Pure Python implementation of scPPIN for single-cell protein-protein interaction network analysis
scPPIN-py detects functional modules in protein-protein interaction networks by integrating single-cell RNA sequencing data. This is a reimplementation of the original R package with an object-oriented Python API.
Original method: Klimm et al. (2020), BMC Genomics
- Fast — Vectorized NumPy operations and igraph backend for speed and efficiency
- Class-Based API — Object-oriented design with method chaining
- Edge Weights — Supports pre-computed edge weights from dictionaries
- Scanpy Integration — Works with AnnData objects for p-value extraction
- Easy Installation — Single
pip installcommand - Standalone PCST — Direct PCST implementation without dependencies on expression data
# Install directly from GitHub
pip install git+https://github.com/shahrozeabbas/scppin-py.git
# Or clone and install locally
git clone https://github.com/shahrozeabbas/scppin-py.git
cd scppin-py
pip install .from scppin import scPPIN
# Create model and load network
model = scPPIN()
model.load_network('edges.csv')
# Set node weights (p-values from differential expression)
pvalues = {'TP53': 0.0001, 'MDM2': 0.001, 'CDKN1A': 0.005}
model.set_node_weights(pvalues)
# Optionally set edge weights (from pre-computed dictionary)
edge_weights = {('TP53', 'MDM2'): 0.9, ('TP53', 'CDKN1A'): 0.8}
model.set_edge_weights(weights=edge_weights)
# Detect functional module using PCST
model.detect_module(fdr=0.01, edge_weight_attr='weight')
# Visualize
model.plot_module()Full documentation: https://scppin-py.readthedocs.io
If you use scPPIN-py in your research, please cite the original paper:
@article{klimm2020functional,
title={Functional module detection through integration of single-cell RNA sequencing data with protein--protein interaction networks},
author={Klimm, Florian and Toledo, Enrique M and Monfeuga, Thomas and Zhang, Fang and Deane, Charlotte M and Reinert, Gesine},
journal={BMC Genomics},
volume={21},
number={1},
pages={756},
year={2020},
publisher={BioMed Central},
doi={10.1186/s12864-020-07144-2}
}GPL-3.0 (same as original R package)