Skip to content

calc_3d_distance

jaybgager edited this page Sep 25, 2026 · 1 revision

Calculates the 3D distance between two points at a given xyz resolution in nanometers.

Arguments:

  • point_a (list of 3 integers, mandatory)
    • xyz coordinates of first point e.g. [1,2,3]
  • point_b (list of 3 integers, mandatory)
    • xyz coordinates of second point e.g. [1,2,3]
  • res (list of 3 integers, mandatory)
    • resolution in nm/voxel of each spatial dimension e.g. [4,4,45] for brain_and_nerve_cord

Returns:

  • distance (float)
    • the distance between the two points in nanometers

Example

Let's say we wanted to get the approximate width of the neck connective as it exits the VNC in the brain_and_nerve_cord datastack. You could pick two points on opposite sides of the overall mesh and feed them into this function with the viewer resolution to get the result.

INPUT:

import tracertools as tt

distance = tt.calc_3d_distance(
    point_a = [119837, 127150, 2758],
    point_b = [134142, 123812, 2724],
    res = [4,4,45],
)

print("Your points are", distance, "nanometers apart.")

OUTPUT:

Your points are 58777.08060120033 nanometers apart.

Clone this wiki locally