https://github.com/nevrome/ggpointgrid

R Package - Rearrange scatter plot points on a regular grid
https://github.com/nevrome/ggpointgrid

ggplot2 plotting r

Last synced: 3 months ago
JSON representation

R Package - Rearrange scatter plot points on a regular grid

• Host: GitHub
• URL: https://github.com/nevrome/ggpointgrid
• Owner: nevrome
• License: gpl-3.0
• Created: 2020-10-31T21:00:58.000Z (about 4 years ago)
• Default Branch: main
• Last Pushed: 2023-05-18T08:49:38.000Z (over 1 year ago)
• Last Synced: 2024-07-16T23:02:13.229Z (4 months ago)
• Topics: ggplot2, plotting, r
• Language: R
• Homepage:
• Size: 500 KB
• Stars: 36
• Watchers: 4
• Forks: 0
• Open Issues: 2
• Metadata Files:
  • Readme: README.Rmd
  • License: LICENSE

Awesome Lists containing this project

README

        
---

output: github_document

editor_options: 

  chunk_output_type: console

---

# ggpointgrid

This package provides geoms to rearrange scatter-plot coordinates on regular grids while strictly avoiding over-plotting. The applications are similar to `geom_jitter`.

### Installation

You can install the development version from github with the following command (in your R console):

```

if(!require('remotes')) install.packages('remotes')

remotes::install_github("nevrome/ggpointgrid")

```

### Examples

```{r}

library(ggplot2)

set.seed(5)

```

#### `geom_pointgrid`

`geom_pointgrid` aims to optimize the arrangement of observations on a regular grid. This works well for figures with continuously scaled x- and y-axes, so for scatter-plots or even map plots. Just as in `geom_jitter` the rearrangement of the points reduces accuracy and precision of positional information on x and y in favour of making **every** observation visible.

The grid properties are controlled with the parameters `grid_x` and `grid_y`, which allow to precisely specify the desired graticules.

```{r, warning=FALSE}

df <- tibble::tibble(

  x = rep(c(1,1,2,3,3), times = 10),

  y = rep(c(1,3,2,1,3), times = 10),

  var = sample(c("A", "B", "C"), size = 50, replace = T)

)

coord <- coord_fixed(xlim = c(0.5,3.5), ylim = c(0.5,3.5))

p1 <- ggplot(df) +

  geom_point(aes(x, y, color = var)) +

  coord + ggtitle("geom_point")

p2 <- ggplot(df) +

  geom_jitter(aes(x, y, color = var), width = 0.3, height = 0.3) +

  coord + ggtitle("geom_jitter")

p3 <- ggplot(df) +

  ggpointgrid::geom_pointgrid(aes(x, y, color = var), grid_x = 15, grid_y = 15) +

  coord + ggtitle("geom_pointgrid")

p4 <- ggplot(df) +

  ggpointgrid::geom_pointgrid(

    aes(x, y, color = var),

    grid_x = seq(min(df$x) - 0.3, max(df$x) + 0.3, length.out = 18),

    grid_y = seq(min(df$y) - 0.3, max(df$y) + 0.3, length.out = 18)

  ) +

  coord + ggtitle("geom_pointgrid, grid specified")

cowplot::plot_grid(p1, p2, p3, p4)

```

`geom_textgrid` performs the same arrangement operation on text data. It is to `geom_text` what `geom_pointgrid` is to `geom_point`.

#### `geom_pointrect`

`geom_pointrect` was designed for a slightly different use-case than `geom_pointgrid`. Here all observations that share the x- and y-coordinate are spread out into a rectangular grid, representing only this one position. This is especially useful, when the x- and y- axis are ordinally scaled.

The order within each rectangle can be set by the order of the input data.frame and the arguments `scale_x` and `scale_y` control the size of the per-position box. The arguments `round_x` and `round_y` allow to specify how data on continuously scaled x- and y-axes should be aggregated.

```{r, warning=FALSE}

df <- tibble::tibble(

  x = rep(letters[c(1,1,2,3,3)], times = 10),

  y = rep(letters[c(1,3,2,1,3)], times = 10),

  var = sample(c("A", "B", "C"), size = 50, replace = T)

) |> dplyr::arrange(var)

coord <- coord_fixed()

p4 <- ggplot(df) +

  geom_point(aes(x, y, color = var)) +

  coord + ggtitle("geom_point")

p5 <- ggplot(df) +

  geom_jitter(aes(x, y, color = var), width = 0.3, height = 0.3) +

  coord + ggtitle("geom_jitter")

p6 <- ggplot(df) +

  ggpointgrid::geom_pointrect(aes(x, y, color = var)) +

  coord + ggtitle("geom_pointrect")

p7 <- ggplot(df) +

  ggpointgrid::geom_pointrect(

    aes(x, y, color = var),

    scale_x = 0.2,

    scale_y = 0.2

  ) +

  coord + ggtitle("geom_pointrect, scaling set")

cowplot::plot_grid(p4, p5, p6, p7)

```