How to make Ladder plot? - r

Question

how can you make a ladder plot in R, please.
Here is a data example:
d <- data.frame("Subject" = 1:10,
"Group" = c(rep(1, 6), rep(2, 4)),
"Gender" = c(rep("male" ,2), rep("female", 6), rep("male", 2)),
"Y1" = rnorm(10, 100, 30),
"Y2" = rnorm(10, 3000, 1000))
in this ladderplot:
Need to show those with high Y1 also have high Y2
display the correlations for the "Group" and "Gender" factors
show the Y axis ticks on the left for Y1 and on the right for Y2
The two variables (Y1 and Y2) are to be connected for each subject with solid lines for Group 1 and dotted lines for Group 2, red for males and blue for females.
There is something about package plotrix but I cant seem to find details.

Answer 1

The ggplot2 library is so flexible, I'd use that, instead of looking for a canned routine. Here's code that has the basics of the ladder graph. Look at the existing documentation for things like secondary axes, and changing the scales of color & linetype. I'm leaving this so your post feels like it's asking a focused question (and won't be flagged as a request for a code writing service).
The important step below actually comes before the graphing call. Change your 'wide' format to 'long.
library(magrittr)
library(ggplot2)
set.seed(100)
d <- data.frame(
Subject = 1:10,
Group = c(rep(1, 6), rep(2, 4)),
Gender = c(rep("male" ,2), rep("female", 6), rep("male", 2)),
Y1 = rnorm(10, 100, 30),
Y2 = rnorm(10, 3000, 1000)
)
d_long <- d %>%
tidyr::gather(key=Time, value=Score, -Subject, -Group, -Gender) %>%
dplyr::mutate(
Group = factor(Group)
)
ggplot(d_long, aes(x=Time, y=Score, group=Subject, linetype=Group, color=Gender)) +
geom_line()
Results
> head(d_long)
Subject Group Gender Time Score
1 1 1 male Y1 84.93423
2 2 1 male Y1 103.94593
3 3 1 female Y1 97.63249
4 4 1 female Y1 126.60354
5 5 1 female Y1 103.50914
6 6 1 female Y1 109.55890

Answer 2

Thank you wibeasley for your input! it was greatly helpful. I used however the following code to generate my results.
# Melt dataset for plot:
library(reshape)
melted_data<-melt(d, id.vars=c("Subject","Group","Sex"),measure.vars= c("Y1","Y2"))
melted_data$Group<-as.factor(melted_data$Group)
# calcuate R2 per Group and Sex combination
require(plyr)
func <- function(xx)
{ return(data.frame(R2 = round (cor(xx$Y1, xx$Y2),6)))}
CorrDataset<-ddply(d, .(Group,Sex), func)
# plot:
library(gridExtra)
library(ggplot2)
set.seed(1)
p <-ggplot(melted_data, aes(x=variable, y=value, group=Subject, linetype=Group, color=Sex)) +
geom_line(size=1)
#p <- p + scale_y_continuous(sec.axis = sec_axis(~ scale(.), name = "Y2"))
p+theme(legend.position="top",
axis.line.x = element_line(color="black", size = 2),
axis.line.y = element_line(color="black", size = 2),
panel.grid.major = element_blank(),
panel.grid.minor = element_blank(),
panel.border = element_blank(),
panel.background = element_blank())+
annotation_custom(tableGrob(CorrDataset,rows = rownames(CorrDataset)), ymin=4000, ymax=4000)
Please note: For the second Y axis, its commented out in the code above.