navis.symmetrize_brain

navis.symmetrize_brain(x, template, via='auto', verbose=False)

Symmetrize 3D object (neuron, coordinates).

The way this works is by:

1. Finding the closest mirror transform (unless provided)

2. Mirror data on the left-hand-side to the right-hand-side using the proper (warp) mirror transform to offset deformations

3. Simply flip that data back to the left-hand-side

This works reasonably well but may produce odd results around the midline. For high quality symmetrization you are better off generating dedicated transform (see navis-flybrains for an example).

Parameters:

• x (Neuron/List | Volume/trimesh | numpy.ndarray | pandas.DataFrame) – Data to transform. Dataframe must contain ['x', 'y', 'z'] columns. Numpy array must be shape (N, 3).

• template (str | TemplateBrain) – Source template brain space that the data is in. If string will be searched against registered template brains.

• via ("auto" | str) – By default (“auto”) it will find and apply the closest mirror transform. You can also specify a template that should be used. That template must have a mirror transform!

Returns:

Same object type as input (array, neurons, etc) but hopefully symmetrical.

Return type:

xs

Examples

This example requires the flybrains library to be installed: pip3 install flybrains

>>> import navis
>>> import flybrains
>>> # Get the FAFB14 neuropil mesh
>>> m = flybrains.FAFB14.mesh
>>> # Symmetrize the mesh
>>> s = navis.symmetrize_brain(m, template='FAFB14')
>>> # Plot side-by-side for comparison
>>> m.plot3d()                                              
>>> s.plot3d(color=(1, 0, 0))