Population size in QTL detection using quantile regression in genome-wide association studies.

Area of Science:

• Genetics
• Biostatistics
• Quantitative Genetics

Background:

• Genome-Wide Association Studies (GWAS) are crucial for identifying genetic variants associated with phenotypic traits.
• Traditional methods like the General Linear Model (GLM) may have limitations in detecting Quantitative Trait Loci (QTLs) under various conditions.
• Quantile Regression (QR) offers an alternative statistical approach for association analysis.

Purpose of the Study:

• To evaluate the performance of Quantile Regression (QR) compared to the General Linear Model (GLM) for QTL detection in GWAS.
• To assess the impact of population size and heritability on the power and accuracy of QR for identifying QTLs.
• To determine the effectiveness of QR in detecting trait-associated QTLs across different quantiles.

Main Methods:

• Simulated genetic data with varying heritability (0.30, 0.50) and numbers of QTLs (3, 100).
• Population sizes ranging from 1,000 to 200 individuals, with random reductions.
• Analysis using QR at quantiles 0.10, 0.50, and 0.90, and comparison with GLM.

Main Results:

• QR models consistently showed higher QTL detection power across all evaluated scenarios.
• QR exhibited a relatively low false positive rate, particularly in larger populations.
• Extreme quantiles (0.10, 0.90) in QR models yielded the highest detection power for true QTLs.
• GLM detected few or no QTLs, especially in smaller populations, while QR maintained high power even with low heritability.

Conclusions:

• Quantile Regression (QR) is an effective method for QTL detection in Genome-Wide Association Studies (GWAS).
• QR demonstrates superior performance over GLM, particularly in scenarios with limited sample sizes or low heritability.
• The use of QR facilitates the identification of QTLs associated with phenotypic traits, even in populations with fewer genotyped and phenotyped individuals.