y = prop, fill =ORR)) +
  geom_col(position = position_dodge()) +
  geom_text(aes(label = round(100 * prop)),
            position = position_dodge(.9), vjust = -.2
  )

I have been trying different attempts but no success yet. Any ideas?

ggplot

• 1,495 views • link

written 1.8 years ago by Nona •

If i understand correctly, why not just change the mutate bit to:

dplyr::mutate(prop = 100 * n / sum(n))

As well as removing the 100 from the geom_text command.

You can also remove the legend by using:

theme(legend.position = "none")

• 0 views • link

written 1.8 years ago by biofalconch

Thanks for your reply. But this is not what I meant. My aim is to get something like this:

enter image description here

I am not interested in the percentage of the non-responder, just show the percentage of the responder:

• 0 views • link

written 1.8 years ago by Nona •

have you tried something like this before the ggplot command?

dplyr::filter( ORR == "Responder" ) %>%

• 0 views • link

written 1.8 years ago by biofalconch

Great, yes like this it is working! Thanks so much.

Sharing the code in case someone is interested:

metadata %>%
dplyr::count(triplet_douplet, ORR) %>% 
 dplyr::group_by(triplet_douplet) %>%
  dplyr::mutate(prop = n / sum(n)) %>%
  dplyr::filter( ORR == "Responder" ) %>%
  ggplot(aes(x =triplet_douplet, y = prop, fill =ORR)) +
  geom_col(position = position_dodge()) +
  geom_text(aes(label = round(100 * prop)),
            position = position_dodge(.9), vjust = -.2
  )

I have one final question: In case I want to replace "ORR" in fill with "height" it gives me this error, even so "height" is a variable in metadata. Maybe it happens because I filter and it cannot longer recognise it?!

Thanks again!

Error in `geom_col()`:
! Problem while computing aesthetics.
Error occurred in the 1st layer.
Caused by error in `FUN()`:
! object 'height' not found

• 0 views • link

written 1.8 years ago by Nona •

Hard to say without looking at the data and the ggplot command, are you sure this is column is around?

• 0 views • link

written 1.8 years ago by biofalconch

Ok got it. Need to add "height" here:

dplyr::count(triplet_douplet,height, ORR) %>%

• 0 views • link

written 1.8 years ago by Nona •

0 answers

No answers yet.