A Semi-Supervised Category Identification and Assignment Tool for single-cell RNA-Seq and Cytof/FACS data.

You can install SCINApy from PyPI using pip:

pip install SCINApy

SCINApy requires Python 3.7 or higher and the following dependencies:

• numpy>=2.1.2
• pandas>=2.2.3
• scipy>=1.15.3
• anndata>=0.11.4
• seaborn>=0.13.2
• matplotlib>=3.10.0

SCINApy provides tools for cell type assignment and visualization of single-cell data. Key features include:

• SCINA: Core algorithm for semi-supervised cell type identification.
• plotheat_scina: Visualize SCINA results with a heatmap.
• Command-line interface via scinapy for easy execution.(still working in progress!!!)

Run the command-line interface with sample data:

scinapy --data data/matrix.csv --signatures data/signatures.json --output results.pkl --job_id test

An example Jupyter Notebook (example.ipynb) is included to demonstrate the usage of SCINApy.

The SCINA algorithm implemented in this package is based on the methodology originally developed by Zhang et al. (2019), where the technical details are comprehensively elaborated (SCINA: A Semi-Supervised Subtyping Algorithm of Single Cells and Bulk Samples). This package, SCINApy, is also developed based on the same author's R package SCINA, adapting its functionality to the Python ecosystem for enhanced usability and integration with modern single-cell analysis tools.

• PyPI Page: SCINApy 0.1.1
• Source Code: GitHub Repository
• Issues: Report Issues

• Fixed ValueError in SCINA function when anndata.X is a numpy.ndarray.

• Fixed ValueError in SCINA function when no signature genes are found in adata.

• Initial release of SCINApy with core SCINA algorithm and visualization tools.

This project is licensed under the MIT License.

Contributions are welcome! Please submit issues or pull requests on the GitHub repository.

For support or questions, please open an issue on GitHub or contact the author at hwr9912@gmail.com.