Home

Home

Welcome to the gcPCA Wiki!

Generalized contrastive PCA (gcPCA) is a new dimensionality reduction method designed to compare high-dimensional datasets collected under different experimental conditions. Unlike PCA, which works on a single dataset, gcPCA directly compares two conditions to highlight key differences.

🔹 Key Features

• Hyperparameter-free: No manual tuning required.
• Symmetric comparison: Both conditions are treated equally.
• Sparse solutions: Helps simplify and interpret results.
• Multi-language support: Implemented in Python and MATLAB, with R implementation coming soon.

📖 More details in our paper: PLOS Computational Biology

🔗 Quick Links

• Installation
• Usage Guide
• FAQ

---

Installation

Python Installation (via Conda)

1. Install Conda.
2. Download this repository.
3. Open a conda terminal and navigate to your copy of the repository.
4. Run the following command:

   conda env create -f environment.yml