Getting started with ggseg

In ggseg: Plotting Tool for Brain Atlases

#| label: setup
#| include: false
knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.width = 6
)

If you work with neuroimaging data, you've probably spent time wrestling region-level results into brain plots. ggseg handles that plumbing for you. It stores brain atlas geometries and plots them as ggplot2 layers, so you get brain figures with the same code you'd use for any other ggplot.

#| label: load-packages
library(ggseg)
library(ggplot2)

What's in a brain atlas

A brain_atlas object bundles four things: a name, a type (cortical, subcortical, or tract), region geometry, and an optional colour palette.

#| label: dk-atlas
dk()

The dk atlas (Desikan-Killiany) ships with the package. Call plot() for a quick look:

#| label: fig-dk-plot
#| fig-cap: "Quick overview of the Desikan-Killiany cortical atlas."
plot(dk())

Two helpers pull out the names you'll need for matching data:

#| label: dk-names
library(ggseg.formats)

atlas_regions(dk())
atlas_labels(dk())

Your first brain plot

geom_brain() works like any ggplot2 geom. Pass it an atlas and it draws the regions:

#| label: fig-basic-brain
#| fig-cap: "Default brain plot using geom_brain() with the dk atlas."
ggplot() +
  geom_brain(atlas = dk(), show.legend = FALSE)

Arranging views

Brain atlases have multiple views (lateral, medial) and hemispheres. position_brain() controls how they're laid out, using formula syntax borrowed from facet_grid():

#| label: fig-position-formula
#| fig-cap: "Brain views arranged using formula syntax in position_brain()."
ggplot() +
  geom_brain(
    atlas = dk(),
    position = position_brain(hemi ~ view),
    show.legend = FALSE
  )

Or reorder views with a character vector:

#| label: fig-position-character
#| fig-cap: "Brain views arranged using a character vector to specify order."
ggplot() +
  geom_brain(
    atlas = dk(),
    position = position_brain(c(
      "right lateral",
      "right medial",
      "left lateral",
      "left medial"
    )),
    show.legend = FALSE
  )

See vignette("positioning-views") for the full set of layout options, including grid layouts for subcortical and tract atlases and per-view zoom.

Showing only certain hemispheres or views

#| label: fig-filter-lateral
#| fig-cap: "Brain plot showing only lateral views."
ggplot() +
  geom_brain(atlas = dk(), view = "lateral", show.legend = FALSE)

#| label: fig-filter-left
#| fig-cap: "Brain plot showing only the left hemisphere."
ggplot() +
  geom_brain(atlas = dk(), hemi = "left")

For subcortical atlases, views are slice identifiers. Check what's available with ggseg.formats::atlas_views():

#| label: aseg-views
atlas_views(aseg())

Plotting your own data

Create a data frame with at least one column that matches the atlas (typically region or label). Pass it to geom_brain() through the data argument and it joins to the atlas automatically:

#| label: fig-external-data
#| fig-cap: "Brain plot coloured by external p-values using a viridis scale."
library(dplyr)

some_data <- tibble(
  region = c(
    "transverse temporal",
    "insula",
    "precentral",
    "superior parietal"
  ),
  p = sample(seq(0, 0.5, 0.001), 4)
)

ggplot() +
  geom_brain(
    atlas = dk(),
    data = some_data,
    position = position_brain(hemi ~ view),
    aes(fill = p)
  ) +
  scale_fill_viridis_c(option = "cividis", direction = -1) +
  theme_void()

See vignette("external-data") for more on data preparation and matching.

Faceting

ggplot2 faceting works as you'd expect. Group your data by the faceting variable and geom_brain() replicates the full atlas in each panel:

#| label: fig-faceting
#| fig-cap: "Brain plots faceted by group, each panel showing the full atlas."
some_data <- tibble(
  region = rep(
    c(
      "transverse temporal",
      "insula",
      "precentral",
      "superior parietal"
    ),
    2
  ),
  p = sample(seq(0, 0.5, 0.001), 8),
  group = c(rep("A", 4), rep("B", 4))
)

ggplot() +
  geom_brain(
    atlas = dk(),
    data = group_by(some_data, group),
    position = position_brain(hemi ~ view),
    aes(fill = p)
  ) +
  facet_wrap(~group)

Highlighting specific regions

To colour a few regions with their atlas colours, use two columns: one for the join (region) and one for the fill aesthetic:

#| label: fig-highlighting
#| fig-cap: "Highlighting selected regions using the atlas colour palette."
data <- data.frame(
  region = atlas_regions(dk())[1:3],
  reg_col = atlas_labels(dk())[1:3]
)

palette <- atlas_palette(dk())[1:3]

ggplot() +
  geom_brain(atlas = dk(), data = data, aes(fill = reg_col)) +
  scale_fill_brain_manual(palette)

Scales and themes

Atlas colour palettes are applied automatically by geom_brain(). For custom palettes, use scale_fill_brain_manual().

Built-in themes strip axes and grids for cleaner figures:

#| label: themes-example
#| eval: false
ggplot() +
  geom_brain(atlas = dk()) +
  theme_brain()

Going further

For full control over the data before it reaches ggplot2 — adding region labels, layering other sf geoms, or using geom_sf() directly — see vignette("geom-sf").

Finding more atlases

ggseg ships with dk() (cortical), aseg() (subcortical), suit() (cerebellar flamap), and tracula() (white matter tracts). Many more are available through the ggsegverse r-universe:

#| label: install-extra
#| eval: false
install.packages("ggsegYeo2011", repos = "https://ggsegverse.r-universe.dev")

ggseg documentation built on June 22, 2026, 9:09 a.m.