Restricted maximum likelihood estimation of additive genetic variance when selected base animals are considered fixed

Area of Science:

• Animal Breeding and Genetics
• Quantitative Genetics
• Statistical Genetics

Background:

• Accurate estimation of genetic parameters is vital for effective animal breeding programs.
• Traditional models may not adequately account for selection in base populations, potentially biasing results.
• Understanding the impact of base animal selection on genetic variance estimation is essential.

Purpose of the Study:

• To investigate a method for estimating genetic parameters using a model that treats selected base animals as fixed.
• To evaluate the bias introduced by assuming base animals as fixed versus random in genetic variance estimation.
• To assess the impact of selection on descendants of fixed base animals.

Main Methods:

• A simulation study was conducted with five generations of selection.
• Genetic variance was estimated using a model where base animals were treated as fixed and compared to a random model.
• Performance records were available for Generations 4 and 5.
• The Mendelian sampling of gametes from base parents was considered.

Main Results:

• Estimated genetic variance was lower when base animals were assumed fixed (6.03) compared to random (8.58), with simulated values of 10.
• Residual variance was overestimated when base animals were assumed fixed.
• Bias was introduced when descendants of fixed base animals were selected, but not by the selection of base animals themselves.
• Real-world dairy data showed up to 8% differences in genetic variance estimation after accounting for selected base animals.

Conclusions:

• A model assuming fixed base animals can provide unbiased estimates of genetic parameters if selection in descendants is properly handled.
• Overestimation of residual variance occurs when base animals are fixed and their descendants are selected.
• Adjusting for selected base animals in genetic parameter estimation, particularly in dairy cattle, can significantly improve accuracy.