A Python package for identification Differential Spatial Expression Pattern (DESP) gene by interpretable deep learning from multi-slice spatial omics data.

---

Preprint

River is able to identify Differential Spatial Expression Pattern (DSEP) across multi-slice dataset, and offers the downstream analysis based on obtained DSEP genes.

Getting started

Please refer to the

• Stereo-seq 3D dataset Tutorial (Can be downloaded by pysodb package)
• Stereo-seq development dataset Tutorial (Can be downloaded by pysodb package)
• Slide-seq mouse diabetes disease dataset Tutorial. (Can be downloaded by pysodb package)
• MIBI TNBC disease dataset Tutorial. (Can be downloaded by pysodb package)
• CODEX lupus dataset Tutorial. (Can be downloaded by pysodb package)

Installation

1. Create a conda environment

conda create -n river python=3.8 -y && conda activate river

2. Install the River dependency

pip install scSLAT
python -c "import torch; print(torch.__version__)"
pip install pyg_lib torch_scatter torch_sparse torch_cluster torch_spline_conv -f https://data.pyg.org/whl/torch-2.0.0+cu117.html  # replace torch and CUDA version to yours
pip install captum ipykernel 

Install the pysodb for efficient download processed Anndata in h5ad format (https://pysodb.readthedocs.io/en/latest/) Install the CellCharter for multi-slice co-clustering in Slide-seq analysis (https://github.com/CSOgroup/cellcharter)

Contribution

If you found a bug or you want to propose a new feature, please use the issue tracker.