API Reference

Ossify is a Python package for working with neuromorphological data, providing tools for analyzing and visualizing neuron structures across multiple geometric representations.

Quick Start

import ossify

example_cell_path = "https://github.com/ceesem/ossify/raw/refs/heads/main/864691135336055529.osy"
# Load a cell from file locally or in the cloud
cell = ossify.load_cell(example_cell_path)

# Create from CAVE client (requires caveclient)
cell = ossify.load_cell_from_client(root_id=12345, client=cave_client)

# Analyze morphology
strahler = ossify.strahler_number(cell)
is_axon = ossify.label_axon_from_synapse_flow(cell)

# Create visualizations
fig, ax = ossify.plot_cell_2d(cell, color="compartment")

Package Structure

Core Classes

The foundation classes for representing neuromorphological data:

• Cell: Main container for morphological data with multiple data layers
• Link: Manages relationships between different data layers

Data Layer Classes

Specialized classes for different geometric representations:

• SkeletonLayer: Rooted tree-structured neuronal skeletons
• GraphLayer: Graph-based representation for spatial connectivity
• MeshLayer: 3D mesh surfaces with face-based geometry
• PointCloudLayer: Sparse point annotations and markers

Analysis & Algorithms

Computational methods for morphological analysis:

• Morphological Analysis: Strahler number, compartment classification

Visualization & Plotting

Plotting and visualization utilities:

• Cell Plotting: Integrated cell visualization with multiple projections
• Figure Management: Multi-panel layouts and precise sizing

File I/O Operations

Loading and saving morphological data:

• Core I/O Functions: load_cell(), save_cell()
• File Management: CellFiles for cloud and local storage

External Integrations

Interfaces with external data sources and tools:

• CAVE Integration: cell_from_client() for connectome data

Key Features

Multi-Scale Data Integration

Seamlessly work with data at different scales - from high resolution meshes to coarse skeletons, with automatic mapping between representations.

Flexible Analysis Pipeline

Chain operations across different data types with consistent APIs and automatic data propagation between layers.

Publication-Ready Visualization

Create high-quality figures with precise unit control, multiple projections, and customizable styling.

Cloud-Native I/O

Load and save data from local files, cloud storage (S3, GCS) via cloud-files, or directly from CAVE.

---

Navigation

Module	Description	Key Classes
Core Classes	Foundation classes and containers	Cell, Link
Data Layers	Spatial and graph representation classes	SkeletonLayer, GraphLayer, MeshLayer, PointCloudLayer
Algorithms	Analysis and computation functions	strahler_number, label_axon_*, smooth_features
Plotting	Visualization and figure creation	plot_cell_*, plot_morphology_*
File I/O	Data loading and saving	load_cell, save_cell, CellFiles
External	Third-party integrations	cell_from_client

Best Practices

• Use Cell.apply_mask() for non-destructive filtering
• Use mask_context() for temporary operations
• Use annotations for sparse point-like data and features for dense data where every vertex has a value.
• Leverage Link objects for complex data relationships
• Take advantage of method chaining for concise workflows