In Middleton-Lab/MuscleTernary: Plot Muscle Resultants in Ternary Space

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)

To animate a ternary plot, you need two sets of points to interpolate between (the start and end points). Here we use the Alligator data for the adult AL_008 specimen and the juvenile AL_031 specimen.

Load AL008 and AL031 data, calculate means, and merge into one data.frame.

library(MuscleTernary)

AL_008 <- read_csv(system.file("extdata",
                               "AL_008_data.csv",
                               package = "MuscleTernary"),
                   show_col_types = FALSE) |>
  dplyr::select(-side, -force) |>
  coords_to_ternary(grouping = c("muscle"))

AL_031 <- read_csv(system.file("extdata",
                               "AL_031_data.csv",
                               package = "MuscleTernary"),
                   show_col_types = FALSE) |>
  dplyr::select(-side, -force) |>
  coords_to_ternary(grouping = c("muscle"))

M <- left_join(AL_031, AL_008, by = "muscle", suffix = c("_1", "_2")) |>
  as.data.frame()
M

x_1, y_1, and z_1 are the starting locations in ternary space. x_2, y_2, and z_2 are the ending locations.

We then interpolate each row into length_out new rows. Here we make 100 steps in between.

length_out <- 100
D <- list()

for (i in 1:nrow(M)) {
  D[[i]] <- interpolate_ternary(M[i, ],
                                length_out = length_out)
}
D <- do.call(rbind, D)

Basically we are making a list of data.frames and then binding them all together at the end with do.call().

To make the animation, first we make a set of all the plots. Here we iterate through .frame and make a plot for each. Add to a list of plots.

P <- list()

for (i in 1:length_out) {
  d <- D |> filter(.frame == i)
  P[[i]] <- ggtern(d, aes(x = x, y = y, z = z,
                          color = muscle)) +
    geom_point(size = 5) +
    muscle_color_map() +
    labs( x       = "ML",
          xarrow  = "Mediolateral",
          y       = "DV",
          yarrow  = "Dorsoventral",
          z       = "RC",
          zarrow  = "Rostrocaudal") +
    theme_bw(base_size = 16) +
    theme_showarrows() +
    guides(colour = guide_legend(override.aes = list(size = 5),
                                 ncol = 2, byrow = TRUE))
}

P now is a list of plots (here with a length of 100, because we used 100 steps). We can plot a few as examples:

P[[1]]
P[[50]]
P[[100]]

Saving the animation

To save the animation to a .gif file, we have to use additional software: ImageMagick. Installation of ImageMagick is beyond the scope of this article, but it's not too difficult.

You have to set options with ani.options differently for MacOS vs. Windows. Then call saveGIF().

# Set interval to 1/24 s.
ani.options(interval = 1/24)

# For MacOS with ImageMagick installed somewhere on the path
# e.g., using homebrew.

# ani.options(convert = "convert")

# For Windows, install the ImageMagick standalone release:
# (http://www.imagemagick.org/script/binary-releases.php). Use a
# variation of the next line to set the absolute path to convert.exe.

# ani.options(convert = 'C:\\Program Files\\ImageMagick\\convert.exe')

saveGIF({for (i in 1:length_out) print(P[[i]])},
        movie.name = "ternary_animation.gif",
        ani.width = 800, ani.height = 600)