Power of QTL detection by either fixed or random models in half-sib designs

Area of Science:

• Quantitative genetics
• Statistical genetics
• Bioinformatics

Background:

• Quantitative trait locus (QTL) mapping is crucial for understanding genetic architecture.
• Half-sib designs are commonly used in animal and plant breeding for genetic analysis.
• Comparing different statistical approaches is essential for optimizing QTL detection.

Purpose of the Study:

• To compare the variance component approach with the simple regression method for QTL linkage mapping in granddaughter designs.
• To evaluate the empirical power of these methods under various genetic and experimental parameters.

Main Methods:

• Monte Carlo simulation was employed to determine empirical power.
• Simulations included varying numbers of sires and sons per sire, QTL variance ratios, marker spacing, and QTL allele frequencies.
• A single bi-allelic QTL and six equally spaced markers were simulated.

Main Results:

• The variance component method consistently showed slightly higher empirical power than the regression method across different numbers of sires and QTL variance ratios.
• Power differences were marginal at high QTL variance ratios and allele frequencies.
• Linkage analysis in half-sib designs proved inadequate for fine-mapping QTL, especially at close marker spacing.

Conclusions:

• The variance component method offers a slight advantage over the simple regression method for QTL detection in half-sib designs.
• Effective QTL fine-mapping requires methods beyond simple linkage analysis in these designs.
• The findings provide insights for optimizing QTL mapping strategies in breeding programs.