[R] ps or pdf

[Stefan Grosse]

On Monday 31 March 2008 10:17:50 pm Francois Pepin wrote:
FP> Example of row names that are truncated in Illustrator (* denoting 
FP> truncation):
FP> CCDC3*4 (2nd row)
FP> MUC2*0 (3rd row)
FP> MBNL*1 (8th row)

Those are not truncated with me but there are some truncated, this is true. 
Note that if one uses the option paper="special" the papersize will set at 
the plot size, I always would recommend that.

After having a look at your code I tried:

matrix1<-matrix(rnorm(2500),50)

# this is what I use for eps creation normally
postscript('/home/sgrosse/heatmap.ps',paper="special",horizontal=F,onefile=T)
heatmap(matrix1,labRow=row)
dev.off()

# for comparison
pdf('/home/sgrosse/heatmap.pdf',paper='special')
heatmap(matrix1,labRow=row)
dev.off()

what I get with the ps creation is an error:

"Error in plot.new() : outer margins too large (fig.region too small)
Error in par(op) : invalid value specified for graphical parameter "mai""

So I would guess it is a bug in heatmap .

But what you could do is increasing the size of the ps plot:

postscript('/home/sgrosse/heatmap.ps',paper="special",horizontal=F,onefile=T,width=8,height=8)
heatmap(matrix1,labRow=row)
dev.off()

for example works.

FP> The top right cell (26, TXNRD2) is grouped with the cell just below it 
FP> (26, CCDC34). It's more of a curiosity than anything else.

I do not get what you mean here with grouping. 

Cheers
Stefan

(R-2.6.2, Fedora 8)
-- 
Microeconomics
University of Erfurt