Planning incomplete block experiments when treatments are genetically related

Area of Science:

• Agricultural science
• Genetics
• Statistical genetics

Background:

• Selection trials in plant and animal breeding often use incomplete block designs analyzed with linear models.
• These models incorporate random effects for treatments with known genetic covariance structures.
• Information on genetic relatedness among treatments can improve analysis, but its impact on block design optimality is understudied.

Purpose of the Study:

• To investigate the consequences of genetical relatedness among treatments on the optimality of incomplete block designs.
• To determine if designs optimal for unrelated treatments are also optimal when relatedness is considered.

Main Methods:

• Utilized a suitable optimality criterion to evaluate block designs.
• Analyzed linear models with random effects parameters reflecting genetic covariance structure.
• Compared optimal designs under relatedness versus unrelatedness assumptions.

Main Results:

• Knowledge of genetical relatedness among treatments can alter the optimal block design.
• The optimal design considering relatedness may differ from designs optimal for unrelated treatments.
• This suggests a need to reconsider design choices in breeding programs.

Conclusions:

• Genetical relatedness is a crucial factor influencing the choice of optimal block designs in breeding trials.
• Standard design selection methods may not be appropriate when treatments are genetically related.
• Practical implications include the need for specialized designs to maximize efficiency in selection trials with related individuals.