Map points between FlyWire v1 and FAFB14 space (xyz nm)

flywire2fafb maps points FlyWire->FAFB

fafb2flywire maps points FAFB->FlyWire

Usage

flywire2fafb(
  xyz,
  method = c("mapmany", "map1"),
  chunksize = 40000,
  swap = FALSE,
  ...
)

fafb2flywire(
  xyz,
  method = c("mapmany", "map1"),
  chunksize = 40000,
  swap = FALSE,
  ...
)

Arguments

xyz: A Nx3 matrix of points
method: Whether to map many points at once (default) or just one
chunksize: The number of points to send to the server when mapping many points at once.
swap: When TRUE applies the deformation field in the opposite direction e.g. to give a coarse mapping of points FAFB->FlyWire. This is wrong but may be useful.
...: Additional arguments passed to httr::GET/POST operation

Value

an Nx3 matrix of points

Details

Note that you can also access FlyWire<->FAFB bridging registrations via the xform_brain series of functions. This will allow you to transform most kinds of 3D data objects, whereas the flywire2fafb function is restricted to plain 3D coordinates. See examples.

Mapping single points is unlikely to be useful, but you may wish to adjust the chunksize argument to send more points at once at the risk of possible server timeouts. The default value is quite conservative.

When swap=TRUE displacements will be applied in the opposite direction to what is intended. This can be used to provide a coarse inverse mapping if you feed in FAFB points. This is wrong but was useful before the inverse mapping was available as it can get you closer to the right place in FlyWire than just assuming that FAFB14 and FlyWire are in the same space. This works because deformations are mostly fairly smooth at the scale of FAFB-FlyWire displacements. Operationally we find that residual displacements are typically of the order 100 nm using this procedure. Since 2020-08-08 we have a real inverse available, so this is now only of historical interest.

Examples

# identified location in FAFB14
p.fafb.nm <- cbind(477042, 284535, 90680)
p.fafb.raw <- p.fafb.nm/c(4,4,40)
# corresponding location in FlyWire
p.flywire.raw <- cbind(118865, 71338, 2267)
p.flywire.nm <- p.flywire.raw * c(4,4,40)

# check displacement
flywire2fafb(p.flywire.nm)-p.fafb.nm
#>      X  Y Z
#> [1,] 3 -1 0

# check what happens when you apply the inverse
fafb2flywire(p.fafb.nm)-p.flywire.nm
#>       X Y Z
#> [1,] -3 2 0

data("AV4b1", package='catmaid')
set.seed(42)
before=xyzmatrix(AV4b1)[sample(nvertices(AV4b1), size=2000), ]
after=fafb2flywire(before)
d=sqrt(rowSums((before-after)^2))
hist(d, br=20, main="FAFB14 - FlyWire Displacements", xlab="d /nm")


if (FALSE) {
AV4b1.flywire <- xform_brain(AV4b1, reference="FlyWire", sample="FAFB14")
plot3d(neuronlist(AV4b1.flywire, AV4b1))
}