A fast expectation-maximum algorithm for fine-scale QTL mapping.

Area of Science:

• Genetics and Genomics
• Statistical Genetics
• Bioinformatics

Background:

• Single-nucleotide-polymorphism (SNP) arrays enable high-density genotyping, necessitating efficient fine-scale quantitative trait loci (QTL) mapping methods.
• Single-marker association (SMA) is simple but suffers from low power and high false-positive rates.
• Variance component (VC) methods offer higher power but are computationally intensive and difficult to converge.

Purpose of the Study:

• To develop a computationally fast and efficient method for fine-scale QTL mapping.
• To compare the performance of the proposed method against existing approaches like VC and SMA.

Main Methods:

• Development of an Expectation-Maximization algorithm under Fixed effect model (EMF) for biallelic QTL.
• Utilized an expectation-maximization (EM) algorithm to solve model effects.
• Simulation experiments were conducted to evaluate EMF's performance.

Main Results:

• EMF demonstrated significantly faster computation compared to the VC method.
• EMF and VC methods exhibited similar QTL detection power and parameter estimation accuracy.
• Both EMF and VC outperformed paired-marker analysis and SMA in simulations.

Conclusions:

• EMF provides a computationally efficient alternative for fine-scale QTL mapping with performance comparable to VC methods.
• EMF effectively reduces false-positive signals and enhances QTL detection power.
• The performance of EMF may decrease for multiallelic QTL compared to VC methods.