Installing SAIGE-QTL

Choose the installation method that best fits your environment and requirements. We recommend Docker for most users as it works on all platforms with zero setup, followed by Pixi Binary for Linux and Mac users seeking the fastest installation.

Current Version: 0.3.5

Updated: February 2, 2025

Installation Methods (Ordered by Ease of Use)

🥇 Easiest: Docker/Singularity (Recommended)

Zero setup, works on Linux and macOS - everything included in the container.

→ Install with Docker

🥈 Easy: Pixi Binary

Fastest installation for Linux and Mac - no compilation required, pre-built packages.

→ Install with Pixi Binary

🔧 Advanced: Source Code

Requires specific system configurations, unless using the pixi source-code installation approach outlined in this page.

→ Install from Source

Platform Support & Quick Start

Platform Support Level Recommended Methods Notes
Linux ✅ Full Support Docker/Singularity → Pixi Binary → Source Pre-built binaries available
macOS ✅ Good Support Docker/Singularity → Pixi Binary → Source No pre-built binaries
Windows ⚠️ Limited Support WSL2 + Linux methods Not natively supported

Installation Comparison

Method Platform Setup Time Dependencies Best For
Docker/Singularity All platforms ~5-10 minutes Docker Image only Zero setup, containers, HPC
Pixi Binary Linux, macOS ~5 minutes Pixi + GLIBC 2.28+ Fastest installation from pre-built binaries
Pixi Source Linux, macOS ~15 minutes Pixi only Managed dev environment
Source Linux, macOS ~30 minutes C++ compiler + libraries Traditional development, requires specific system configuration

System Requirements

Minimum Requirements

  • OS: Linux andmacOS (can be attempted on Windows with WSL)
  • RAM: 8GB minimum, 16GB+ recommended for large datasets
  • Storage: 2GB for software, additional space for data
  • R: Version 4.0+ (installed automatically with Pixi/Docker)
  • RAM: 32GB+ for large-scale analyses
  • CPU: Multi-core processor for parallel processing
  • Storage: SSD storage for improved I/O performance

📋 Complete Changelog

For detailed logs of all bug fixes and improvements, see the Installation Logs.

Post-Installation

Verify Installation

After installation, verify SAIGE-QTL is working correctly:

  1. Check help information: 
    # Replace with your installation method's command prefix
Rscript step1_fitNULLGLMM_qtl.R --help
  2. Run example analysis: Follow the Step 1 tutorial with provided example data

Next Steps

  1. Review the workflow overview - Understand the analysis pipeline
  2. Learn how to call SAIGE-QTL - Execute scripts in different environments
  3. Start with Step 1 - Begin your first analysis

Getting Help

Installation Issues

  • Check the specific installation guide for your chosen method
  • Verify system requirements are met
  • Review error messages carefully - they often contain helpful information

Technical Support

  • Email: wzhou@broadinstitute.org
  • GitHub Issues: For bug reports and feature requests
  • Documentation: Browse all available guides in the navigation menu

Changelog

Version 0.3.5 (February 2, 2025)

  • New: Added --solverMethod option allowing step 1 null modeling fitting using either Sherman-Morrison-Woodbury approach (“smw” option) or the original preconditioned conjugate gradient approach (“pcg” option). For data sets with unrelated donors, for which the sparse GRM becomes the identity matrix, SMW further reduces to closed-form block-wise operations with optimal complexity. The default is to automatically detect given phenotype file, if multiple rows of cell data for any individual is provided, it will automatically use the SMW approach.
  • Performance: Optimized matrix inversion algorithm (SMW) for faster and user-specified model fitting
  • Improved: Enhanced Docker support with updated documentation and cross-platform compatibility
  • Fixed: Various stability and performance improvements

Version 0.3.4 (January 11, 2025)

  • New: Comprehensive installation documentation with platform-specific guidance and troubleshooting
  • New: Binary installation method for faster Linux setup via Pixi
  • Performance: Optimized matrix inversion algorithms for faster model fitting
  • Performance: Improved memory efficiency for large-scale analyses
  • Improved: Enhanced Docker support with updated documentation and cross-platform compatibility
  • Fixed: Various stability and performance improvements

Version 0.3.2 (July 28, 2025)

  • New: Added --offsetCol option for using log of total read counts per cell as an offset in the model
  • Improved: Enhanced installation process using pixi
  • Fixed: Various bug fixes and stability improvements

July 31, 2025:

  • Export extdata/ in the docker container so users do not need to download the git repo when using the docker image wzhou88/saigeqtl:latest on example data
This site uses Just the Docs, a documentation theme for Jekyll.