SNT: Python Notebooks

Revision as of 17:44, 9 May 2019 by Carshadi (talk | contribs) (Advanced Examples)


Direct access to the SNT API from the Python programming language is made possible through the pyimagej module. This enables full integration between SNT and any library in the Python ecosystem.

Installing pyimagej

Follow the instructions given here

Getting Started

To initialize Fiji from Python with GUI capabilities:

import imagej
ij = imagej.init('sc.fiji:fiji', headless=False)

Then, use pyjnius (bundled with pyimagej) to import the SNT (Java) classes you need. For example, to download a neuron reconstruction from the MouseLight database and calculate summary statistics on it, you would import the MouseLightLoader and TreeStatistics classes:

from jnius import autoclass
MouseLightLoader = autoclass('')
TreeStatistics   = autoclass('tracing.analysis.TreeStatistics')

Now you can access all the attributes and methods these classes offer. Let's get a summary of the inter-node distances for a specific mouse cortical motor neuron (ID = "AA0100" in the MouseLight database).

def run():
    loader = MouseLightLoader("AA0100")
    if not loader.isDatabaseAvailable():
        print("Could not connect to ML database", "Error")
    if not loader.idExists():
        print("Somehow the specified id was not found", "Error")

    a_tree = loader.getTree('axon', None) # compartment, color
    s_stats = TreeStatistics(a_tree)
    metric = TreeStatistics.INTER_NODE_DISTANCE
    summary_stats = s_stats.getSummaryStats(metric)
    print("The average inter-node distance is %d" % summary_stats.getMean())

Advanced Examples

The Notebooks directory in the SNT GitHub repository contains examples which are executable on Binder. For example, to run the convex_hull notebook via, follow this link: convex_hull.ipynb