passing a character generated using paste() to mtext()

Question

I want to trying to wtire a function that will generate heatmaps by calling heatmap.2. Among other things, I want to generate the title of the plots by calling add.expr, for example:

add.expr=c(mtext(text=titlestring, side=3, line=4, cex=1)

With titlestring being a charqacter vector passed to the function by another function:

titlestring<-paste("Mean bin methylation",samplename, "on 5kb flanked CpG Island promoters in mm9")

When I try to run my function I get the following error:

Error in as.graphicsAnnot(text) : object 'titlestring' not found I do know that titlestring is defined in the scope of my funciton , as I tested this using print() I thouhgt that the problem my be related to the fact the mtext() expects an exprssion object, so I coerced titlestring to an expression using as.expression(). But I still get this error.

Any idea what might be the problem?

Thanks in advance

Dolev Rahat

Your titlestring command gives an error because samplename is not known. Why not use something like main=titlestring in the heatmap call? — Ruthger Righart, Feb 25 '15 at 12:29
@RuthgerRighart: Thanks. Using main is indeed a possibility, but since I want to add a secondary title I still prefer to use add.expr() because it is more fleixble. samplename is a character vector geneated by my calling function based on one of its arguments, so it is defined (as I stated, I also verified that titlestring is visible to the internal function). Is there no possibilty to call mtext() with character vectors passed from another function? — user1614062, Feb 25 '15 at 14:14
I am not sure about what you want exactly, it would be good to have a data example. Please find a possible workaround in the answer box, hope that helps ! — Ruthger Righart, Feb 25 '15 at 14:42
@RuthgerRighart: I added an example of what I want below. However, I think the difference between what I want to do using add.expr() and what I get using main is not that crucial, so I will not waste any furhter time on this. Thank you very much for all your help! — user1614062, Feb 25 '15 at 15:58

score 0 · Accepted Answer · answered Feb 25 '15 at 14:40

If your main question is about adding a secondary title using a function you could use main = c(titlestring, substring), where you can change the vector strings as you like. The example is taken from the mtcars set (it can be found using ?heatmap).

require(graphics) 
require(grDevices)

x  <- as.matrix(mtcars)
rc <- rainbow(nrow(x), start = 0, end = .3)
cc <- rainbow(ncol(x), start = 0, end = .3)

titlestring<-c("Mtcars dataset")
substring<-c("First example")
par(cex.main=0.9)
heatmap(x, Colv = NA, col = cm.colors(256), scale = "column",
    RowSideColors = rc, margins = c(5,10),
    xlab = "specification variables", ylab =  "Car Models",
    main = c(titlestring, substring))

enter image description here )

score 0 · Answer 2 · answered Feb 25 '15 at 15:56

Not that worth all this fuss, but Ideally i would like to be able to have different font sizes for the main and secndary title, and also to control thier position. I could do this if I added the title with add.expr() using litterals, for example:

 heatmap.2 (as.matrix(matToPlot[,(1:totCols)]),
         Rowv=FALSE, Colv=FALSE, dendrogram="none", 
         breaks=seq(0,1,by=1/length(methColors)),
         col=methColors, trace="none",colsep=colsep,sepcolor=c("sky blue"),
         #main=paste(titlestring,subtitlestring,chromstring,sep="\n"),
         add.expr=c(mtext(text="Read coverage of Dox Plus on 5kb flanked CpG
          Island promoters in mm9", side=3, line=4, cex=1),
         mtext(text="Island set represents island for which ChIP data is avilable
          and is order by mean(H3K4me3)-mean(H3K27me3) within islands", side=3, line=2, cex=0.8),
         mtext(text="H3K27me3", side=1, adj=0.125, line=3),
         mtext(text="H3K4me3", side=1, adj=0.375, line=3),
         mtext(text="H3K4me1", side=1, adj=0.625, line=3),
         mtext(text="Methylation", side=1,adj=0.9, line=3)))

Whereas using main I get a result which I can live with, but which is not as good (compare these two plots):

plot generated using add.expr

plot generated using main

passing a character generated using paste() to mtext()

2 Answers2