🚀 This package is part of
BIOMERO 2.0 — For complete deployment and FAIR infrastructure setup, start with the NL-BIOMERO Documentation 📖
Shared, versioned Pydantic contracts for the BIOMERO ecosystem.
Documentation · Python API · Pixel identity reference
The package keeps two contract areas separate:
biomero_schema.modelsis BIOMERO's normalized representation of a workflow descriptor.biomero_schema.zarr,biomero_schema.imports, andbiomero_schema.shallowerdefine internal cross-service contracts for managed Zarr sources, pixel identity, shallow collections, importer lifecycle operations, and remote shallowing reports and receipts.
Workflow providers normally describe their tools using an external workflow
format. The adapters in biomero_schema.adapters detect and convert supported
formats into the validated WorkflowSchema from this package:
BIAFLOWS descriptor.json ─┐
├─> biomero adapters ─> WorkflowSchema ─> BIOMERO services
BILAYERS config.yaml ─────┘
Currently supported inputs are:
| Provider format | BIOMERO support |
|---|---|
BILAYERS config.yaml |
Converts the container, command, citations, inputs, parameters, and declared outputs. This is the preferred route for Zarr-to-Zarr and Plate-aware workflows. |
| BIAFLOWS/Cytomine descriptor | Converts the supported legacy cytomine-0.1 subset. It remains useful for established TIFF-oriented BIAFLOWS workflows. |
Native biomero-0.1 descriptor |
Validates directly. This is primarily BIOMERO's normalized service representation, although integrations may produce it explicitly. |
CWL and OpenAPI descriptors are not yet supported. The adapters were moved
here from biomero.schema_parsers
so descriptors can be converted without the BIOMERO runtime:
from biomero_schema.adapters import parse_descriptor
workflow = parse_descriptor(yaml.safe_load(open("config.yaml")))The adapter determines which fields from an external format are currently converted.
See Workflow descriptors for the conversion boundary and supported mappings.
Python 3.11 or newer is required.
pip install biomero-schemaFor development:
git clone https://github.com/NL-BioImaging/biomero-schema.git
cd biomero-schema
pip install -e .The repository also supports Pixi:
pixi run testThe CLI validates the normalized BIOMERO descriptor. BILAYERS (.yaml) and
BIAFLOWS descriptors are detected and converted before validation, and
parse --json writes the converted descriptor.
biomero-schema validate config.yaml
biomero-schema parse config.yaml --json > descriptor.json
biomero-schema validate descriptor.json
biomero-schema parse descriptor.json
biomero-schema parse descriptor.json --pretty
biomero-schema schemaA maintained native example is available at
tests/example_workflow.json.
Consumers should import the shared models rather than duplicating JSON fields:
from biomero_schema.zarr import CanonicalZarrSource
source = CanonicalZarrSource.from_dict(payload)
wire_payload = source.to_dict()
json_schema = CanonicalZarrSource.model_json_schema()The camelCase output from to_dict() is the stable wire representation. Every
contract family carries its own integer schema, independently of the workflow
descriptor version.
The shallow-Zarr manifest is an experimental, versioned BIOMERO storage contract, not RFC-8 metadata. Its portable image/label graph is separate from BIOMERO storage and provenance bindings so that a future standards adapter can reuse the graph without redefining the operational data.
See Remote shallower contracts for the report formats, trusted receipt hand-off, and compatibility requirements. These models do not enable shallow storage or remote execution; feature configuration belongs to the consuming services.
python -m pip install -e .
python -m pip install -r docs/requirements.txt
mkdocs serveUse mkdocs build --strict to reproduce the documentation CI build. See
docs/contributing.md for publishing details.