SELECCIÓN DE MODELOS Y ESTIMACIÓN DE PARÁMETROS GENÉTICOS PARA MEDIDAS ZOOMÉTRICAS DE CABALLOS PURA RAZA ESPAÑOLA EN MÉXICO.

The prototype of Pura Raza Española horse depends on the equestrian disciplines and establishes the breeding selection objectives; the selection criteria are developed from zoometric measures (ZM), which are subject to multiple direct and maternal genetic effects. The objective of this study was to evaluate the magnitude of the maternal genetic effects, based on the (co)variances, if they are equal to or different from zero. Our database included information of 1806 equines born from 2004 to 2013, with data of eight ZM (mean in cm ± SD): thoracic perimeter (TP; 189.3±8.9), knee perimeter (KP; 32.3±1.8), cannon perimeter (CP; 20.4±1.3), scapular-ischial length (SIL; 160.7±5.9), chest width (CW; 43.1±3.0), height of the substernal space (HSS; 83.1±5.1), chest height (CH; 73.2±4.3), and withers height (WH; 157.8±4.3). We conducted the analysis with a mixed model that included the fixed effects of contemporary groups (ranch, year, time, and sex) and the age covariable of the individual; as random, the direct (D) and maternal (M) genetic effects, plus the covariance between them, (C). By alternating maternal genetic effects, we compared the models: D, DM, and DMC. We selected the models using the likelihood-ratio test. The pedigree was made up of 7498 animals. We estimated the direct (h²) and maternal (m²) heritabilities, as well as the genetic correlations (rg). For the analysis, we used the software MTDFREML. The maternal genetic effects in CH and CP were statistically not different from zero (p±0.05); in the other ZM, the DMC model was the most fit. The h² oscillated between 0.09 (CH) and 0.65 (WH), with an average of 0.40. The m² fluctuated between 0.17 (HSS) and 0.23 (WH), with an average of 0.19. All the rg were negative, between ±0.37 (WH) and ±0.99 (HSS), which must be considered in the genetic improvement of the ZM. [ABSTRACT FROM AUTHOR]