A mixed model method to predict QTL-cluster effects using trait and marker information in a multi-group population

Area of Science:

• Animal genetics
• Quantitative genetics
• Statistical genetics

Background:

• Genetic evaluation of crossbred animals is complex due to diverse founder genetic groups.
• Accurate genetic parameter estimation is crucial for effective animal breeding programs.
• Existing methods may not fully capture the effects of linked quantitative trait loci (QTLs).

Purpose of the Study:

• To develop a mixed model method for genetic evaluation in crossbred populations.
• To incorporate marker information for linked QTLs into the genetic evaluation model.
• To improve the accuracy of genetic predictions by considering founder genetic group and marker data.

Main Methods:

• A mixed model approach was developed utilizing trait phenotype and marker data.
• The method considers a cluster of QTLs flanked by two markers.
• It accounts for conditional expectation of identity-by-descent proportion and genetic variances/covariances.
• The segregation variance structure differs from single-QTL models.

Main Results:

• The method provides best linear unbiased estimation (BLUE) of fixed effects.
• It offers best linear unbiased prediction (BLUP) of additive effects for marked QTL clusters and remaining polygenes.
• The approach properly accounts for genetic group and marker information.
• A numerical example demonstrates the prediction procedure's efficacy.

Conclusions:

• The developed mixed model method offers enhanced genetic evaluation for crossbred animals.
• Incorporating linked QTL information and genetic group data improves prediction accuracy.
• This method provides a robust framework for complex genetic evaluations in livestock populations.