This module reads a SWC(.swc) or NEURON (.nrn) morphology file commonly used in Light Microscopic (LM) image reconstruction, then detects and labels the intersected branches based on spherical interpolation and graph theory. The detected intersections are written to an output file for curation.

• Python3.6 or above
• numpy
• networkx
• rtree
• (optional) NEURON with python integration if the morphology is in .nrn format

git clone https://github.com/CNS-OIST/SWCIntersectDetect.git
cd SWCIntersectDetect
python setup.py install

in Python interface, import the module

import swc_intersect_detect

If you want to convert NEURON .nrn morphology to .swc file, use

from swc_intersect_detect import morph_io
morph_io.nrn2swc(NRN_FILE, SWC_OUTPUT)

To perform intersection detection, use

swc_intersect_detect.run(SWC_FILE)

for more parameter control details, use

help(swc_intersect_detect.run)

An example is provided in example/purkinje.py, which first converts the example/Purkinje19b972-1.nrn morphology to a swc file, then performs intersection detection on the morphology data. The curation file can be visualized in neuTube as below, with curation tag (intersections) colored red.