stories-jax 0.1.0

Learn spatially informed Waddington-like potentials for single-cell gene expression

Learning cell fate landscapes from spatial transcriptomics using Fused Gromov-Wasserstein

STORIES is a novel trajectory inference method for spatial transcriptomics data profiled at several time points, relying on Wasserstein gradient flow learning and Fused Gromov-Wasserstein. Read the preprint here and the documentation here!

Install the package

STORIES is implemented as a Python package seamlessly integrated within the scverse ecosystem. It relies on JAX for fast GPU computations and JIT compilation, and OTT for Optimal Transport computations.

via PyPI (recommended)

pip install stories-jax

via GitHub (development version)

git clone [e-mail]:cantinilab/stories.git
pip install ./stories/

Getting started

STORIES takes as an input an AnnData object, where omics information and spatial coordinates are stored in obsm, and obs contains time information, and optionally a proliferation weight. Visit the Getting started and API sections for tutorials and documentation.