Skip to content

ggplot2 reorder factors for faceting

4 messages · Iain Gallagher, Kenneth Takagi, ONKELINX, Thierry +1 more

Iain Gallagher
# 
Dear List

I am trying to draw a heatmap using ggplot2. In this heatmap I have faceted my data by 'infection' of which I have four. These four infections break down into two types and I would like to reorder the 'infection' column of my data to reflect this. 

Toy example below:

library(ggplot2)

# test data for ggplot reordering
genes <- (rep (c(rep('a',4), rep('b',4), rep('c',4), rep('d',4), rep('e',4), rep('f',4)) ,4))
fcData <- rnorm(96)
times <- rep(rep(c(2,6,24,48),6),4)
infection <- c(rep('InfA', 24), rep('InfB', 24), rep('InfC', 24), rep('InfD', 24))
infType <- c(rep('M', 24), rep('D',24), rep('M', 24), rep('D', 24))

# data is long format for ggplot2
plotData <- as.data.frame(cbind(genes, as.numeric(fcData), as.numeric(times), infection, infType))

hp2 <- ggplot(plotData, aes(factor(times), genes)) + geom_tile(aes(fill = scale(as.numeric(fcData)))) + facet_wrap(~infection, ncol=4)

# set scale
hp2 <- hp2 + scale_fill_gradient2(name=NULL, low="#0571B0", mid="#F7F7F7", high="#CA0020", midpoint=0, breaks=NULL, labels=NULL, limits=NULL, trans="identity") 

# set up text (size, colour etc etc)
hp2 <- hp2 + labs(x = "Time", y = "") + scale_y_discrete(expand = c(0, 0)) + opts(axis.ticks = theme_blank(), axis.text.x = theme_text(size = 10, angle = 360, hjust = 0, colour = "grey25"), axis.text.y = theme_text(size=10, colour = 'gray25'))

hp2 <- hp2 + theme_bw()

In the resulting plot I would like infections infA and infC plotted next to each other and likewise for infB and infD. I have a column in the data - infType - which I could use to reorder the infection column but so far I have no luck getting this to work.

Could someone give me a pointer to the best way to reorder the infection factor and accompanying data into the order I would like?

Best

iain
R version 2.13.2 (2011-09-30)
Platform: x86_64-pc-linux-gnu (64-bit)

locale:
?[1] LC_CTYPE=en_GB.utf8?????? LC_NUMERIC=C???????????? 
?[3] LC_TIME=en_GB.utf8??????? LC_COLLATE=en_GB.utf8??? 
?[5] LC_MONETARY=C???????????? LC_MESSAGES=en_GB.utf8?? 
?[7] LC_PAPER=en_GB.utf8?????? LC_NAME=C??????????????? 
?[9] LC_ADDRESS=C????????????? LC_TELEPHONE=C?????????? 
[11] LC_MEASUREMENT=en_GB.utf8 LC_IDENTIFICATION=C????? 

attached base packages:
[1] grid????? stats???? graphics? grDevices utils???? datasets? methods? 
[8] base???? 

other attached packages:
[1] ggplot2_0.8.9 proto_0.3-9.2 reshape_0.8.4 plyr_1.6???? 

loaded via a namespace (and not attached):
[1] digest_0.5.0 tools_2.13.2
Kenneth Takagi
# 
Hi, Iain-



You might want to have a look at 

?relevel



i.e.? plotData$infection<-relevel(plotData$infection, ref = 'InfC')



Ken
On 11/08/11, Iain Gallagher wrote:
ONKELINX, Thierry
# 
Dear Iain,

RSiteSearch("ggplot2 reorder", restrict = c("Rhelp10", "Rhelp08")) gives you the solution.

Best regards,

Thierry
Dennis Murphy
# 
Hi:

(1) Here is one way to reorganize the levels of a factor:
plotData[['infection']] <- factor(plotData[['infection']],
                               levels = c('InfA', 'InfC', 'InfB', 'InfD'))

Do this ahead of the call to ggplot(), preferably after plotData is defined.

relevel() resets the baseline category of a factor, but here you want
to make multiple changes.

(2) You probably want a better title for the legend. Assuming you want
'Scale' as the title, you can add the following to labs:

labs(..., fill = 'Scale')

HTH,
Dennis


On Tue, Nov 8, 2011 at 3:51 AM, Iain Gallagher
<iaingallagher at btopenworld.com> wrote: