No document available.
Abstract :
[en] The data used in the present study were recorded at the Algerian Horse Racing Society for 36492 racing performance of 913 Arabian horses from 1995 to 2007. The aim of the study was to identify the genetic parameters underlying three traits: two earnings traits, namely the logarithm of average annual virtual earnings (LAEV) and the logarithm of average annual virtual earnings per start (LAEV/S), and a ranking trait, the normalized ranking (PERF). To identify the fixed effects to be included in the genetic mixed model, a preliminary analysis was conducted using the General Linear Models (GLM) procedure from SAS software. The effects of age, sex, year and the interaction between year of the race and age and between sex and age were included in the model for the three traits. Afterwards, two random effects, a direct genetic effect of the animal and a permanent environmental effect were included in the mixed model. The variance components and genetic parameters were estimated using the restricted maximum likelihood (REML) procedure with the MTDFREML program. The analyses with this repeatable animal model led to the following estimation of the genetic parameters: for LAEV, heritability was 0.225 (±0.041), while estimate of repeatability was 0.330 (±0.040). For LAEV/S, heritability was 0.164 (±0.027), while estimate of repeatability was 0.215 (±0.022). The heritability for the normalized ranking was higher, 0.369 (±0.054), indicating that this trait might provide faster progress for breeding programs of Arabian horses in Algeria. The repeatability estimate for the normalized ranking was 0.587 (±0.045). The genetic correlation between LAEV and LAEV/S was 0.99, revealing a almost complete genetic dependence between these two traits, 0.69 between PERF and LAEV and 0.79 between PERF and LAEV/S.
