Histopia

Histology Spatial Topology for Omics Profiling and Inter-section Alignment

Histopia provides reusable tools for serial-section alignment, cell segmentation, stain quantification, protein prediction, and spatial tissue reconstruction.

Documentation · Interactive demo · QuPath extension

Install

The base package is lightweight. Install dependencies for the workflows you need:

pip install "histopia[registration,wsi]"

For the latest source version:

git clone https://github.com/oncologylab/histopia.git
cd histopia
pip install -e ".[registration,wsi]"

Other optional profiles include stain, cells, semantic, uni2h, protein, and topology. See installation details.

Workflows

Workflow	Guide
Tissue masking and serial-section alignment	Registration
Chromogenic stain quantification	Stain quantification
Cell boundaries and segmentation merging	Cell segmentation
Histology embeddings and tissue regions	Semantic atlas
Protein-expression models	Protein prediction
Spatial topology and reconstruction	Topology
Per-cell protein visualization	Cellular protein atlas

Start with the example configurations. Each command supports --help, for example histopia-register --help.

Development

pip install -e ".[dev,registration,semantic,topology,stain,wsi]"
ruff check .
pytest

Reusable code lives in src/histopia, with small fixtures and tests in tests. Datasets, checkpoints, generated reports, and local research operations stay outside version control. Heavy image and model dependencies are loaded only when needed.

Histopia is research software under active development. Registration, prediction, and reconstructed or interpolated tissue require task-specific validation; a displayed result does not establish biological accuracy.

License

BSD 3-Clause.