How to subset data in a ggplot panel chart? - r

Question

I am trying create a panel chart in ggplot with four variables which all have their own scale for the y axis. I can get the structure of the panel chart to work but am having trouble actually getting each data set onto the gird. I have been following a script I found online. See below however I am getting the following error when I try and use the subset function further down in the script.
Error in .(variable == "Count") : could not find function "."
#load data
#Data source: data analysis-gullies > R Stats Input > Panel Chart
df <- read.csv(file.choose(), header = T)
View(df)
#load library
library(ggplot2)
library(reshape2)
dfm <- melt(df, id.vars =c("Interval"))
View(dfm)
test <- ggplot(dfm, aes(Interval, value, ymin = 0,
ymax = value, colour = "grey20"))+ scale_colour_identity() +
xlim(5,1115)+ facet_grid(variable ~ ., scales = "free", as.table = FALSE)+
theme_bw() + theme(panel.spacing = unit(0, "lines"), axis.title.x = element_blank(),
axis.title.y = element_text())
test
test1 <- test + geom_col(subset = .(variable == "Count"))
test2 <- test1 + geom_col(subset = .(variable == "Length"))
test3 <- test2 + geom_col(subset = .(variable == "Area"))
test4 <- test3 + geom_col(subset = .(variable == "Volume"))

Answer 1

You can use the patchwork package to merge individual ggplot2 objects to get individual axes for each panel:
library(tidyverse)
library(patchwork)
iris %>%
nest(-Species) %>%
mutate(
plt = data %>% map2(Species, ~ {
.x %>%
ggplot(aes(Sepal.Width, Sepal.Length)) +
geom_point() +
labs(title = .y)
})
) %>%
pull(plt) %>%
wrap_plots()
You can also add logic to plot different plots per panel:
library(tidyverse)
library(patchwork)
iris %>%
nest(-Species) %>%
mutate(
plt = data %>% map2(Species, ~ {
if(.y == "setosa") {
.x %>%
ggplot(aes(Sepal.Width, Sepal.Length)) +
geom_point() +
labs(title = .y)
} else {
.x %>%
ggplot(aes(Sepal.Width, Sepal.Length)) +
geom_line() +
labs(title = .y)
}
})
) %>%
pull(plt) %>%
wrap_plots()
If the panel plots are very different from each other (e.g. different variables for the x and y axes), it is recommended to create each plot individually and then call wrap_plots of all the plot objects:
plt1 <- qplot(Sepal.Length, Sepal.Width, data = iris, geom = "point")
plt2 <- qplot(Petal.Length, Petal.Width, data = iris, geom = "line")
wrap_plots(plt1, plt2, nrow = 1)