Gelex: High-Performance Genomic Analysis

Gelex is a high-performance C++ library and CLI tool for genomic prediction and genome-wide association studies (GWAS). It integrates advanced Bayesian models (BayesAlphabet series) and frequentist approaches (GBLUP) with memory-mapped genotype data, delivering state-of-the-art performance for large-scale genomic datasets.

Quick Links

• Installation - Get Gelex running on your system.

• GWAS Tutorial - Step-by-step guide to running your first GWAS.

• Command Line Interface - Comprehensive command-line reference.

Note

This project is under active development. APIs and features are subject to change.

Installation

Install the latest version via pixi (recommended) or conda:

# Using pixi (Global install)
pixi global install -c conda-forge -c https://prefix.dev/gelex gelex

# Using conda
conda install -c conda-forge -c https://prefix.dev/gelex gelex

Quick Start

Here is how to fit a Bayesian model (BayesR) in one command:

gelex fit \
  --bfile data/genotypes \
  --pheno data/phenotypes.tsv \
  --method R \
  --iters 10000 \
  --burnin 2000 \
  --o result/my_analysis

For more examples, check out the GWAS Tutorial.

Key Features

• Bayesian Models: Full BayesAlphabet suite (A, B, C, R, RR) with dominance effect variants.

• Frequentist Models: GBLUP with REML-based variance component estimation.

• GWAS: Mixed linear model-based association testing with LOCO correction.

• High Performance: AVX512/AVX2 vectorized I/O, OpenMP parallel processing, and optimized MKL/OpenBLAS backends.

• Memory Efficiency: Memory-mapped BED file reading with chunk-based processing.

Getting Started

• Installation
  • Binary Installation (Recommended)
  • Build from Source
  • Verification
• GWAS Tutorial
  • Background
  • Step 1: Compute GRM
  • Step 2: Association Analysis
  • Output Format
• Genomic Selection Tutorial
  • Background
  • Workflow Overview
  • Step 1: Model Fitting
  • Step 2: Genomic Prediction
  • Output Format

Reference

• Data Formats
  • Quick Reference
  • Input Data Formats
  • Output Data Formats
• Command Line Interface
  • Available Subcommands
  • Global Options
  • Basic Usage Pattern
• API Reference
• Genotype Processing Methods
  • What --geno-method Changes
  • HWE vs Sample: Practical Difference
  • Quick Selection Guide
  • Method Families
  • Orthogonal vs Non-orthogonal Dominance
  • Method Matrix (User View)
  • Practical Recommendations
  • Minimal Technical Notes
  • Example Commands
  • See Also

Citing Gelex

Citation

Please use the following BibTeX template to cite Gelex in scientific discourse:

@misc{gelex,
   author = {RuLei Chen},
   year = {2026},
   note = {https://github.com/r1cheu/gelex},
   title = {Gelex: A high-performance C++ genomic analysis toolkit}
}