TwinEQTL

TwinEQTL performs genome-wide association study (GWAS) and expression quantitative trait loci (eQTL) analyses in twin study datasets by partitioning twin samples and integrating results to account for genetic relatedness.

Key Features:

• Efficient Data Partitioning: Partitions twin samples into two independent groups and performs separate multiple linear regression analyses on each group to reduce computational complexity compared with linear mixed-effects models.
• Meta-Analysis Approach: Integrates the results from the two non-independent test statistics using a meta-analysis-like technique to combine evidence while accounting for relatedness.
• Mathematical Validation: Includes mathematical derivations demonstrating that the correlation between dependent test statistics at each single-nucleotide polymorphism (SNP) is constant across SNPs and independent of minor allele frequency (MAF).
• Controlled Type I Error: Empirical simulations confirm well-controlled type I error rates with negligible loss of statistical power relative to traditional linear mixed-effects models.
• Enhanced Computational Efficiency: Implemented as an R package to improve computational speed for GWAS and eQTL analyses involving twin samples.

Scientific Applications:

• GWAS in twin cohorts: Identifying genetic variants associated with traits or diseases in twin study populations.
• eQTL mapping in twin cohorts: Mapping expression quantitative trait loci within twin datasets to link genetic variation to gene expression.
• Heritability and gene–environment studies: Estimating heritability and exploring gene–environment interaction effects using related samples.

Methodology:

Partitions twin samples into two groups, runs separate multiple linear regression analyses per group, integrates the two non-independent test statistics via a meta-analysis-like technique, validates method behavior by mathematical derivation showing constant SNP-wise correlation independent of MAF, and assesses type I error and power using empirical simulations; implementation provided as an R package.