EMMAX

EMMAX implements an efficient linear mixed-model association method to perform genome-wide association studies while accounting for genetic relatedness and sample structure.

Key Features:

• Variance Component Approach: Employs a variance component mixed model that uses pairwise relatedness estimated from high-density markers to account for genetic relatedness.
• Computational Efficiency: Builds on the EMMA algorithm to expedite computations and substantially reduce runtime for large GWAS datasets by avoiding repetitive per-locus re-estimation of variance components.
• Correction for Sample Structure: Corrects for population structure and cryptic relatedness, providing calibration superior to principal component analysis and genomic control in many settings.
• Application to Large Datasets: Applicable to large human GWAS datasets and has been applied to cohorts such as the Northern Finland Birth Cohort and the Wellcome Trust Case Control Consortium covering quantitative traits and common diseases.

Scientific Applications:

• Genome-wide association studies (GWAS): Performs association testing across the genome while controlling for sample structure and relatedness.
• Complex trait mapping: Facilitates identification of genetic variants influencing quantitative traits and common diseases by accounting for polygenic background and relatedness.
• Large-cohort analyses: Enables analysis of extensive human cohorts, as demonstrated with the Northern Finland Birth Cohort and the Wellcome Trust Case Control Consortium.

Methodology:

Builds on the efficient mixed-model association (EMMA) framework and avoids repetitive variance component estimation across loci by estimating variance components once and reusing them, with relatedness derived from high-density markers.