Related
I'm new to flowCore + R. I would like to mimic a histogram plot after gating that can be manually done in FlowJo software. I got something similar but it doesn't look quite right because it is a "density" plot and is shifted. How can I get the x axis to shift over and look similar to how FlowJo outputs the plot? I tried reading this document but couldn't find a plot similar to the one in FlowJo: howtoflowcore Appreciate any guidance. Thanks.
code snippet:
library(flowCore)
parentpath <- "/parent/path"
subfolder <- "Sample 1"
fcs_files <- list.files(paste0(parentpath, subfolder), pattern = ".fcs")
fs <- read.flowSet(fcs_files)
rect.g <- rectangleGate(filterId = "main",list("FSC-A" = c(1e5, 2e5), "SSC-A" = c(3e4,1e5)))
fs_sub <- Subset(fs, rect.g)
p <- ggcyto(fs_sub[[15]], aes(x= `UV-379-A`)) +
geom_density(fill='black', alpha = 0.4) +
ggcyto_par_set(limits = list(x = c(-1e3, 5e4), y = c(0, 6e-5)))
p
FlowJo output:
R FlowCore output:
The reason that for the "shift" is that the x axis is logarithmic (base 10) in the flowJo graph. To achieve the same result in R, add
+ scale_x_log10()
after the existing code. This might interact weirdly with the axis limits you've set, so bare that in mind.
To make the y-axis "count" rather than density, you can change the first line of your ggcyto() call to:
aes(x= `UV-379-A`, y = after_stat(count))
Let me know if that works - I don't have your data to hand so that's all from memory!
For any purely aesthetic changes, they are relatively easy to look up.
I am not entirely sure if these kind of questions are allowed at SO, as I have no reproducible data at the moment.
My question is in regards to how one might go about creating a scatterplot of raster images in R. I am not familiar with any packages that allow you to do this. This is the only example I have come across so far in my search. Essentially, this is what I would like to do, however, I am wondering if it's possible for R to simply take the input data and plot the image rather than be fed coordinates in my plot area.
My end goal is to create raster image scatterplots of sports teams using their logos instead of labels. My first thought is to create a data frame including team name, X variable, Y variable, and .png image URL location.
Here is an example of what I am ultimately hoping to do. I'm not sure what program the OP uses, but obviously I would like to do something like this in R.
UPDATE
With the help of Greg Snow's suggestion, I was able to reproduce his example with my own logos.
The the my.symbols and ms.image functions in the TeachingDemos package are one possible starting place. There is an example on the help page for ms.image that shows how to use the R logo as the plotting symbol. Currently it only does one image at a time, so you could either start with a blank plot and loop through the set of images, or a wrapper function can be written that takes a list of images and an indicator of which to plot. Here is a first stab at a wrapper function:
ms.image2 <- function(imgs, transpose=TRUE,
which=1, ...) {
ms.image(imgs[[which]], transpose=transpose, ...)
}
Then we can create a list of images with code like:
require(png)
img1 <- readPNG(system.file("img", "Rlogo.png", package="png"))
logos <- list( img1, img1[76:1,,], img1[,100:1,],
img1[76:1,100:1,], img1[,,c(3:1,4)])
These are all variations on the logo, but for your example you could pass a vector of file names of the .png files to lapply to produce a similar list.
