README.md

neuromorpholib

Installation

pip3 install neuromorpholib

Usage

Downloading

A simple download of known dataset and name

from neuromorpholib import neuromorpho

nmo = neuromorpho.NeuroMorpho()
acc1 = nmo.download_swc("martone", "ACC1")

A query for all species=mouse neurons

mouse_neurons = nmo.search({"species": "mouse"})

Download a SWC for a mouse neuron

swc_demo = nmo.download_swc(
    mouse_neurons[0]
)

If you know the archive name and neuron name, you can also download the swc directly by passing archive and neuron_name arguments.

If you only want the SWC text and don't want it to be converted into a NeuronMorphology object, you can pass text_only=True.

SWC Management

Read a SWC file from disk

from neuromorpholib.swc import load_swc, NeuronMorphology

my_morphology = load_swc("my_neuron.swc")
# This is a NeuronMorphology object.

Get a list of branch points

branch_points = my_morphology.get_branch_points()

Get a simplified graph (only leaves and branch points) of a morphology

morphology_graph = my_morphology.smoothed()
# nx.DiGraph

SWC Geometry Operations

Rotate, translate, or scale a geometry with NeuronMorphology functions:

n = NeuronMorphology()
n.scale(3)
n.translate([4, 44, 10])
n.rotate([0, 0, math.pi/2])

Rendering SWCs

All of the following visualization techniques assume you have a single NeuronMorphology object, downloaded perhaps like this:

from neuromorpholib import neuromorpho

nmo = neuromorpho.NeuroMorpho()
acc1 = nmo.download_swc("martone", "ACC1")

Using matplotlib

import networkx as nx

g = acc1.get_graph()
nx.draw(g, pos={n: a['xyz'][:2] for n, a in g.nodes(data=True)}, node_size=0)

Using pytri

This will return an interactive 3D turntable in a Jupyter Notebook.

from pytri import Figure

fig = Figure()
fig.graph(acc1.get_graph(), pos_attribute="xyz")
fig.show()