Haruka

A Python package for identifying salient and background spatial domains in large-scale spatial omics analysis.

Haruka identifies salient and background spatial domains across multi-slice datasets and provides downstream analyses based on the discovered domains.

Getting started

Please refer to the

• MERFISH simulation Tutorial (Can be downloaded by )
• MERFISH brain aging dataset Tutorial (Can be downloaded via link)

Installation

1. Create a conda environment

conda env create -f environment.yml --name haruka
conda activate haruka

2. Install the modified scVI-tools dependency

cd scvi-tools
pip install -e ".[dev]"

3. Install the mini-batch version scENVI COVET calculation

pip install 'scenvi==0.4.5' --no-deps

(Optional)

Install MaxFuse link and scSLAT link to reproduce the downstream analysis

Acknowledgement:

This project is built based on the codebase of scVI link. CellCharter link and contrastiveVI link.

Contribution

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