This vignette provides an overview of basic usage of the
ggbrain package. The package seeks to generate 2D
volumetric plots of brain data stored in the NIfTI format. Additional
details about the structure of NIfTI data are here: https://neuroconductor.org/tutorials/nifti_basics. The
ggbrain package uses
ggplot2 to generate its
plots, allowing users to access the wide array of popular features in
ggplot2, including annotations and fine control over
coloration and themes.
Note that this package is unrelated to the ggBrain
package developed by Aaron Fisher in 2014. I had developed most of the
ggbrain package and had named it prior to
discovering Aaron’s work. The earlier
ggBrain package was
also not released publicly to CRAN and has not been updated much over
the years, whereas
ggbrain is hosted on CRAN and also
offers a wider array of functionality.
The package includes a few images that can be used for demonstrating its capabilities. The locations of the images on your machine are listed below.
print(c(underlay_3mm, decision_onset_3mm, echange_overlay_3mm, pe_overlay_3mm, abspe_overlay_3mm))
##  "/private/var/folders/y0/pkh_h_511bn27zjj8m0w2wx00000gn/T/RtmpAD6adc/Rinsta14a21951453/ggbrain/extdata/mni_template_2009c_3mm.nii.gz" ##  "/private/var/folders/y0/pkh_h_511bn27zjj8m0w2wx00000gn/T/RtmpAD6adc/Rinsta14a21951453/ggbrain/extdata/decision_onset_zstat_3mm.nii.gz" ##  "/private/var/folders/y0/pkh_h_511bn27zjj8m0w2wx00000gn/T/RtmpAD6adc/Rinsta14a21951453/ggbrain/extdata/echange_overall_zstat_3mm.nii.gz" ##  "/private/var/folders/y0/pkh_h_511bn27zjj8m0w2wx00000gn/T/RtmpAD6adc/Rinsta14a21951453/ggbrain/extdata/pe_ptfce_fwep_0.05.nii.gz" ##  "/private/var/folders/y0/pkh_h_511bn27zjj8m0w2wx00000gn/T/RtmpAD6adc/Rinsta14a21951453/ggbrain/extdata/abspe_ptfce_fwep_0.05.nii.gz"
You can install the latest public version of the
You can also to install the development version. It may may have new features and bugfixes, though it may also have new bugs or include works in progress.
One can create an empty ggbrain object using the
function. This is akin to creating an empty ggplot object using
gg_obj <- ggbrain()
At this point, trying to plot the
ggbrain yields a
warning given that no data or geometric layers have been added.
## Warning in self$render(guides): No brain layers added to this object yet. Use + ## geom_brain() or + geom_outline() to add.
Consistent with the
ggplot2 approach, additional
elements can be added using the
+ operator, as we elaborate
below. The main thing to know at this point is that you need to
add all ggbrain-specific elements before adding typical ggplot
components such as
Also note that the order of the addition operations for images and
slices, respectively) is
not important. The example order below – adding images, slices, layers,
and labels – is conceptually intuitive, but could be reordered without
altering the plut. As in
ggplot2, the order of all geom_*
layers does matter, such that geom_* layers that are added
later are rendered on top of layers added earlier. For example, the
geom_brain() layer for an underlay image should be added
before an overlay.
A ggbrain plot must have at least one 3D image stored in NIfTI
format. These images are read into ggbrain using the
package. All images for a given plot must have the same image dimensions
(otherwise, an error will occur). At present,
provides no internal support for interpolation to combine images with
different resolutions. We encourage you to use
from AFNI for this purpose (see details in the ‘Other considerations’
The file names containing relevant images can be passed in as
ggbrain or they can be appended using
images() In either, it is important to provide a name for
the image that is used internally to refer to the image in contrasts,
labels, or other operations. If you do not provide a name, you will see
Warning message: In private$set_images(images, fill_holes, clean_specks) : The images vector does not contain any names. This may lead to weird behaviors downstream if 'underlay' and 'overlay' are requested.
gg_obj <- gg_obj + images(c(underlay = underlay_3mm))
Note that the image name of ‘underlay’ has a special meaning in the package. Specifically, using this name will set a default color scale that is grayscale and will not show the legend of this scale in the plot. You don’t have to use ‘underlay’ but it can make things a bit simpler if you have a conventional plot that shows an anatomical scan underneath functional activations.
