scxpand 0.1.0.dev1

Pan-cancer detection of T-cell clonal expansion from single-cell RNA sequencing

scXpand

Pan-cancer detection of T-cell clonal expansion from single-cell RNA sequencing without paired single-cell TCR sequencing

Documentation • Installation • Quick Start • Usage Examples • Data Format • Output Format • Model Architectures • Citation

A framework for predicting T-cell clonal expansion from single-cell RNA sequencing data.

Manuscript in preparation - detailed methodology and benchmarks coming soon.

View full documentation for comprehensive guides and API reference.

Features

• Multiple Model Architectures: Autoencoder, MLP, LightGBM, Logistic Regression, and SVM for comprehensive analysis
• Scalable Processing: Handles millions of cells with memory-efficient data streaming from disk during training
• Automated Hyperparameter Optimization: Built-in Optuna integration for model tuning

Installation

pip install scxpand

Quick Start

import scxpand

# List available pre-trained models
scxpand.list_pretrained_models()

# Run inference with automatic model download
results = scxpand.run_inference_with_pretrained(
    model_name="autoencoder_pan_cancer",
    data_path="your_data.h5ad"
)

Or via command line:

# Pre-trained model inference (curated models)
scxpand predict --data_path your_data.h5ad --model_name autoencoder_pan_cancer

# Direct DOI inference (any Zenodo model - seamless sharing!)
scxpand predict --data_path your_data.h5ad --model_doi 10.5281/zenodo.1234567

# Local model inference
scxpand predict --data_path your_data.h5ad --model_path results/my_model

Development

For development installation and model training, see the documentation.

Model Architectures

scXpand provides multiple model architectures to suit different use cases and data characteristics:

Autoencoder-based Classifiers

Architecture featuring an encoder with auxiliary decoder for reconstruction and classifier head for expansion prediction. This approach leverages representation learning to capture complex patterns in single-cell data.

Multi-Layer Perceptron (MLP)

Standard feed-forward neural networks for direct expansion prediction.

LightGBM

Gradient boosting for classification tasks with excellent performance on tabular data.

Linear Models

Classical machine learning approaches including logistic regression and support vector machines.

License

This project is licensed under the MIT License – see the LICENSE file for details.

Citation

If you use scXpand in your research, please cite:

@article{scxpand2024,
  title={scXpand: Pan-cancer detection of T-cell clonal expansion from single-cell RNA sequencing without paired single-cell TCR sequencing},
  author={[Your Name]},
  journal={[Journal Name]},
  year={2024},
  doi={[DOI]}
}

This project was created in favor of the scientific community worldwide, with a special dedication to the cancer research community. We hope you’ll find this repository helpful, and we warmly welcome any requests or suggestions - please don’t hesitate to reach out!