Skip to content

Usage

This pipeline supports both command-line execution and interactive analysis via Jupyter notebooks.

Command-Line Usage

To run the pipeline from the command line, use the provided scripts:

Prepare Cores

python scripts/prepare_cores.py --config config/analysis_pipeline.yaml

This script wraps the CoreController class, which handles:

  • Reading image and metadata files.
  • Cutting out cores with appropriate margins and masking.
  • Writing intermediate TIFFs.
  • Assembling per-core Zarr datasets using the SpatialData model.

For more details, see the core_cutter.py source and its configuration.

Jupyter Notebook Usage

For interactive inspection, prototyping, or educational purposes, the following notebooks illustrate how to use the components directly:

  • core_selection_demo.ipynb: An interactive notebook that enables users to define cores as rectangles or polygons using the Napari viewer. It supports automatic core detection via Segment Anything v2, with the option to manually correct the detected shapes or draw new ones from scratch. This step is interactive and only available as a notebook. The result is a core_info.csv file containing core metadata for use in subsequent steps via either Jupyter or CLI.
  • core_cutting_demo.ipynb: Demonstrates how to load a single image and metadata entry and apply the core cutting logic.