Now we can run my.symbols like this (though obviously you will use real data rather than random numbers for the locations):
my.symbols( runif(10), runif(10), ms.image2,
MoreArgs=list(imgs=logos), which=rep(1:5,2),
inches=0.3, symb.plots=TRUE, add=FALSE)
And that produces a plot along the lines of your example:
Edit
For a speed up you can use rasterImage, here is some new code that ran in about half the time as the above (compared using microbenchmark):
ms.rasterImage <- function(imgs, which=1, ...) {
rasterImage(imgs[[which]], -1, -1, 1, 1)
}
logos2 <- list(as.raster(img1), as.raster(img1[76:1,,]),
as.raster(img1[,100:1,]),
as.raster(img1[76:1,100:1,]),
as.raster(img1[,,c(3:1,4)])
)
my.symbols( runif(10), runif(10), ms.rasterImage,
MoreArgs=list(imgs=logos2), which=rep(1:5,2),
inches=0.3, symb.plots=TRUE, add=FALSE)
And here is some code using ggplot2 based on the link in the comment above, but using the list of logos:
ggplot(mtcars, aes(mpg, wt)) +
mapply(function(xx, yy, i)
annotation_raster(logos[[i]], xmin=xx-1, xmax=xx+1, ymin=yy-0.2, ymax=yy+0.2),
mtcars$mpg, mtcars$wt, mtcars$gear-2)
And mainly for my curiosity, here are the timings:
> microbenchmark(
+ my.symbols( mtcars$mpg, mtcars$wt, ms.image2,
+ MoreArgs=list(imgs=logos), which=mtcars$gear-2,
+ inches=0.3, symb.plots=TRUE, add=FALSE),
+ my.symbols( mtcars$mpg, mtcars$wt, ms.rasterImage,
+ MoreArgs=list(imgs=logos2), which=mtcars$gear-2,
+ inches=0.3, symb.plots=TRUE, add=FALSE),
+ plot(ggplot(mtcars, aes(mpg, wt)) +
+ mapply(function(xx, yy, i)
+ annotation_raster(logos[[i]], xmin=xx-1, xmax=xx+1, ymin=yy-0.2, ymax=yy+0.2),
+ mtcars$mpg, mtcars$wt, mtcars$gear-2) )
+ )
Unit: milliseconds
min lq mean median uq max neval cld
ms.image 518.9137 530.5549 661.9333 545.3890 751.7116 1737.7430 100 b
ms.rasterImage 158.7097 162.4493 244.6673 171.6103 381.6499 544.1656 100 a
ggplot2 478.3005 606.3831 896.8793 772.7210 1359.8888 1714.5647 100 c
I'm attempting to step through a dataset and create a histogram and summary table for each factor and save the output as a .svg . The histogram is created using ggplot2 and the summary table using summary().
I have successfully used the code below to save the output to a single .pdf with each page containing the relevant histogram/table. However, when I attempt to save each histogram/table combo into a set of .svg images using ggsave only the ggplot histogram is showing up in the .svg. The table is just white space.
I've tried using dev.copy Cairo and svg but all end up with the same result: Histogram renders, but table does not. If I save the image as a .png the table shows up.
I'm using the iris data as a reproducible dataset. I'm not using R-Studio which I saw was causing some "empty plot" grief for others.
#packages used
library(ggplot2)
library(gridExtra)
library(gtable)
library(Cairo)
#Create iris histogram plot
iris.hp<-ggplot(data=iris, aes(x=Sepal.Length)) +
geom_histogram(binwidth =.25,origin=-0.125,
right = TRUE,col="white", fill="steelblue4",alpha=1) +
labs(title = "Iris Sepal Length")+
labs(x="Sepal Length", y="Count")
iris.list<-by(data = iris, INDICES = iris$Species, simplify = TRUE,FUN = function(x)
{iris.hp %+% x + ggtitle(unique(x$Species))})
#Generate list of data to create summary statistics table
sum.str<-aggregate(Sepal.Length~Species,iris,summary)
spec<-sum.str[,1]
spec.stats<-sum.str[,2]
sum.data<-data.frame(spec,spec.stats)
sum.table<-tableGrob(sum.data)
colnames(sum.data) <-c("species","sep.len.min","sep.len.1stQ","sep.len.med",
"sep.len.mean","sep. len.3rdQ","sep.len.max")
table.list<-by(data = sum.data, INDICES = sum.data$"species", simplify = TRUE,
FUN = function(x) {tableGrob(x)})
#Combined histogram and summary table across multiple plots
multi.plots<-marrangeGrob(grobs=(c(rbind(iris.list,table.list))),
nrow=2, ncol=1, top = quote(paste(iris$labels$Species,'\nPage', g, 'of',pages)))
#bypass the class check per #baptiste
ggsave <- ggplot2::ggsave; body(ggsave) <- body(ggplot2::ggsave)[-2]
#
for(i in 1:3){
multi.plots<-marrangeGrob(grobs=(c(rbind(iris.list[i],table.list[i]))),
nrow=2, ncol=1,heights=c(1.65,.35),
top = quote(paste(iris$labels$Species,'\nPage', g, 'of',pages)))
prefix<-unique(iris$Species)
prefix<-prefix[i]
filename<-paste(prefix,".svg",sep="")
ggsave(filename,multi.plots)
#dev.off()
}
Edit removed theme tt3 that #rawr referenced. It was accidentally left in example code. It was not causing the problem, just in case anyone was curious.
Edit: Removing previous answer regarding it working under 32bit install and not x64 install because that was not the problem. Still unsure what was causing the issue, but it is working now. Leaving the info about grid.export as it may be a useful alternative for someone else.
