Installation

This repository currently provides the spora_io Python package from source. Install it in editable mode from the repository root:

git clone https://github.com/bunnelab/spora-io.git
cd spora-io
pip install -e .

Required Data Configuration

The loaders need a root directory containing one or more curated datasets. Set SPORA_DATASETS_DIR to that root:

export SPORA_DATASETS_DIR=/path/to/datasets_folder

Dataset constructors resolve dataset roots from this environment variable by default. For example, HEImagingDataset(name="my_dataset", ...) loads $SPORA_DATASETS_DIR/my_dataset unless an explicit path override is provided.

Runtime Dependencies

The core loaders expect common scientific Python packages, including numpy, pandas, torch, zarr, and einops. Multiplex standardization and parquet-backed metadata also require parquet support through pandas, typically provided by pyarrow.

Verify Install

from spora_io import HEImagingDataset, MultiplexImagingDataset, ComposedImagingDataset, SporaDataset
print("spora_io import OK")

Compatibility Note

Use Python 3.10 or newer. The current loaders call a pandas grouping API introduced in pandas 2.2, so install pandas>=2.2 even though the package metadata currently permits older versions. Optional processing scripts can require additional dependencies; see Tools and Scripts.