You can add many images at once by providing a named vector:
gg_obj <- ggbrain() + images(c(underlay = underlay_3mm, echange = echange_overlay_3mm))
Or you can add many images in independent
Note that the order of addition is not important.
gg_obj <- ggbrain() + images(c(underlay = underlay_3mm)) + images(c(echange = echange_overlay_3mm))
After adding images to the ggbrain object, the next step is to
specify which slices should be displayed on the plot. This is achieved
slices() function. Following standard
neuroimaging terminology, we can add axial, sagittal, and coronal
The order of the addition operations, whether within a single
slices() call or in a sequence of calls like
slices("x=12") + slices(c("y=15", "z=10")), determines the
order of panels on the rendered plot from upper left to lower right
(following the standard
facet_wrap() conventions in
ggplot2). Thus, consider the order in which you wish the panels to
appear so that the slice additions follow this order.
The most conventional approach is add slices based on their spatial position along one of three cardinal axes, x/sagittal, y/coronal, or z/axial. You can determine the locations of interest using an interactive image viewer such as afni or fsleyes. Or, as described below, you could add a montage of images, add the coordinate labels on the panels, then choose the coordinates of the slices that are most relevant.
You can add one or more slices using
slices() by passing
a character vector of positions.
gg_obj <- gg_obj + slices(c("x = 10", "y = 70", "z = 15"))
This will add one sagittal slice (x=10), one coronal (y=70), and one axis (z=15).
If you want to use standardized coordinates in terms of percentiles,
rather than spatial coordinates, you can append
% at the
end of the
slices() specification. This will find the
position of the corresponding slice along the axis between 0% and 100%.
These percentiles are computed based on the slices that are not entirely
empty (i.e., not all zero).
gg_obj <- gg_obj + slices(c("x = 25%", "y = 50%", "z = 90%"))
If you would like to alter specific aesthetic elements of the slices
such as the panel title, border, or axis labels, these can be added as
slices(). Aesthetic arguments that will alter
the appearance of the relevant slices include:
title: The panel title
bg_color: The background color of the panel
text_color: The color of text on the panel
border_color: The color of a border around the panel
border_size: The size (thickness) of the border around the panel. Default is NULL (none).
xlab: The x axis label of the panel
ylab: The y axis label of the panel
theme_custom: A ggplot2
theme()(or list of theme calls) added to this panel that customizes its appearance.
gg_obj <- gg_obj + slices( c("x = 25%", "x = 50%"), bg_color = "gray25", xlab = "x axis", theme_custom = theme(axis.text.x = element_text(size = 24)) )
If you would like a set of slices along one of the image planes to
form a montage, you can use the
function to specify the inputs to
slices(). The arguments
plane: “sagittal”, “coronal”, or “axial” (“x”, “y”, and “z” also work)
n: the number of slices to render along this plane
min: the lowest quantile to be included (between 0 and 1). Default: 0.1
max: the highest quantile to be included (between 0 and 1). Default: 0.9
min_coord: the lowest spatial position to be included. Mutually exclusive with quantile approach. Default: NULL
max_coord: the highest spatial position to be included. Mutually exclusive with quantile approach. Default: NULL
The code below adds 10 axial slices between the 10th and 90th percentiles.
gg_obj <- gg_obj + slices(montage("axial", 10, min = 0.1, max = 0.8))
This code adds 5 sagittal slices between the x = -10 and x = 10 (millimeters, in image coordinates).
gg_obj <- gg_obj + slices(montage("sagittal", 5, min_coord = -10, max_coord = 10))
Now that we have defined the images to read and the slices to render,
we need to map the numeric values in the slices to the points on the
plot. This is achieved by
geom_brain(), which is
essentially an extension of
ggplot2. The numeric values in the slices fill squares on
the plot, essentially forming ‘pixels’.
