A new selection index percent emphasis method using subindex weights and genetic evaluation accuracy

Area of Science:

• Animal breeding and genetics
• Quantitative genetics
• Selection index theory

Background:

• Current selection index methodology quantifies trait emphasis using economic value and genetic standard deviation.
• This standard method fails to account for trait evaluation accuracy and genetic correlations, misrepresenting actual selection emphasis.
• Economic values do not accurately reflect selection effort when trait accuracies vary significantly.

Purpose of the Study:

• To propose and evaluate a novel method for quantifying trait percent emphasis in selection indexes.
• The new method aims to incorporate trait accuracy and genetic correlations for a more practical estimation of selection emphasis.
• To compare the proposed method against the conventional approach using a real-world sheep breeding selection index.

Main Methods:

• A hierarchical clustering method is applied to the genetic correlation matrix to group genetically correlated traits into subindexes.
• Trait emphasis within subindexes is calculated, incorporating a weighting for trait accuracy.
• Subindex emphasis is then converted to full index emphasis based on subgroup relative emphasis.

Main Results:

• The new method significantly reduced emphasis on low-heritability traits (e.g., survival from 51% to 19%) and increased emphasis on growth traits (from 30% to 49%) in a New Zealand sheep index.
• Accounting for accuracy adjusted within-subindex trait rankings, while clustering for correlations affected all traits within a subgroup equally.
• The method's effectiveness in reflecting actual selection pressure was demonstrated, particularly when trait genetic correlations were distinct.

Conclusions:

• The proposed method, incorporating trait accuracy and genetic correlations, offers a more practical and realistic assessment of selection index emphasis.
• This approach provides a better indication of the likely outcomes of selection based on the index.
• The standard method's limitations are highlighted, particularly in scenarios with varying trait accuracies and distinct genetic correlations.