ggmagnify

geom_magnify() stores the plot state when it is added. Consequently, the order in which you add layers to your ggplot object determines whether those layers appear in the main plot or inside the magnified inset.

• To include a layer (like geom_smooth()) inside the inset: Add the layer before geom_magnify().
• To include a layer only in the main plot: Add the layer after geom_magnify().

# Layer inside the inset:
ggpi + geom_smooth() + geom_magnify(from = from3, to = to3)

# Layer only in the main plot:
ggpi + geom_magnify(from = from3, to = to3) + geom_smooth()

By default, the inset may use different grid line breaks than the main plot. To ensure the inset uses the same grid lines as the main graph, explicitly set the breaks in scale_x_continuous() and scale_y_continuous() for both the main plot and the magnification parameters.

# Fix the grid lines by defining explicit breaks
ggp2 +
  scale_x_continuous(breaks = seq(2, 5, 0.5)) + 
  scale_y_continuous(breaks = seq(5, 8, 0.5)) + 
  geom_magnify(from = c(2.45, 3.05, 5.9, 6.6), to = c(3.4, 4.4, 5.5, 6.6), shadow = TRUE)

To position a magnified inset outside the standard plot boundaries, you must disable clipping in the coordinate system and adjust the plot margins using theme(plot.margin = ...).

# Use coord_cartesian(clip = "off") to allow drawing outside boundaries
ggp + 
  coord_cartesian(clip = "off") + 
  theme(plot.margin = ggplot2::margin(10, 60, 10, 10)) + 
  geom_magnify(from = from, to = to + c(0.5e8, 0.5e8, 0, 0), shadow = TRUE)

You can call geom_magnify() multiple times on the same plot to create multiple insets. To create an 'inset within an inset', pass a ggplot object that already contains a geom_magnify() layer into the plot argument of a second geom_magnify() call.

# Multiple insets
ggm_unif + 
  geom_magnify(from = c(0.05, 0.15, 0.05, 0.15), to = c(0.2, 0.4, 0.2, 0.4)) +
  geom_magnify(from = c(0.25, 0.35, 0.25, 0.35), to = c(0.45, 0.85, 0.45, 0.85))

# Nested inset (inset within an inset)
ggp + 
  geom_magnify(plot = ggpm, 
               from = c(-0.3, 0.3, -0.3, 0.3),
               to = c(-3, -1, -3, -1),
               expand = FALSE, colour = "white")

You can install the latest stable release from r-universe or the development version from GitHub.

From r-universe (Recommended):

install.packages("ggmagnify", repos = c("https://hughjonesd.r-universe.dev", 
                 "https://cloud.r-project.org"))

From GitHub (Development version):

# install.packages("remotes")
remotes::install_github("hughjonesd/ggmagnify")

Because geom_magnify() uses masks, it requires R version 4.1.0 or higher and a graphics device that supports masking.

If you are using knitr, the ragg_png device is recommended. If your device does not support masks, you must use shape = "rect", though note that the plot inset will not be clipped to the panel area in this mode.

Instead of providing static coordinates, you can use aes() to select points to magnify based on logical conditions in your data.

# Magnify points where Species is 'setosa' and Sepal.Length < 5
ggpi + geom_magnify(aes(from = Species == "setosa" & Sepal.Length < 5), 
                    to = c(4, 6, 6, 7.5))

By default, geom_magnify() uses the statistical transformations (like smoothers or densities) calculated for the original plot. If you want the inset to recalculate these statistics based specifically on the zoomed-in data range, set recompute = TRUE.

# Recalculates the smooth (e.g., loess) specifically for the inset area
ggp2 + geom_magnify(
  from = c(-1.25, 1.25, -1, 1), 
  to = c(2, 5, 1, 5),
  recompute = TRUE
)

For complex or highly customized insets, use the plot argument in geom_magnify() to pass an explicit ggplot object. To ensure the inset looks appropriate (e.g., adjusting backgrounds or margins), use inset_theme() on the plot object you intend to use as the inset.

# 1. Create the custom inset plot
booms_inset <- booms + 
  geom_point(data = faithful, color = "red", ...) + 
  coord_cartesian(expand = FALSE) + 
  inset_theme()

# 2. Pass it to geom_magnify
booms + geom_magnify(
  from = c(78, 90, 4.0, 4.8), 
  to = c(70, 90, 1.7, 3.3),
  plot = booms_inset
)

Use geom_magnify(from, to) to create an inset of a specific area in a ggplot2 object.

• from: A vector defining the area to magnify. For a rectangle, use c(xmin, xmax, ymin, ymax). You can also map this inside aes() to select points based on data.
• to: A vector defining the location and size of the inset. For a rectangle, use c(xmin, xmax, ymin, ymax).

Note: The names xmin, xmax, ymin, and ymax are optional in these vectors.

library(ggplot2)
library(ggmagnify)

ggp <- ggplot(dv, aes(Position, NegLogP)) + 
  geom_point(color = "darkblue", alpha = 0.8, size = 0.8) + 
  labs(title = "GWAS p-values for cognitive function",
       subtitle = "Davies et al. (2018).", y = "-log(p)")

from <- c(xmin = 9.75e7, xmax = 9.95e7, ymin = 16, ymax = 28)
to <- c(2e8 - 2e7, 2e8 + 2e7, 10, 26)

ggp + geom_magnify(from = from, to = to)