Below is the loop for saving the .svg's using grid.export(), although I was having some text formatting issues with this (different dataset).
for(i in 1:3){
multi.plots<-marrangeGrob(grobs=(c(rbind(iris.list[i],table.list[i]))),
nrow=2, ncol=1,heights=c(1.65,.35), top =quote(paste(iris$labels$Species,'\nPage', g,
'of',pages)))
prefix<-unique(iris$Species)
prefix<-prefix[i]
filename<-paste(prefix,".svg",sep="")
grid.draw(multi.plots)
grid.export(filename)
grid.newpage()
}
EDIT: As for using arrangeGrob per #baptiste's comment. Below is the updated code. I was incorrectly using the single brackets [] for the returned by list, so I switched to the correct double brackets [[]] and used grid.draw to on the ggsave call.
for(i in 1:3){
prefix<-unique(iris$Species)
prefix<-prefix[i]
multi.plots<-grid.arrange(arrangeGrob(iris.list[[i]],table.list[[i]],
nrow=2,ncol=1,top = quote(paste(iris$labels$Species))))
filename<-paste(prefix,".svg",sep="")
ggsave(filename,grid.draw(multi.plots))
}
I would like to add LaTeX typesetting to elements of plots in R (e.g: the title, axis labels, annotations, etc.) using either the combination of base/lattice or with ggplot2.
Questions:
Is there a way to get LaTeX into plots using these packages, and if so, how is it done?
If not, are there additional packages needed to accomplish this.
For example, in Python matplotlib compiles LaTeX via the text.usetex packages as discussed here: http://www.scipy.org/Cookbook/Matplotlib/UsingTex
Is there a similar process by which such plots can be generated in R?
The CRAN package latex2exp contains a TeX function that translate LaTeX formulas to R's plotmath expressions. You can use it anywhere you could enter mathematical annotations, such as axis labels, legend labels, and general text.
For example:
x <- seq(0, 4, length.out=100)
alpha <- 1:5
plot(x, xlim=c(0, 4), ylim=c(0, 10),
xlab='x', ylab=TeX(r'($\alpha x^\alpha$, where $\alpha \in \{1 \ldots 5\}$)'),
type='n', main=TeX(r'(Using $\LaTeX$ for plotting in base graphics!)', bold=TRUE))
for (a in alpha) {
lines(x, a*x^a, col=a)
}
legend('topleft',
legend=TeX(sprintf(r'($\alpha = %d$)', alpha)),
lwd=1,
col=alpha)
produces this plot.
Here's an example using ggplot2:
q <- qplot(cty, hwy, data = mpg, colour = displ)
q + xlab(expression(beta +frac(miles, gallon)))
As stolen from here, the following command correctly uses LaTeX to draw the title:
plot(1, main=expression(beta[1]))
See ?plotmath for more details.
You can generate tikz code from R:
http://r-forge.r-project.org/projects/tikzdevice/
Here's something from my own Lab Reports.
tickzDevice exports tikz images for LaTeX
Note, that in certain cases "\\" becomes "\" and "$" becomes "$\" as in the following R code: "$z\\frac{a}{b}$" -> "$\z\frac{a}{b}$\"
Also xtable exports tables to latex code
The code:
library(reshape2)
library(plyr)
library(ggplot2)
library(systemfit)
library(xtable)
require(graphics)
require(tikzDevice)
setwd("~/DataFolder/")
Lab5p9 <- read.csv (file="~/DataFolder/Lab5part9.csv", comment.char="#")
AR <- subset(Lab5p9,Region == "Forward.Active")
# make sure the data names aren't already in latex format, it interferes with the ggplot ~ # tikzDecice combo
colnames(AR) <- c("$V_{BB}[V]$", "$V_{RB}[V]$" , "$V_{RC}[V]$" , "$I_B[\\mu A]$" , "IC" , "$V_{BE}[V]$" , "$V_{CE}[V]$" , "beta" , "$I_E[mA]$")
# make sure the working directory is where you want your tikz file to go
setwd("~/TexImageFolder/")
# export plot as a .tex file in the tikz format
tikz('betaplot.tex', width = 6,height = 3.5,pointsize = 12) #define plot name size and font size
#define plot margin widths
par(mar=c(3,5,3,5)) # The syntax is mar=c(bottom, left, top, right).
