Kvantitativno-genetska analiza ostvarenog vremena u trci kod kasača | TI Quantitative - genetic analysis for racing times of trotter horses

This thesis encompasses research based on quantitative genetics methods for racing times of trotter horses, which is the most important measure for the selection of racehorses. The aim of this study was to determine the most suitable model for the evaluation of genetic parameters and prediction of breeding values for trotter horses in Serbia. A total of five different models were used for this genetic analysis (univariate sire model for repeated measures, univariate animal model for repeated measures, multivariate sire model for repeated measures, multivariate animal model for repeated measures and a random regression model), by which phenotypic parameters and the effects of different fixed factors on racing times of trotter horses were initially analyzed. Results of descriptive statistics, interpreted in relation to other studies on this subject, indicate that the average value of the observed trait (speed), which in the study population was 84.13 s/km, was low compared with the possible level, and that there is scope for improvement. Fixed factors included the effects of sex, year and month of birth, year and season of the race, age, racetrack, distance and start method, and single factor ANOVA analysis showed statistically highly significant effects (P <0.001) of each these individual factors. Using a general linear model, the fixed factors were combined in different ways, and based on the coefficient of determination, the best model from a total of ten surveyed models was selected for further genetic analysis. The highest coefficient of determination (0.51) was produced by a model that included the fixed effects of sex, age and interactions: the month and year of birth, season, year of race and racetrack, and the interaction of distance and start method. By using the five different models for genetic analysis, a medium to high level of heritability was calculated, which is desirable from the aspect of selection effects because it indicates that in the study population, direct selection on the basis of race time should be effective. The results obtained in this study indicate the advantage of the univariate repeatability animal model for genetic evaluation of trotters, given the fact that this model achieved full convergence of data, as was calculated by the high heritability value (0.551) and low error of this coefficient (0.058). Regardless of the applied model, based on the average breeding values by year of birth, the estimated genetic trend was positive, and it can be concluded that over the reporting period, genetic improvement of racing times, i.e. speed, was achieved in the observed trotter horse population.