Accurate and fast multiple-testing correction in eQTL studies.

Area of Science:

• Genetics
• Bioinformatics
• Statistical genomics

Background:

• Identifying expression quantitative trait loci genes (eGenes) is crucial for understanding gene regulation and biological processes.
• Current methods for detecting eGenes, particularly permutation tests for multiple-testing correction, face computational challenges with increasing sample sizes in eQTL studies.

Purpose of the Study:

• To develop an efficient and accurate computational method for correcting multiple testing in eQTL studies.
• To assess eGene p-values using a novel approach that accounts for linkage disequilibrium and is independent of sample size.

Main Methods:

• Proposed an efficient approach utilizing a multivariate normal distribution for multiple-testing correction.
• The method incorporates linkage disequilibrium structure among genetic variants.
• Time complexity is independent of sample size, enabling scalability for large datasets.

Main Results:

• The novel method achieves high accuracy (over 98%) in determining eGene p-values.
• Demonstrated consistent accuracy across three diverse human eQTL datasets (GTEx pilot, multi-region brain, HapMap 3) with varying sample sizes and SNP densities.
• The approach effectively addresses computational bottlenecks associated with large sample sizes.

Conclusions:

• The proposed method provides an accurate and computationally efficient solution for eGene detection in eQTL studies.
• This advancement facilitates more robust downstream analyses and prioritization of genes in genetic studies.
• The method's scalability and accuracy make it suitable for current and future large-scale genomic research.