Genetic parameter estimates for racing time by restricted maximum likelihood in the Thoroughbred horse of Japan

Abstract

Summary Genetic parameters were estimated for repeated racing times on Thoroughbred horses for several racing distances both on turf and dirt. An additive genetic and permanent environmental model of multiple‐trait derivative‐free restricted maximum likelihood (MTDFREML) was used. The data used were collected by the Japan Racing Association from 1986 to 1990 and the effects of individual race, sex, age, jockey and weight carried were used. The generation 2 pedigree information was preferable for variance estimates. Heritability decreases as the racing distance increases. These values were 0.25, 0.16, 0.10, 0.12, 0.09 and 0.08 at 1000, 1200, 1400, 1600, 1800, and 2000 m, respectively, on turf, and 0.19, 0.22, 0.12, 0.09 and 0.17 at 1000, 1200, 1400, 1600, and 1800 m, respectively, on dirt track. The values of repeatability were from 0.43 to 0.70 on turf, and from 0.51 to 0.67 on dirt track. These suggest that racing times at the different racing distances may be regarded as different traits when the horse is to be evaluated genetically. Zusammenfassung Schätzung genetischer Parameter für Rennzeit von japanischen Vollblütern durcb eingeschrïhte max‐imale Wahrscheinlichkeit Die Schätzung der genetischen Parameter erfolgte für wiederholte Rennzeiten von Vollblütern über mehrere Renndistanzen sowohl auf Rasen als auch auf Sandbahn. Zusätzlich wurde ein Permanent‐Umweltmodell (multiple‐trait derivative‐free restricted maximum likelihood (MTDFREML)) angewendet. Die Daten wurden von 1986 bis 1990 von JRA erhoben und die Wirkungen von Rasse, Geschlecht, Alter, Jockey und Tragegewicht geschätzt. Stammbaumdaten der Generation 2 dienten bevorzugt zur Varianzschätzung. Der Einfluß des Erbguts nimmt mit zunehmender Renndistanz ab, die betraffender h2 Werte waren 0,25, 0,16, 0,10, 0,12, 0,09 und 0,08 bei 1000, 1200, 1400, 1600, 1800 und 2000 m auf Rasen sowie 0,19, 0,22, 0,12, 0,09 und 0,17 bei 1000, 1200, 1400, 1600 und 1800 m auf Sandbahn. Der Wiederholbarkeitskoeffizient lag zwischen 0,43 und 0,70 auf Rasen und 0,51 bis 0,67 auf Sandbahn. Dies legt die Vermutung nahe, daß bei Zuchtwertschätzung Rennzeiten über verschiedene Distanzen als unterschiedliche Merkmale betrachtet werden sollen. Résumé Estimation des paramètres génétiques sur le temps de course chez les Pur‐sang au Japon, selon la méthode de probabilité maximum restreinte On a estimé les paramètres génétiques pour les temps de course répétés chez les chevaux Pur‐sang pour plusieurs distances de course sur gazon et sur terre battue. On a utilisé un modèle additif génétique et environnementiel permanent de vraisemblence maximum restreinte à traits multiples et sans dérivé (MTDFREML: ‘multiple‐trait derivative‐free restricted maximum likelihood’). Les données utilisées ont été rassemblées par la JRA de 1986 à 1990 et on a utilisé les effets de la race, du sexe, de l'âge, du jockey et de la charge portée. L'information de pédigrée de génération 2 s'est avérée préférable pour les estimations de variance. L'effet de l'hérédité diminue avec l'augmentation de la distance pa...

Phenotypic and Genetic Parameter Estimates for Racing Traits of Thoroughbred Horses in Turkey

“…Estimates of heritability obtained from the analysis of dirt track dataset were clearly higher than those of turf track for racing time and best racing time, whereas they were similar for rank. In contrast with the current study, estimates of heritability for racing time on dirt and turf tracks were reported to be similar by OKI et al (1995). Estimates of heritability for earning traits varied from low to medium in literature.…”

“…HINTZ (1980) reported lower estimate (0.15) for Thoroughbred horses and similar estimate (0.32) for Trotters compared with the current study. The estimate of heritability for racing time on dirt track (0.396) in the current study was higher than estimates of OKI et al (1995) for dirt track, which varied between 0.086 and 0.217 depending on racing distance. On the other hand, estimate of heritability for racing time on turf track (0.231) was within the range of estimates reported by OKI et al (1995) for racing time on turf track (0.081-0.254).…”

“…The estimate of heritability for racing time on dirt track (0.396) in the current study was higher than estimates of OKI et al (1995) for dirt track, which varied between 0.086 and 0.217 depending on racing distance. On the other hand, estimate of heritability for racing time on turf track (0.231) was within the range of estimates reported by OKI et al (1995) for racing time on turf track (0.081-0.254). Estimate of heritability for best racing time (0.467) was also higher than the literature estimates, except reports of OJALA et al (1987), which varied from 0.28 to 0.50 depending on model for Finnish horses.…”

Genetic correlations between racing performance at different racing distances in Thoroughbreds and Arab horses

“…The effect of environmental factors connected with specific race such as number and level of competitors, track and weather conditions may be higher for long distance races than for short distance races. Oki et al (1995) suggested that the racing performance at the different distances might be regarded as different traits. Also, a trait measured in two different environments is to be regarded as two traits (Falconer, 1989).…”

“…However, the horse's racing performance is not always best at one racing distance. Since performance is unlikely to be constant over a distance (some horses are sprinters, some stayers, and the ability to win a 1 000m race does not imply the ability to win a 2 000m race), some authors (Oki et al, 1995;Williamson and Beilharz, 1996;Oki et al, 1997) analysed the racing performance separately in each distance range. The effect of environmental factors connected with specific race such as number and level of competitors, track and weather conditions may be higher for long distance races than for short distance races.…”