Cancer-specific expression quantitative loci are affected by expression dysregulation

Area of Science:

• Genomics
• Systems Biology
• Genetic Epidemiology

Background:

• Expression quantitative trait loci (eQTLs) link genetic variants to gene expression, crucial for understanding phenotypes.
• Reproducibility of eQTLs is vital for their reliable application in genetic studies.
• Existing eQTL databases primarily use normal tissue data, limiting their use in disease contexts with altered gene expression.

Purpose of the Study:

• To evaluate the reproducibility of eQTLs across different tissue types (normal vs. tumor).
• To assess the impact of disease-specific gene expression on eQTL identification.
• To investigate the concordance of eQTLs between major databases like GTEx and TCGA.

Main Methods:

• Conducted an eQTL study on 5178 samples from The Cancer Genome Atlas (TCGA).
• Compared eQTLs identified in normal tissues versus tumor tissues.
• Analyzed eQTL directional concordance between GTEx and TCGA datasets.

Main Results:

• Low reproducibility of eQTLs was observed between normal and tumor tissues.
• Shared eQTLs exhibited generally concordant effect directions.
• Good directional concordance was found between GTEx and TCGA eQTLs.
• Multi-tissue eQTLs can display opposing effects across different tissue types.

Conclusions:

• Tissue source and disease status significantly influence detectable eQTLs and their effect directions.
• eQTL prediction models must account for tissue and disease-specific dependencies.
• Development of tissue-disease-specific eQTL databases is recommended for improved prediction accuracy.