Expression Quantitative Trait Loci (eQTL)

eQTLs are classified by their genomic proximity to the target gene:

• Cis-eQTLs: Located within ~1 Mb of the gene's transcription start site. They typically affect local chromatin structure or promoter activity.
• Trans-eQTLs: Located far from the target gene, often on different chromosomes. They usually encode transcription factors or signaling molecules that regulate distant gene networks.

Cis-eQTLs generally have larger effect sizes and are more replicable, making them preferred instruments in Cis-Mendelian Randomization studies.

A defining feature of eQTLs is their context-dependent activity. The same genetic variant may strongly associate with gene expression in liver tissue but have no effect in brain tissue.

This tissue-specificity arises from:

• Differential chromatin accessibility across cell types
• Tissue-specific transcription factor availability
• Cell-type-specific enhancer-promoter looping

Projects like GTEx (Genotype-Tissue Expression) have cataloged eQTLs across 49 human tissues, enabling tissue-aware causal inference.

eQTLs are often detected through allele-specific expression (ASE) analysis, which measures the relative expression of maternal vs. paternal alleles within a heterozygous individual.

This approach:

• Uses the individual as their own control, eliminating environmental confounding
• Directly demonstrates cis-regulatory effects
• Provides higher statistical power than total expression QTL mapping

ASE is particularly valuable for identifying rare regulatory variants that would be missed by standard eQTL studies requiring large sample sizes.

eQTL mapping involves systematic association testing between millions of genetic variants and thousands of gene expression phenotypes:

• Linear regression: Tests additive genetic effects on log-transformed expression
• Matrix eQTL: Efficient R package for ultra-fast eQTL analysis using matrix operations
• Probabilistic estimation of expression residuals (PEER): Removes hidden confounding factors like batch effects and cell-type composition

Multiple testing correction is critical—a typical study tests >10^10 variant-gene pairs, requiring stringent Bonferroni or Benjamini-Hochberg thresholds.

A powerful application of eQTL data is colocalization analysis, which tests whether the same causal variant drives both gene expression and complex disease risk.

Methods like COLOC and eCAVIAR compute posterior probabilities for competing hypotheses:

• H0: No association with either trait
• H3: Two distinct causal variants (no shared mechanism)
• H4: A single shared causal variant (evidence for a causal gene)

Strong colocalization (PP.H4 > 0.8) nominates target genes for drug development and validates Drug Target Mendelian Randomization instruments.

Response eQTLs are genetic variants that modulate gene expression changes in response to an environmental stimulus, drug treatment, or disease state.

Unlike static eQTLs measured at baseline, reQTLs capture gene-environment interactions:

• A variant may show no effect in untreated cells but strongly influence expression after immune stimulation
• reQTLs reveal context-specific regulatory mechanisms invisible in steady-state data
• They are critical for understanding differential drug response and personalized therapeutic strategies

This dynamic dimension makes reQTLs particularly valuable for pharmacogenomics and precision medicine applications.

An instrumental variable analysis method that uses genetic variants, such as eQTLs, as proxies for gene expression levels to estimate the causal effect of transcript abundance on complex traits. By leveraging the random assortment of alleles at conception, MR minimizes confounding and reverse causation, treating eQTLs as natural genetic instruments to test if a gene's expression directly influences disease risk.

A statistical method to assess whether an eQTL and a GWAS hit for a complex trait share the same causal genetic variant at a specific genomic locus. Colocalization distinguishes between a shared causal mechanism—where a variant alters gene expression to drive disease—and mere linkage disequilibrium, providing stronger evidence that a specific transcript mediates the observed phenotypic association.

An analytical framework that integrates eQTL reference panels with GWAS summary statistics to identify genes whose genetically predicted expression levels are significantly associated with a complex trait. TWAS imputes transcript abundance into large genotyped cohorts, effectively bridging the gap between genetic variation and phenotypic outcome by testing for expression-trait associations without requiring direct transcriptomic measurement in the outcome study.

A specialized MR design that restricts genetic instruments to cis-eQTLs—variants located within a narrow window around the gene's transcription start site. This approach minimizes horizontal pleiotropy because the biological mechanism is constrained to the local regulatory machinery of the target gene, providing a more robust and specific causal estimate of a protein or transcript's effect on disease compared to genome-wide instruments.

A violation of the exclusion restriction in MR where an eQTL influences the outcome through pathways independent of the gene expression under investigation. For example, a variant might affect both the expression of Gene A and directly modify the function of Protein B. Methods like MR-Egger regression and MR-PRESSO are specifically designed to detect and correct for this bias when using eQTL instruments.

The foundational econometric technique underlying eQTL-based causal inference. A valid instrument (the eQTL) must satisfy three core assumptions: (1) it is robustly associated with the exposure (gene expression), (2) it is independent of confounders, and (3) it affects the outcome only through the exposure. Weak instrument bias occurs when eQTLs explain only a small fraction of expression variance, requiring large sample sizes.