Spanve: A Statistical Method for Detecting Downstream-friendly Spatially Variable Genes in Large-scale Spatial Transcriptomics Data

Citation

@preprint{b.23.SpanveStatistical,
  title = {Spanve: An {{Statistical Method}} to {{Detect Clustering-friendly Spatially Variable Genes}} in {{Large-scale Spatial Transcriptomics Data}}},
  author = {{Guoxin Cai} and {Yichang Chen} and {Shuqing Chen} and {Xun Gu} and {Zhan Zhou}},
  date = {2023-01-01},
  journaltitle = {bioRxiv},
  pages = {2023.02.08.527623},
  doi = {10.1101/2023.02.08.527623},
  url = {http://biorxiv.org/content/early/2023/03/08/2023.02.08.527623.abstract},
}

Analysis code for the paper is available at Evaluate directory.

Installation

• Install by pip (recommend):

pip install Spanve

• no install usage:

# install required packages
cd Evaluate/Softs
pip install spanve_requirements.txt
cp Spanve.py $your_path

Usage

cli usage

spanve --help

Usage: Spanve [OPTIONS]

Options:
  -i, --input_file PATH       input anndata file(h5ad file.)
  -r, --running_mode TEXT     running mode, default is f(c:cluster;
                              i:imputation; f:fitting)
  -s, --save_path PATH        save path
  -v, --verbose BOOLEAN       verbose
  -n, --njobs INTEGER
  -p, --preprocessed BOOLEAN  int preprocessed or not.
  --help                      Show this message and exit.****

command line usage can only run in a standard h5ad file, where there is a anndata.AnnData.obsm key named 'spatial'.

python usage

Quick Start

from Spanve import Spanve
adata = sc.read_h5ad('data.h5ad')
spanve = Spanve(adata)
# -- fitting for spatial genes
spanve.fit()
spanve.save('result.csv')
# -- spatial imputation
X = adata_preprocess(adata)
X_ = spanve.impute_from_graph(X[:,spanve.rejects])

Details

see tutorial notebook or html page.