As with standard
ggplot2 objects, the fill color of
these pixels is determined by a
allowing you to have fine control over the coloration.
layer supports the following standards:
interpolate: if TRUE,
geom_rastersmooths the layer, rather than rendering discrete squares.
alpha: A number between 0 and 1 controlling the alpha transparency of the layer
show_legend: If TRUE (default), a legend mapping the numeric values to a color scale is included in the panel
fill: A color to be used to fill all non-NA pixels. In ggplot2 terms, this is ‘setting’ the color of all pixels, as opposed to ‘mapping’ the color to the numeric value via a scale.
mapping: The aesthetic mapping between the data and the raster layer. In general, the default of
aes(fill=value)should be retained for parametric layers (i.e., where the color of pixels should map to a continuous scale). But as detailed in a subsequent section, the fill can mapped to label columns in a categorical image (e.g., atlas).
Relative to standard ggplot2 conventions, the use of
geom_brain() requires that you specify details about the
fill color scale when the layer is added. This results from the fact
ggbrain objects re-use the ‘fill’ geometry when layers
are overlaid. For example, if you have an anatomical underlay with a
grayscale fill and a statistical overlay with a color fill, each layer
needs to have its own fill scale (with a
object). This is achieved internally using the
The ‘take-home message’ is that you need to specify fill scale
information as part of the
geom_brain() call. Here are the
limits: a two-element vector that if provided, sets the upper and lower bounds on the scale
breaks: a function to draw the breaks on the scale. The default is
integer_breaks(), which draws a few integer-valued breaks (usually, 5) between the min and max of the scale.
unify_scale: if TRUE (default), the color scale across panels (slices) will be equated. If FALSE, each panel will apply the color scale independently such that the min and max are slice-dependent.
scale_fill_*object use for mapping the numeric values in the data to the fill channel on the plot. Standard scales such as
scale_fill_viridis_c()can be used for continuous layers. Note that the package also provides a
scale_fill_bisided()option that, if added, creates separate color scales for the positive and negative pixels on the plot. This is a conventional approach in displays of fMRI activation maps where, for example, positive modulation is mapped to warm colors (reds) while negative modulation is mapped to cool colors (blues). Note that
scale_fill_bisided()has arguments of
pos_scalethat allow the scales used for negative and positive values, respectively, to be specified directly.
geom_brain() function includes an argument called
definition. This controls the numeric values that are
selected for display on the layer. The simplest definition is to name
the image to be displayed. Recall that when you use
images(), you provide names for the images. These are used
as a reference by
ggbrain to interpret your layer
For example, here I want to render the underlay and overlay layers with their numeric values:
# define a reusable object that has the same images and slices gg_base <- ggbrain(bg_color = "gray80", text_color = "black") + images(c(underlay = underlay_3mm, overlay = echange_overlay_3mm)) + slices(c("x = 10", "y = 70", "z = 15"))
gg_obj <- gg_base + geom_brain(definition = "underlay", fill_scale = scale_fill_gradient(low = "grey8", high = "grey62"), show_legend = FALSE) + geom_brain(definition = "overlay", fill_scale = scale_fill_bisided(), show_legend = TRUE)
Now that we have
geom_* layers in our
ggbrain object, we can finally plot our figure to see the
Oftentimes, we wish to modify the display by selecting a subset of
values to display. One of the most common applications is thresholding
of voxelwise statistics so that small/nonsignificant values are omitted.
To achieve this,
geom_brain() supports standard subsetting
syntax used in R. Here is an example of thresholding the overlay layer
so that the absolute (z) statistic is above 2.5:
gg_obj <- gg_base + geom_brain(definition = "overlay[abs(overlay) > 2.5]", fill_scale = scale_fill_bisided(), show_legend = TRUE) plot(gg_obj)
Note how ‘overlay’ is treated as an R object that is subset using square brackets. The subsetting expression inside the brackets uses standard R syntax and can reference other images currently in the ggbrain object. For example, you could display overlay values only where the underlay is greater than 10:
gg_obj <- gg_base + geom_brain(definition = "underlay") + geom_brain(definition = "overlay[underlay > 10]", fill_scale = scale_fill_bisided(), show_legend = TRUE) plot(gg_obj)