Estimation of genetic parameters for racing speed at different distances in young and adult Spanish Trotter horses using the random regression model.

This study examined the genetic parameters related to race speed at various distances in young and adult Spanish Trotter horses using a model known as Random Regression Model (RRM). The research found that genetic factors do play a role in the race speed of these horses, but this contribution varies with distance and age, suggesting a need for specific breeding value estimations for different racing characteristics.

Research Design and Data Collection

• The study was based on a large dataset, consisting of 71,522 records, representing 3154 Spanish Trotter horses.
• These records included times per kilometer for individual races between the years 1991 to 2007.
• The variable “times per kilometer” for various age groups (young and adult) and different distances (1600-2700 meters) were treated as distinct traits.

Analysis Model and Parameters

• A bi-character random regression model was applied to the data.
• Fixed parameters considered included the combination of the hippodrome and race date, gender of the horses, type of race start, and a fixed regression of Legendre polynomials.
• Random parameters considered involved the random regression Legendre polynomial for the animals, individual environment factors, and the driver. The number of animals in the pedigree was 9201 while the number of animals with data was 3154. The drivers were 957 in number.
• In terms of error variance, it was treated as heterogeneous with two class categories: ages.

Results and Implication

• The estimated heritability by distance showed variance in range from 0.12 to 0.34, having a different trajectory for two age groups.
• Within each age group, the genetic correlations for adjacent distances were high (>0.90), but it decreased when the differences between them exceeded 400 meters.
• The genetic correlations for the same distance across age groups ranged from 0.47 to 0.78, thus TPK (time per kilometer) can be construed as genetically related.
• However, the trait presents differently across a continuum of distance and age, implying potential re-ranking in the breeding value of the horses subjected to different racing characteristics.
• The researchers recommend the use of the Random Regression Model (RRM) for estimation as it allows estimation of breeding value across the entire racing competition trajectory.