unravel.coordinates.points_to_sphere_masks module

Use coords_points_to_sphere_masks or points_to_sphere_masks from UNRAVEL to create one binary sphere/ellipsoid mask per point in a CSV.

This is useful for generating local-neighborhood masks around transformed coordinates, e.g. MERFISH-space unit locations.

Input:

• CSV with voxel coordinate columns

• Reference image for output shape/header

Output:

• One .nii.gz binary mask per input row

Note

• Radius and voxel size must be in the same physical units, usually µm.

• Coordinates are voxel/index coordinates, not physical coordinates.

• For MERFISH-CCF, voxel size is often 10 10 200 µm.

Usage:

points_to_sphere_masks

-i points_MERFISH.csv -ri resampled_annotation.nii.gz -x merfish_x -y merfish_y -z merfish_z -id voxel_label -r 100 -vx 10 10 200 -o unit_sphere_masks_MERFISH -v

unravel.coordinates.points_to_sphere_masks.parse_args()

unravel.coordinates.points_to_sphere_masks.parse_voxel_size(voxel_size)

Return voxel size as a length-3 numpy array.

unravel.coordinates.points_to_sphere_masks.sanitize_filename(text)

Make a safe filename component.

unravel.coordinates.points_to_sphere_masks.make_mask_name(row, row_index, id_col=None, name_cols=None, radius=None)

Create a filename stem for one point/mask.

unravel.coordinates.points_to_sphere_masks.sphere_mask_for_coord(shape, center, radius, voxel_size)

Create a binary physical-radius sphere/ellipsoid mask around a voxel coordinate.

unravel.coordinates.points_to_sphere_masks.main()