Guides¶
Task-oriented walkthroughs for each part of the mudm core data model. Every
guide is grounded in the real public API and runs top-to-bottom against the
current mudm. Pick the guide that matches what you are trying to do:
| Guide | Use it when you want to… |
|---|---|
| Examples | Browse a gallery of complete muDM documents in both JSON wire format and equivalent Python, from a single feature to a full provenance-tagged collection. |
| Metadata & Properties | Attach arbitrary metadata to features and collections, and use the muDM members featureClass, parentId, and ref. |
| Ontology Vocabularies | Map free-text property values (e.g. "pyramidal") to stable ontology URIs with OntologyTerm and Vocabulary, without changing the human-friendly strings you store. |
| Coordinate Transforms | Build a 4x4 AffineTransform, apply it to GeoJSON and muDM 3D geometry, and convert between voxel and physical (micrometre) space with VoxelCoordinateSystem. |
| Spatial Layout | Arrange the features in a collection side-by-side without overlap using geometry_bounds and apply_layout, while keeping every shape unchanged. |
| Tile Metadata | Describe tiled, multi-resolution datasets with the TileJSON-style models in mudm.tilemodel (TileModel, TileLayer, PyramidJSON). |
| Validation | Validate muDM and plain GeoJSON with the MuDM and GeoJSON root models, read Pydantic errors, and round-trip to and from JSON. |
| Provenance & Traceability | Attach a structured, machine-readable record of the workflows and artifacts that produced your features, and link it back to specific fields. |
Not sure where to start?
Read Getting Started first — it takes you from install to a validated muDM document in a few minutes. The Specification is the formal reference for the data model. For exact class and function signatures, see the API Reference.
Looking for processing pipelines?
These guides cover the core data model only — building, validating, and
transforming muDM documents in pure Python. Tiling engines, format
converters (Xenium / OBJ / GeoJSON), and exporters (GeoParquet, glTF,
Neuroglancer) live in the separate mudm-tools package, documented at
https://novagenresearch.github.io/mudm-tools/.