ggplot(AR, aes(x=IC, y=beta)) + # define data set
geom_point(colour="#000000",size=1.5) + # use points
geom_smooth(method=loess,span=2) + # use smooth
theme_bw() + # no grey background
xlab("$I_C[mA]$") + # x axis label in latex format
ylab ("$\\beta$") + # y axis label in latex format
theme(axis.title.y=element_text(angle=0)) + # rotate y axis label
theme(axis.title.x=element_text(vjust=-0.5)) + # adjust x axis label down
theme(axis.title.y=element_text(hjust=-0.5)) + # adjust y axis lable left
theme(panel.grid.major=element_line(colour="grey80", size=0.5)) +# major grid color
theme(panel.grid.minor=element_line(colour="grey95", size=0.4)) +# minor grid color
scale_x_continuous(minor_breaks=seq(0,9.5,by=0.5)) +# adjust x minor grid spacing
scale_y_continuous(minor_breaks=seq(170,185,by=0.5)) + # adjust y minor grid spacing
theme(panel.border=element_rect(colour="black",size=.75))# border color and size
dev.off() # export file and exit tikzDevice function
Here's a cool function that lets you use the plotmath functionality, but with the expressions stored as objects of the character mode. This lets you manipulate them programmatically using paste or regular expression functions. I don't use ggplot, but it should work there as well:
express <- function(char.expressions){
return(parse(text=paste(char.expressions,collapse=";")))
}
par(mar=c(6,6,1,1))
plot(0,0,xlim=sym(),ylim=sym(),xaxt="n",yaxt="n",mgp=c(4,0.2,0),
xlab="axis(1,(-9:9)/10,tick.labels,las=2,cex.axis=0.8)",
ylab="axis(2,(-9:9)/10,express(tick.labels),las=1,cex.axis=0.8)")
tick.labels <- paste("x >=",(-9:9)/10)
# this is what you get if you just use tick.labels the regular way:
axis(1,(-9:9)/10,tick.labels,las=2,cex.axis=0.8)
# but if you express() them... voila!
axis(2,(-9:9)/10,express(tick.labels),las=1,cex.axis=0.8)
I did this a few years ago by outputting to a .fig format instead of directly to a .pdf; you write the titles including the latex code and use fig2ps or fig2pdf to create the final graphic file. The setup I had to do this broke with R 2.5; if I had to do it again I'd look into tikz instead, but am including this here anyway as another potential option.
My notes on how I did it using Sweave are here: http://www.stat.umn.edu/~arendahl/computing
I just have a workaround. One may first generate an eps file, then convert it back to pgf using the tool eps2pgf. See http://www.texample.net/tikz/examples/eps2pgf/
h <- rnorm(mean = 5, sd = 1, n = 1000)
hist(h, main = expression(paste("Sampled values, ", mu, "=5, ", sigma,
"=1")))
Taken from a very help article here https://stats.idre.ucla.edu/r/codefragments/greek_letters/
You can use the following, for example:
title(sub=TeX(sprintf(paste("Some latex symbols are ", r'(\lambda)', "and", r'(\alpha)'))))
Just remember to enclose LaTeX expressions in paste() using r'()'
You can also add named objects in the paste() function. E.g.,
lambda_variable <- 3
title(sub=TeX(sprintf(paste(r'(\lambda=)', lambda_variable))))
Not sure if there are better ways to do this, but the above worked for me :)
I would like to add LaTeX typesetting to elements of plots in R (e.g: the title, axis labels, annotations, etc.) using either the combination of base/lattice or with ggplot2.
Questions:
Is there a way to get LaTeX into plots using these packages, and if so, how is it done?
If not, are there additional packages needed to accomplish this.
For example, in Python matplotlib compiles LaTeX via the text.usetex packages as discussed here: http://www.scipy.org/Cookbook/Matplotlib/UsingTex
Is there a similar process by which such plots can be generated in R?
The CRAN package latex2exp contains a TeX function that translate LaTeX formulas to R's plotmath expressions. You can use it anywhere you could enter mathematical annotations, such as axis labels, legend labels, and general text.
For example:
x <- seq(0, 4, length.out=100)
alpha <- 1:5
plot(x, xlim=c(0, 4), ylim=c(0, 10),
xlab='x', ylab=TeX(r'($\alpha x^\alpha$, where $\alpha \in \{1 \ldots 5\}$)'),
type='n', main=TeX(r'(Using $\LaTeX$ for plotting in base graphics!)', bold=TRUE))
for (a in alpha) {
lines(x, a*x^a, col=a)
}
legend('topleft',
legend=TeX(sprintf(r'($\alpha = %d$)', alpha)),
lwd=1,
col=alpha)
produces this plot.
Here's an example using ggplot2:
q <- qplot(cty, hwy, data = mpg, colour = displ)
q + xlab(expression(beta +frac(miles, gallon)))
As stolen from here, the following command correctly uses LaTeX to draw the title:
plot(1, main=expression(beta[1]))
See ?plotmath for more details.
