ggethos

The goal of ggethos is to provide a user-frienldy way to plot ethograms using ggplot2.

Installation

At this point, this is an experimental package, and you can only install the development version from GitHub with:

# install.packages("devtools")
devtools::install_github("matiasandina/ggethos")

Once it’s on CRAN, you can install the released version of ggethos from CRAN with:

install.packages("ggethos")

And the development version from GitHub with:

# install.packages("devtools")
devtools::install_github("matiasandina/ggethos")

Example

There are two supported workflows:

1. geom_ethogram() computes ethogram segments for you (quick start).
2. compute_*() lets you precompute segments for debugging or further analysis.

Quick start (stat computes)

ethogram_demo is a small deterministic dataset designed to make outputs easy to verify.

ggplot(ethogram_demo,
       aes(x = seconds, y = subject, behaviour = behaviour, colour = behaviour)) +
  geom_ethogram() +
  facet_wrap(~ trial)

Precompute segments (pipeline)

When you precompute segments, geom_ethogram(stat = "identity") will draw them without recomputing. The computed data keeps the original column names and adds *_end columns plus mapping metadata.

seg <- compute_samples(ethogram_demo,
                       x = seconds,
                       y = subject,
                       behaviour = behaviour,
                       interval = 5)

ggplot() +
  geom_ethogram(data = seg, aes(colour = behaviour), stat = "identity")

Realistic dataset (wombats)

wombats is a larger, more variable dataset with multiple trials.

ggplot(wombats, aes(x = seconds, y = wombat, behaviour = behaviour, color = behaviour)) +
  geom_ethogram() +
  facet_wrap(~ trial)

ggplot(wombats, aes(x = seconds, y = wombat, behaviour = behaviour, color = behaviour)) +
  geom_ethogram(align_trials = TRUE) +
  facet_wrap(~ trial)

Issues

This is a preliminary release and the package is still very much experimental. Please file issues to improve it.