Estimation of variance components and prediction of breeding values using pooled data

Area of Science:

• Animal genetics
• Quantitative genetics
• Livestock breeding

Background:

• Phenotypic data for genetic evaluations are typically individual-based.
• Group-level performance data are common in farm animal situations.
• Integrating group data into genetic evaluations is highly desirable.

Purpose of the Study:

• To demonstrate the use of group-level data for variance component estimation and breeding value prediction.
• To outline a general procedure applicable across farm animal species.
• To compare estimates and accuracies using pooled versus individual data.

Main Methods:

• Developed a general procedure for genetic analysis using group-level data.
• Applied the procedure to Body Weight (BW) and egg production data.
• Compared heritability estimates and Estimated Breeding Values (EBVs) derived from pooled and individual observations.

Main Results:

• Heritability estimates for BW were similar using pooled and individual data at early ages (19, 27 wk).
• Differences in heritability estimates for BW emerged at later ages (43, 51 wk), potentially due to competition.
• Accuracy of EBVs from pooled data was 70-80% of that from individual data.

Conclusions:

• Estimation of variance components and breeding values using pooled data is theoretically and practically feasible.
• Group-level data can supplement individual data in genetic evaluations, albeit with a quantifiable accuracy loss.
• The developed procedure offers a viable method for enhancing genetic evaluations with commonly available group data.