Thanks for your help, Rui! That works and will save me a lot of trouble.
--Kelly
-----Original Message-----
From: Rui Barradas [mailto:ruipbarradas at sapo.pt]
Sent: Tuesday, July 10, 2012 2:24 AM
To: Vining, Kelly
Cc: r-help at r-project.org
Subject: Re: [R] boxplot with "cut"
Hello,
Maybe this iss what you're looking for. GD is your data.frame.
multi.boxplot <- function(x, by, ...){
x <- as.data.frame(x)
sp <- split(x, by)
len <- length(sp) - 1
n <- ncol(x)
n1 <- n + 1
boxplot(x[[ 1 ]] ~ by, at = 0:len*n1 + 1,
xlim = c(0, (len + 1)*n1), ylim = range(unlist(x)), xaxt = "n", ...)
for(i in seq_len(n)[-1])
boxplot(x[[i]] ~ by, at = 0:len*n1 + i, xaxt = "n", add = TRUE, ...)
axis(1, at = 0:len*n1 + n1/2, labels = names(sp), tick = TRUE) }
cols <- grep("ReadCount", names(GD))
multi.boxplot(GD[, cols], cut(GD$GeneDensity, breaks=10))
If this is it and you don't like those x-axis tick lables, use
as.integer(cut(...etc...)).
Hope this helps,
Rui Barradas
Em 09-07-2012 20:51, Vining, Kelly escreveu:> Dear UseRs,
> I'm making box plots from a data set that looks like this:
>
>
> Chr Start End GeneDensity ReadCount_Explant ReadCount_Callus
ReadCount_Regen
> 1 1 1 10000 107.82 1.243 1.047
1.496
> 2 1 10001 20000 202.50 0.835 0.869
0.456
> 3 1 20001 30000 158.80 1.813 1.529
1.131
> 4 1 30001 40000 100.53 1.731 1.752
1.610
> 5 1 40001 50000 100.53 3.056 2.931
3.631
> 6 1 50001 60000 100.53 1.960 2.013
2.459
>
> I'm breaking the "GeneDensity" column into deciles, then
making a box plot of the relationship between the GeneDensity parameter and each
of the three "ReadCount" columns. Here's an example of one of my
boxplot commands:
>
> boxplot(GeneDensity$ReadCount_Explant ~
> cut(GeneDensitySorted$GeneDensity, breaks=10), ylim=c(0,40),
> ylab="RPKM", xlab="GENE DENSITY (LOW -> HIGH)",
main="INTERNODE
> EXPLANT")
>
> Right now, I'm making three separate graphs: one for each of the three
"ReadCount" columns. I'd like to put all three sets on one graph,
so that each decile is represented by three boxes, one for each ReadCount
category, but don't know how to make that work. I tried this:
>
> boxplot(GeneDensitySorted$ReadCount_Explant ~
> cut(GeneDensitySorted$GeneDensity, breaks=10),
> GeneDensitySorted$ReadCount_Callus ~
> cut(GeneDensitySorted$GeneDensity, breaks=10),
> GeneDensitySorted$ReadCount_Regen ~ cut(GeneDensitySorted$GeneDensity,
> breaks=10), ylim=c(0,40), ylab="RPKM", xlab="GENE DENSITY
(LOW ->
> HIGH)")
>
> Not surprisingly, I got this error:
>
> Error in as.data.frame.default(data) :
> cannot coerce class '"formula"' into a data.frame
>
> Does anyone know how to accomplish this box plot?
>
> Any help is much appreciated.
>
> --Kelly V.
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>
