OmicABEL

OmicABEL performs mixed-model–based genome-wide association studies (GWAS) to account for population structure and relatedness and to improve power and accuracy in single- and multiple-trait omics analyses.

Key Features:

• Mixed-Model Based Analysis: OmicABEL employs mixed models to account for population structure and relatedness in GWAS, reducing false positives.
• Single and Multiple Trait Analysis: Supports both single-trait and multiple-trait analyses across genomics, transcriptomics, proteomics and other omics data types.
• Optimized Computational Algorithms: Implements optimal algorithms tailored for single- and multiple-trait scenarios, leveraging state-of-the-art linear algebra kernels and minimizing redundant computations.
• High-Performance Computing Framework: Built on a high-performance computing framework that enhances throughput while reducing memory usage and energy consumption compared to existing libraries.

Scientific Applications:

• Statistical Genomics: Application to GWAS in large cohorts and structured populations where mixed-model correction is required.
• Multi-trait GWAS: Analysis of arbitrary numbers of correlated traits within association studies while retaining computational efficiency.
• Omics Studies: Use in genomics, transcriptomics and proteomics investigations that require correction for relatedness and population structure.

Methodology:

Uses advanced linear algebra techniques and algorithmic optimizations specific to mixed-model GWAS, including state-of-the-art linear algebra kernels and strategies to minimize redundant computations to handle arbitrary numbers of traits.