Legend not showing fill and colors

[Stefanie Trop]

Hi,

I am making a graph like below, where I am manually specifying colors (for lines) and fills (for points). It was working before, but then I made some change that made me lose the colors in the legend. Can you see what mistake I made and how to fix?

Thanks!

Stefanie Trop


  ggplot(filter(forggplot), aes(reorder(cytokine, percent_res), percent_res, group = cytokine, label = percent_res))+
    geom_line(aes(color = responseType))+
  geom_point(aes(shape = MSD, fill = responseType), size = 6)+
  scale_shape_manual(values = c(21, 22))+
    scale_color_manual(values = c('#1b9e77', '#d95f02', '#7570b3', '#66a61e'),
                      labels = c('antiinflam', 'Th1', 'Th2', 'Th17'),
                      name = 'Category')+
  scale_fill_manual(values = c('#1b9e77', '#d95f02', '#7570b3', '#66a61e'),
                    labels = c('antiinflam', 'Th1', 'Th2', 'Th17'),
                    name = 'Category')+
  coord_flip()+
  theme_bw()+
  geom_text(color = 'white', size = 2, face = 'bold')+
  labs(x = 'Cytokine',
       y = '% responders',
       title = paste('Percent responders'))+
  theme(axis.text = element_text(face = 'bold'))

[Brandon Hurr]

The change seems to happen when you take color = responseType out of the main ggplot() and put it only in geom_line().

library(tidyverse)

forggplot <- tibble(cytokine = rep(c("IFNg", "IL17", "IL4", "TNFa","IL10"), each = 2), responseType = rep(c('antiinflam', 'Th1', 'Th2', 'antiinflam','Th17'), each = 2), percent_res = c(65,81,67,70,60,63,48,60, 19, 30), MSD = rep(c("N", "Y"), 5) )

#everything in the ggplot()

ggplot(forggplot, aes(reorder(cytokine, percent_res), percent_res, group = cytokine, color = responseType, fill = responseType, shape = MSD, label = percent_res))+
    geom_line() +
    geom_point(size = 6) +
    geom_text(color = 'white', size = 2)    +
    scale_shape_manual(values = c(21, 22)) +

    scale_color_manual(values = c('#1b9e77', '#d95f02', '#7570b3', '#66a61e')

                            , labels = c('antiinflam', 'Th1', 'Th2', 'Th17')

                            , name = 'Category') +
    scale_fill_manual(values = c('#1b9e77', '#d95f02', '#7570b3', '#66a61e')

                            , labels = c('antiinflam', 'Th1', 'Th2', 'Th17')

                            , name = 'Category') +

    coord_flip() +
    theme_bw() +

    labs(x = 'Cytokine',
       y = '% responders',
       title = paste('Percent responders')) +
    theme(axis.text = element_text(face = 'bold'))
   

# only the colour not

ggplot(forggplot, aes(reorder(cytokine, percent_res), percent_res, fill = responseType, shape = MSD, group = cytokine, label = percent_res))+
    geom_line(aes(colour = responseType)) +
    geom_point(, size = 6) +
    geom_text(color = 'white', size = 2)    +
    scale_shape_manual(values = c(21, 22)) +

    scale_color_manual(values = c('#1b9e77', '#d95f02', '#7570b3', '#66a61e')

                            , labels = c('antiinflam', 'Th1', 'Th2', 'Th17')

                            , name = 'Category') +
    scale_fill_manual(values = c('#1b9e77', '#d95f02', '#7570b3', '#66a61e')

                            , labels = c('antiinflam', 'Th1', 'Th2', 'Th17')

                            , name = 'Category') +

    coord_flip() +
    theme_bw() +

    labs(x = 'Cytokine',
       y = '% responders',
       title = paste('Percent responders')) +
    theme(axis.text = element_text(face = 'bold'))

What I can't do is explain why that is going on. You are unifying the scales by name as you're supposed to do. Someone else may know more, but that's what's happening.

Next time do try and attach a dataset with your code. Making one up from your data took longer than trying to understand what's going wrong with it.

HTH,

B

--
--
You received this message because you are subscribed to the ggplot2 mailing list.
Please provide a reproducible example: https://github.com/hadley/devtools/wiki/Reproducibility
 
To post: email ggp...@googlegroups.com
To unsubscribe: email ggplot2+unsubscribe@googlegroups.com
More options: http://groups.google.com/group/ggplot2

---
You received this message because you are subscribed to the Google Groups "ggplot2" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ggplot2+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Hadley Wickham]

I suspect it's because you're using shape = c(21, 22). You'll also
need to pass one of them to the scale by using override.aes

Hadley

> --
> --
> You received this message because you are subscribed to the ggplot2 mailing
> list.
> Please provide a reproducible example:
> https://github.com/hadley/devtools/wiki/Reproducibility
>
> To post: email ggp...@googlegroups.com

> To unsubscribe: email ggplot2+u...@googlegroups.com

> More options: http://groups.google.com/group/ggplot2
>
> ---
> You received this message because you are subscribed to the Google Groups
> "ggplot2" group.
> To unsubscribe from this group and stop receiving emails from it, send an

> email to ggplot2+u...@googlegroups.com.

> For more options, visit https://groups.google.com/d/optout.

--
http://hadley.nz

[Brandon Hurr]

Hadley and Stefanie,

Yes, that does correct the issue:

ggplot(forggplot, aes(reorder(cytokine, percent_res), percent_res, group = cytokine, label = percent_res))+
    geom_line(aes(colour = responseType)) +
    geom_point(aes(fill = responseType, shape = MSD), size = 6) +

    geom_text(color = 'white', size = 2)    +
    scale_shape_manual(values = c(21, 22)) +

    scale_color_manual(values = c('#1b9e77', '#d95f02', '#7570b3', '#66a61e')

                            , labels = c('antiinflam', 'Th1', 'Th2', 'Th17')
                            , name = 'Category') +

    scale_fill_manual(values = c('#1b9e77', '#d95f02', '#7570b3', '#66a61e')

                            , labels = c('antiinflam', 'Th1', 'Th2', 'Th17')
                            , name = 'Category') +
    coord_flip() +
    theme_bw() +

    labs(x = 'Cytokine',
       y = '% responders',
       title = paste('Percent responders')) +

    theme(axis.text = element_text(face = 'bold')) +
    guides(fill = guide_legend(override.aes = list(shape = 21)))

You can specify color or fill or both and all of those print the plot as desired.

HTH,

B

> To unsubscribe: email ggplot2+unsubscribe@googlegroups.com

> More options: http://groups.google.com/group/ggplot2
>
> ---
> You received this message because you are subscribed to the Google Groups
> "ggplot2" group.
> To unsubscribe from this group and stop receiving emails from it, send an

> email to ggplot2+unsubscribe@googlegroups.com.

> For more options, visit https://groups.google.com/d/optout.

--
http://hadley.nz

--
--
You received this message because you are subscribed to the ggplot2 mailing list.
Please provide a reproducible example: https://github.com/hadley/devtools/wiki/Reproducibility

To post: email ggp...@googlegroups.com

To unsubscribe: email ggplot2+unsubscribe@googlegroups.com

More options: http://groups.google.com/group/ggplot2

---
You received this message because you are subscribed to the Google Groups "ggplot2" group.

To unsubscribe from this group and stop receiving emails from it, send an email to ggplot2+unsubscribe@googlegroups.com.

[Stefanie Trop]

Yes, thanks to you both! I have never used override.aes before.