Mirror yakuba neurons or points to the opposite side of the VNC

mirror_dyak() mirrors neurons, surfaces and other point data in yakuba space to the opposite side of the VNC.

symmetric_dyak() transforms objects onto a symmetrised version of the yakuba template, optionally mirroring across the midline.

Usage

mirror_dyak(x, units = c("nm", "microns"), subset = NULL, ...)

symmetric_dyak(
  x,
  units = c("nm", "microns"),
  mirror = FALSE,
  subset = NULL,
  ...
)

Arguments

x: An object compatible with nat::xform() (neuron, neuronlist, dotprops, surface, coordinate matrix etc).
units: Coordinate units of x and the returned object. Defaults to nm, matching read_dyak_neurons().
subset: Optional subset passed to nat::xform(), e.g. when transforming selected elements of a neuronlist.
...: Additional arguments passed to nat::xform().
mirror: Whether to mirror across the midline of the symmetric space.

Value

A transformed object of the same kind as x, in the same units.

Details

Mirroring maps x into the symmetric yakubasym space with a bundled thin-plate spline registration (yakuba_yakubaSym_1000pts_tps.rds, computed by Sebastian Cachero), flips across the X axis and then maps back with the inverse registration. As for the default method of xform_dyak2manc(), points are first shifted by (-6.5, -7.1, -8.0) µm to match the frame in which the registration was computed. The registration is defined in microns; nm inputs are scaled before and after transformation. Requires the suggested Morpho package.

Denser versions of the registration (10000 points) were also computed but gave < 1 µm change in mirrored positions at ~500x the compute cost.

symmetric_dyak() returns objects in yakubasym space, using units for both input and output.

Examples

if (FALSE) { # \dontrun{
dna02 <- read_dyak_neurons("DNa02")
dna02.m <- mirror_dyak(dna02)
plot3d(dna02, col = "black")
plot3d(dna02.m, col = "red")
} # }
if (FALSE) { # \dontrun{
# DNa02 neurons and their mirror images in symmetric space
dna02.sym <- symmetric_dyak(dna02)
dna02.sym.m <- symmetric_dyak(dna02, mirror = TRUE)
} # }