Skip to content

Latest commit

 

History

History
79 lines (57 loc) · 3.08 KB

README.md

File metadata and controls

79 lines (57 loc) · 3.08 KB
Raw

A Python package for the Scalale and accurate identification condition-relevant niches from spatial omics data.

Preprint

Taichi is able to automatically identify condition-relevant niches, and offers the downstream analysis based on obtained niches.

Getting started

Please refer to the

  • STARmap Simulation dataset Tutorial
  • MERFISH Simulation dataset Tutorial (Simulation perturbated condition data link and original control data link)
  • Slice-seq v2 DKD mouse disease dataset Tutorial. (Can be downloaded by pysodb package)
  • CODEX proteomics CRC dataset Tutorial. (Can be downloaded by pysodb package)
  • ROC curve for spatial proteomics data Tutorial
  • Tensor Decomposition Tutorial
  • Scalability Evaluation Tutorial

Installation

  1. Create a conda environment
conda create -n taichi-env
conda activate taichi-env
  1. Install the Taichi dependency
mamba install squidpy scanpy -c conda-forge (squidpy == 1.3.0 for reproducing CCI in manuscript)
pip insall pygsp ipykernel
  1. Install the MENDER for batch-free niches representation:
cd MENDER
python setup.py install

Install the pysodb for efficient download processed Anndata in h5ad format (https://pysodb.readthedocs.io/en/latest/) if you want to run the DKD and CRC related analysis

We suggest using mamba to install the dependencies. Installing the latest version of the dependencies may lead to dependency conflicts.

  1. Additional Spatial co-embedding methods

If you want to try Taichi with other spatial co-embedding methods (CellChater, STAGATE).You should install them first and run the following code on two simulation benchmarking

python run_sta.py (STAGATE)
python run_cc.py (CellCharter)

Contribution

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