You can generate tikz code from R:
http://r-forge.r-project.org/projects/tikzdevice/
Here's something from my own Lab Reports.
tickzDevice exports tikz images for LaTeX
Note, that in certain cases "\\" becomes "\" and "$" becomes "$\" as in the following R code: "$z\\frac{a}{b}$" -> "$\z\frac{a}{b}$\"
Also xtable exports tables to latex code
The code:
library(reshape2)
library(plyr)
library(ggplot2)
library(systemfit)
library(xtable)
require(graphics)
require(tikzDevice)
setwd("~/DataFolder/")
Lab5p9 <- read.csv (file="~/DataFolder/Lab5part9.csv", comment.char="#")
AR <- subset(Lab5p9,Region == "Forward.Active")
# make sure the data names aren't already in latex format, it interferes with the ggplot ~ # tikzDecice combo
colnames(AR) <- c("$V_{BB}[V]$", "$V_{RB}[V]$" , "$V_{RC}[V]$" , "$I_B[\\mu A]$" , "IC" , "$V_{BE}[V]$" , "$V_{CE}[V]$" , "beta" , "$I_E[mA]$")
# make sure the working directory is where you want your tikz file to go
setwd("~/TexImageFolder/")
# export plot as a .tex file in the tikz format
tikz('betaplot.tex', width = 6,height = 3.5,pointsize = 12) #define plot name size and font size
#define plot margin widths
par(mar=c(3,5,3,5)) # The syntax is mar=c(bottom, left, top, right).
ggplot(AR, aes(x=IC, y=beta)) + # define data set
geom_point(colour="#000000",size=1.5) + # use points
geom_smooth(method=loess,span=2) + # use smooth
theme_bw() + # no grey background
xlab("$I_C[mA]$") + # x axis label in latex format
ylab ("$\\beta$") + # y axis label in latex format
theme(axis.title.y=element_text(angle=0)) + # rotate y axis label
theme(axis.title.x=element_text(vjust=-0.5)) + # adjust x axis label down
theme(axis.title.y=element_text(hjust=-0.5)) + # adjust y axis lable left
theme(panel.grid.major=element_line(colour="grey80", size=0.5)) +# major grid color
theme(panel.grid.minor=element_line(colour="grey95", size=0.4)) +# minor grid color
scale_x_continuous(minor_breaks=seq(0,9.5,by=0.5)) +# adjust x minor grid spacing
scale_y_continuous(minor_breaks=seq(170,185,by=0.5)) + # adjust y minor grid spacing
theme(panel.border=element_rect(colour="black",size=.75))# border color and size
dev.off() # export file and exit tikzDevice function
Here's a cool function that lets you use the plotmath functionality, but with the expressions stored as objects of the character mode. This lets you manipulate them programmatically using paste or regular expression functions. I don't use ggplot, but it should work there as well:
express <- function(char.expressions){
return(parse(text=paste(char.expressions,collapse=";")))
}
par(mar=c(6,6,1,1))
plot(0,0,xlim=sym(),ylim=sym(),xaxt="n",yaxt="n",mgp=c(4,0.2,0),
xlab="axis(1,(-9:9)/10,tick.labels,las=2,cex.axis=0.8)",
ylab="axis(2,(-9:9)/10,express(tick.labels),las=1,cex.axis=0.8)")
tick.labels <- paste("x >=",(-9:9)/10)
# this is what you get if you just use tick.labels the regular way:
axis(1,(-9:9)/10,tick.labels,las=2,cex.axis=0.8)
# but if you express() them... voila!
axis(2,(-9:9)/10,express(tick.labels),las=1,cex.axis=0.8)
I did this a few years ago by outputting to a .fig format instead of directly to a .pdf; you write the titles including the latex code and use fig2ps or fig2pdf to create the final graphic file. The setup I had to do this broke with R 2.5; if I had to do it again I'd look into tikz instead, but am including this here anyway as another potential option.
My notes on how I did it using Sweave are here: http://www.stat.umn.edu/~arendahl/computing
I just have a workaround. One may first generate an eps file, then convert it back to pgf using the tool eps2pgf. See http://www.texample.net/tikz/examples/eps2pgf/
h <- rnorm(mean = 5, sd = 1, n = 1000)
hist(h, main = expression(paste("Sampled values, ", mu, "=5, ", sigma,
"=1")))
Taken from a very help article here https://stats.idre.ucla.edu/r/codefragments/greek_letters/
You can use the following, for example:
title(sub=TeX(sprintf(paste("Some latex symbols are ", r'(\lambda)', "and", r'(\alpha)'))))
Just remember to enclose LaTeX expressions in paste() using r'()'
You can also add named objects in the paste() function. E.g.,
lambda_variable <- 3
title(sub=TeX(sprintf(paste(r'(\lambda=)', lambda_variable))))
Not sure if there are better ways to do this, but the above worked for me :)