Visualizing multiple datasets with scatterbar

Visualizing multiple datasets with scatterbar

This tutorial demonstrates how to visualize multiple datasets together, utilizing the package patchwork.

Below we can load in our datasets provided by scatterbar and create the respective scatterbars using those datasets and save them to a variable.

library(scatterbar)
library(ggplot2)

data("mOB")
data("adult_mouse_brain_ffpe")

# Basic scatterbar plot with default settings
p1 <- scatterbar(mOB$data, mOB$xy) + coord_fixed()
#> Calculated size_x: 1.24034734589208
#> Calculated size_y: 0.930260509419063
#> Applied padding_x: 0
#> Applied padding_y: 0
p2 <- scatterbar(adult_mouse_brain_ffpe$prop, adult_mouse_brain_ffpe$pos) + coord_fixed()
#> Calculated size_x: 302.260275014085
#> Calculated size_y: 323.465991707814
#> Applied padding_x: 0
#> Applied padding_y: 0

We can then load in patchwork and visualize both scatterbars in one plot.

library(patchwork)
p1 + p2