Comparative Evaluation of Genetic Gains through Selection Methods

Introduction

The evaluation of genetic gains following one or more generations of selection introduces challenges, primarily stemming from the variability of generation means. Generation means fluctuate erratically, making accurate measurement challenging. The average response per generation is best measured through the slope of a regression line fitted to the generation means, addressing the inherent variability.

Causes of Variation

Several causes contribute to the variation in response between generations, including random genetic drift, sampling errors in estimating generation means, differences in selection differentials, and environmental factors. Mitigating variation due to random drift and sampling errors requires increasing the number of selected and measured individuals. Environmental differences can be addressed by maintaining an unselected control population, which estimates genetic improvement. However, the use of a control may compromise precision due to limited experimental scale.

Two-Way or Divergent Selection:

• The relative accuracy of response estimation can be improved by employing two-way or divergent selection, where the control is selected in the opposite direction. 
• This approach reduces some of the variation between generations, enhancing precision. However, an unselected control remains preferable if the interest lies solely in one direction of change.

Environmental Trends and Challenges:

• Random changes in the environment decrease estimation precision but do not invalidate the estimate. More significant challenges arise from environmental trends, such as progressive changes over time.
• Distinguishing between responses to selection and environmental trends becomes difficult without a control. Assessing the effectiveness of selection in improving domestic animals and plants becomes complex, particularly when environmental trends are present.

Overlapping Generations:

• When generations overlap, as seen with farm animals, it becomes possible to address some of the challenges associated with environmental trends. 
• Overlapping generations offer a way to discern the contributions of selection and improved management to observed improvements.

Selection Differential Weighing

• In experimental selection, the measurement of the response and selection differential is crucial, as it helps understand the genetic perspective. 
• Distinguishing between the expected and effective selection differential is essential, considering that not all individual parents contribute equally to the next generation. The weighting of the selection differential involves considering the deviations of parents based on the number of offspring measured.

Effects of Natural Selection

• Weighting the selection differential accounts for a significant portion of natural selection effects. Differences in fertility related to parental phenotypic values impact the joint effects of natural and artificial selection. 
• By comparing the effective (weighted) and expected selection differentials, insights into the operation of natural selection can be gained.

Realized Heritability

• The response per generation reflects the outcomes, but a mean is necessary to understand the relationship between the response and the selection differential. The response can be expressed as a proportion of the selection differential, offering insights into the impact of selection. 
• Plotting generation means against the cumulated selection differential helps visualize the total selection applied over successive generations.

Estimating Heritability

• The response to selection serves as an estimate of heritability in the base population. The ratio of the response to the selection differential (BS) provides valuable information. 
• The heritability estimated through this approach, termed realized heritability, primarily describes the response and may not always accurately reflect the heritability in the base population due to factors like environmental trends or inbreeding depression.

Addressing Random Drift Effects

Random drift effects can complicate the estimation of heritability. Replication of the selection in different directions (upward and downward) allows for a more comprehensive understanding of realized heritabilities, providing valid descriptions of the response and offering insights into the base population's heritability.