nichepca 0.0.1

Package for PCA-based spatial domain identification in single-cell spatial transcriptomics data

NichePCA

Package for PCA-based spatial domain identification in single-cell spatial transcriptomics data. The corresonding manuscript was published in Bioinformatics.

Getting started

Given an AnnData object adata, you can run nichepca starting from raw counts as follows:

import scanpy as sc
import nichepca as npc

npc.wf.nichepca(adata, knn=25)
sc.pp.neighbors(adata, use_rep="X_npca")
sc.tl.leiden(adata, resolution=0.5)

Multi-sample support

If you have multiple samples in adata.obs['sample'], you can provide the key sample to npc.wf.nichepca this uses harmony by default:

npc.wf.nichepca(adata, knn=25, sample_key="sample")

If you have cell type labels in adata.obs['cell_type'], you can directly provide them to nichepca as follows (we found this sometimes works better for multi-sample domain identification). However, in this case we need to run npc.cl.leiden_unique to handle potential duplicate embeddings:

npc.wf.nichepca(adata, knn=25, obs_key='cell_type', sample_key="sample")
npc.cl.leiden_unique(adata, use_rep="X_npca", resolution=0.5, n_neighbors=15)

Customization

The nichepca functiopn also allows to customize the original ("norm", "log1p", "agg", "pca") pipeline, e.g., without median normalization:

npc.wf.nichepca(adata, knn=25, pipeline=["log1p", "agg", "pca"])

or with "pca" before "agg":

npc.wf.nichepca(adata, knn=25, pipeline=["norm", "log1p", "pca", "agg"])

or without "pca" at all:

npc.wf.nichepca(adata, knn=25, pipeline=["norm", "log1p", "agg"])

Setting parameters

We found that higher number of neighbors e.g., knn=25 lead to better results in brain tissue, while knn=10 works well for kidney data. We recommend to qualitatively optimize these parameters on a small subset of your data. The number of PCs (n_comps=30 by default) seems to have negligible effect on the results.

Installation

You need to have Python 3.10 or newer installed on your system. If you don't have Python installed, we recommend installing Mambaforge or Miniconda.

To create a new conda environment with Python 3.10:

conda create -n npc-env python=3.10 -y
conda activate npc-env

Install the latest development version:

pip install git+https://github.com/imsb-uke/nichepca.git@main

Contributing

If you want to contribute you can follow this guide. In short fork the repository, setup a dev environment using this command:

conda create -n npc-dev python=3.10 -y
conda activate npc-dev
git clone https://github.com/{your-username}/nichepca.git
pip install -e ".[dev, test]"

And then make your changes, run the tests and submit a pull request.

Release notes

See the changelog.

Contact

For questions, help requests, and bug reports, please use the issue tracker.

Citation

If you use NichePCA in your research, please cite:

@article{schaub2025pca,
  title={PCA-based spatial domain identification with state-of-the-art performance},
  author={Schaub, Darius P and Yousefi, Behnam and Kaiser, Nico and Khatri, Robin and Puelles, Victor G and Krebs, Christian F and Panzer, Ulf and Bonn, Stefan},
  journal={Bioinformatics},
  volume={41},
  number={1},
  pages={btaf005},
  year={2025},
  publisher={Oxford University Press}
}