navis-fastcore (Python)¶

Python bindings for the fastcore Rust core, built with PyO3. Functions here are generalized and do not depend on navis itself, so third-party libraries can use them without that rather heavy dependency — navis-fastcore's only runtime requirement is numpy.

Install¶

Pre-compiled wheels are on PyPI:

pip install navis-fastcore

See the GitHub repo if you want to build from source.

Usage¶

navis will use fastcore automatically where appropriate — you only need this section if you want to call it directly.

Tree functions take node_ids and parent_ids arrays; the mapping from arbitrary IDs onto the core's internal indices happens for you.

import navis_fastcore as fastcore
import numpy as np

node_ids = np.array([1, 2, 3, 4, 5])
parent_ids = np.array([-1, 1, 2, 3, 1])  # -1 marks the root

# All-by-all geodesic distances
dists = fastcore.geodesic_matrix(node_ids, parent_ids)

Available functions¶

Operations on rooted trees:

• geodesic_matrix: geodesic ("along-the-arbor") distances, either all-by-all or between specific sources and targets
• geodesic_pairs: geodesic distances between given pairs of nodes
• geodesic_nearest: distance to the nearest target for each source, without building the full matrix
• geodesic_farthest: distance to the farthest target for each source, without building the full matrix
• dist_to_root: distance from each node to its root
• connected_components: generate the connected components
• classify_nodes: classify nodes into roots, leaves, branch points and slabs
• synapse_flow_centrality: synapse flow centrality (Schneider-Mizell, eLife, 2016)
• strahler_index: calculate Strahler index
• subtree_height: distance from each node down to the farthest leaf below it
• prune_twigs: remove terminal twigs below a certain size
• break_segments: break the neuron into the linear segments connecting leafs, branches and roots
• generate_segments: same as break_segments but maximizing segment lengths, i.e. the longest segment goes from the most distal leaf to the root and so on
• segment_coords: coordinates per linear segment (useful for plotting)

Repairing fragmented skeletons:

• heal_skeleton: reconnect the fragments of a broken skeleton
• stitch_fragments: find the minimal-length edges that reconnect fragments

Meshes:

• mesh_connected_components: connected components of a triangle mesh

Neuron similarity:

• nblast / nblast_allbyall: NBLAST (query-vs-target / all-by-all)
• nblast_smart: two-pass approximate NBLAST for large comparisons
• synblast: synapse-based NBLAST

Interop:

• wrappers.csgraph: drop-in replacements for some scipy.sparse.csgraph routines