0

I want to create a plot and add a secondary z-axis. The problem is that I don't want the secondary axis on the top, but I want both x axis at the bottom of the graph. any ideas how to do it?

I have this data frame:

df <- read.table(text="Control     Stress days sd_control  sd_stress
-0.2866667 -0.2833333   X1 0.11846237 0.05773503
-0.2566667 -1.0333333   X2 0.08144528 0.15275252
-0.4766667 -1.4500000   X3 0.09291573 0.10000000
-0.4900000 -1.2766667   X4 0.21517435 0.22501852
-0.4600000 -1.2666667   X5 0.07549834 0.40722639
-0.2633333 -1.0833333   X6 0.12662280 0.10408330
-0.2833333 -1.0333333   X7 0.03511885 0.07767453", header=T)

I create a simple plot and add a secondary x-axis as follors:

ggplot(data=df,aes(x=Control, y=Stress))+geom_point()+scale_x_continuous(sec.axis = sec_axis(~ .+50,))

and get the following plot:

enter image description here

Now, how do I move the axis from top to bottom?

not sure if its possible in ggplot, but R has a function called axis() that can add a secondary axis with whatever characteristics to a regular plot.

neilfws
  • 32,751
  • 5
  • 50
  • 63
Juan Pablo Rincon
  • 19
  • 5
  • 1
    See https://stackoverflow.com/questions/71630756/is-it-possible-to-add-a-third-y-axis-to-ggplot2 for an approach to add multiple axes on one side of the plot using `patchwork`. – stefan Mar 13 '23 at 21:36
  • [This is interesting, but whether you agree or not, you can see at least one method that might work.](https://stackoverflow.com/questions/3099219/ggplot-with-2-y-axes-on-each-side-and-different-scales) – Kat Mar 13 '23 at 22:33

1 Answers1

0

another possibility:

ggplot(data=df,aes(x=Control, y=Stress))+geom_point() +
  scale_x_continuous(labels = ~paste(., .+50, sep = "\n"),
                     name = "Primary axis\nSecondary axis")

enter image description here

With some more effort this could be extended to any arbitrary set of scales:

labels <- data.frame(
  x = c(
    seq(-0.5, -0.25, length.out = 10),
    seq(-0.5, -0.25, length.out = 11)
  ),
  label = c(
    format(seq(-0.5, -0.25, length.out = 10), digits = 2),
    paste0("\n", LETTERS[1:11])
  )
)


ggplot(data=df,aes(x=Control, y=Stress))+geom_point() +
  scale_x_continuous(breaks = labels$x,
                     labels = labels$label,
                     minor_breaks = NULL)

enter image description here

Or we could overlay the 2nd axis as labels:

ggplot(data=df,aes(x=Control, y=Stress))+geom_point() +
  scale_x_continuous(breaks = seq(-0.5, -0.25, length.out = 10),
                     labels = format(seq(-0.5, -0.25, length.out = 10), digits = 2)) +
  annotate("text", 
           x = seq(-0.5, -0.25, length.out = 11),
           y = -1.62,
           size = 2.5,
           label = LETTERS[1:11]) +
  coord_cartesian(ylim = c(-1.5, NA), clip = "off")

(I haven't bothered to tweak the margin spacing to make room, but you get the idea.)

enter image description here

Jon Spring
  • 55,165
  • 4
  • 35
  • 53