xform.ngscene: Add a transformation to one or more layers in a neuroglancer...

In natverse/fafbseg: Support Functions for Analysis of FAFB EM Segmentation

Add a transformation to one or more layers in a neuroglancer scene

Description

Add a transformation to one or more layers in a neuroglancer scene

Usage

## S3 method for class 'ngscene'
xform(x, reg, layers = NULL, ...)

Arguments

x	A neuroglancer scene as produced by ngl_decode_scene
reg	A registration either as a reglist containing multiple registrations or a single registration in any form handled by xform.
layers	A character vector specifying the layers in the scene to transform. If the elements are named, they names specify the new names of the transformed layer.
...	Additional arguments passed to fit_xform when reg specifies a non-rigid registration. See details.

Details

Neuroglancer only implements homogeneous affine transforms for layers. However these can still be quite useful when a non-rigid transform cannot be applied to a layer e.g. because the underlying neurons are undergoing rapid editing and it is not practical to generate a static set of transformed meshes.

Value

A new ngscene object

Examples

# flywire scene
u='https://ngl.flywire.ai/?json_url=https://globalv1.flywire-daf.com/nglstate/4559706743898112'
scf=ngl_decode_scene(u)
m=matrix(c(-0.9663, -0.0695, 0.17, 0, 0.0351, 1.043, -0.028, 0, 0.1082,
  -0.0093, 0.9924, 0, 1021757.1284, 31409.0911, -85626.0572, 1), ncol=4)
# nb replaces existing layer of this name
xform(scf, m, layers=c('Production-mirrored'))

scu=ngl_decode_scene('https://tinyurl.com/kj9rwn26')
# make a new layer mirroring an existing layer
scu2=xform(scu, m, layers=c('fly_v31_m'='fly_v31'))
scu2

## Not run: 
browseURL(as.character(scu2))

## End(Not run)

## Not run: 
# mirror a flywire scene based on points from a specific pair of neurons
mbon18.dps=read_l2dp('MBON18')
mirror_reg=fit_xform(samplepts = mbon18.dps,
  refpts = nat.jrcbrains::mirror_fafb(mbon18.dps), subsample = 500)
flywire_scene('MBON18') %>%
  ngl_decode_scene %>%
  xform(mirror_reg, layers=c("mirror"="Production-segmentation_with_graph")) %>%
  as.character() %>%
  browseURL()

## End(Not run)

natverse/fafbseg documentation built on Nov. 11, 2024, 9:50 p.m.