removing the white space of panel margin

[Babak]

Hello all,

Sorry for the very basic question.  

Simply, I don't know how to get rid of the white space at the margin. I have tried all the options that I knew, but no progress.

Here is how I generated the plot:

this is my data

> head(counts_m)

  id idsort variable      value

1  1      1    ratio 4.72345845

2  2      2    ratio 5.75093918

3  3   1884    ratio 2.83037365

4  4      3    ratio 5.02301873

5  5    719    ratio 0.02301873

6  6    720    ratio 0.02301873

the command to generate the plot is here. I left all the options that I tried, I don't know why expand does not produce the expected result. 

ggplot(counts_m, aes(x=variable, y=idsort)) + geom_tile(aes(fill=value)) + 

scale_fill_gradient2(expression(log2(Xrcc2mm/WT)), low="blue", mid = "white", high="red", space = "Lab", midpoint = 0, breaks = c(-5,-2,0,2,5)) + 

scale_y_continuous(expand=c(0,0)) + 

scale_x_discrete(expand=c(0,0)) + 

opts(panel.background = theme_blank(), panel.border=theme_blank())

[Dennis Murphy]

Hi:

Just guessing, but variable appears to be a factor with one level
(ratio). Since the extent of the tiles appears to be one horizontal
unit in length, perhaps

scale_x_discrete(limits = c(0.5, 1.5), expand = c(0, 0))

would work for you. Is this 0.8.9 or 0.9.0?

Dennis

> --
> You received this message because you are subscribed to the ggplot2 mailing
> list.
> Please provide a reproducible example: http://gist.github.com/270442
>
> To post: email ggp...@googlegroups.com
> To unsubscribe: email ggplot2+u...@googlegroups.com
> More options: http://groups.google.com/group/ggplot2

[Robert Faryabi]

Hi Dennis,

Thanks for the reply.

Unfortunately, your suggestion doesn't work.

The problem is I loose the graph after changing the limits for x. I tried several values but since I don't know how to get the current value I cannot set the parameters accordingly. 

Here is the result

last_plot() + scale_x_discrete(limits = c(0.5, 1.5), expand=c(0,0))

Warning messages:

1: In min(x) : no non-missing arguments to min; returning Inf

2: In max(x) : no non-missing arguments to max; returning -Inf

3: In min(diff(sort(x))) : no non-missing arguments to min; returning Inf

4: In min(x) : no non-missing arguments to min; returning Inf

5: In max(x) : no non-missing arguments to max; returning -Inf

6: In min(x) : no non-missing arguments to min; returning Inf

7: In max(x) : no non-missing arguments to max; returning -Inf

8: In min(x) : no non-missing arguments to min; returning Inf

9: In max(x) : no non-missing arguments to max; returning -Inf

Thanks,

Robert

On Wed, Mar 14, 2012 at 6:12 PM, Dennis Murphy <djm...@gmail.com> wrote:

Hi:

Just guessing, but variable appears to be a factor with one level
(ratio). Since the extent of the tiles appears to be one horizontal
unit in length, perhaps

scale_x_discrete(limits = c(0.5, 1.5), expand = c(0, 0))

would work for you. Is this 0.8.9 or 0.9.0?

I have version 0.9.0

[Dennis Murphy]

Hi:

Maybe you can take this and run with it to get what you want. Instead
of using a one-level factor that is more than capable of messing
things up, I used geom_segment() instead with a numeric x-axis. Here's
a reproducible example with fake data using ggplot2-0.9.0:

tst <- data.frame(idsort = round(runif(1000, 0, 2000), 0),
value = rnorm(1000, 0, 2))
ggplot(tst, aes(y = idsort, colour = value)) +
geom_segment(aes(x = 1, xend = 0, yend = idsort)) +
scale_colour_gradient2(expression(log2(Xrcc2mm/WT)), low="blue",

mid = "white", high="red",
space = "Lab", midpoint = 0, breaks = c(-5,-2,0,2,5)) +

opts(panel.background = theme_blank(), panel.border=theme_blank())

A couple of things you might try since you're using 0.9.0: (i) use
guide_legend() with the aes.override = argument to increase the
thickness of the lines in the legend (but not in the plot itself);
(ii) the x-labels are rather arbitrary and hence superfluous, so you
could get rid of them altogether.

Dennis