Reproductive Progress of American Yorkshire Swine
M. T. See1, J.W. Mabry2, J. Venner3, T. J. Baas2, K. J. Stalder4, and S. J. Moeller5
1North Carolina State University, Raleigh,
2Iowa State University, Ames,
3National Swine Registry, West Lafayette, IN,
4University of Tennessee, Knoxville,
5Ohio State University, Columbus, U.S.A.
The American Yorkshire Club has been progressive in implementing and utilizing across-herd genetic evaluations for improving both maternal and terminal characteristics. Maternal selections are based upon the reproductive traits of number of pigs born alive (NBA), litter birth weight, number of pigs weaned (NWN) and litter weight at 21 d of age (LWT). Results from the most recent national genetic evaluation that included data from 1985 to 1999 were used to determine genetic trends in the breed.
Materials and Methods
Breeding values (BV) for NBA, NWN and LWT were estimated using an animal model that includes fixed effect of contemporary group (CG), random direct genetic effect of the animal and uncorrelated random effects of sow's permanent maternal environment and mate within CG. Contemporary groups were defined by the producer as sows that were bred, gestated, farrowed and weaned together. Data were pre-adjusted for known sources of variation as described in (1). Components of (co)variance used for BV estimation were specific for the American Yorkshire breed as described in (2). Breeding values were also estimated for the post-weaning traits of days to 114 kg (Days), backfat depth (BF), loin muscle area (LMA), and kg of lean in an 84 kg carcass (Lean) using a model that includes fixed effects of CG and sex within herd, the random genetic effect of pig, and uncorrelated random effect of the pig’s birth litter. All post-weaning data were adjusted to 114 kg as described in (2). Traits were combined into three selection indexes: Sow Productivity Index (SPI) including only reproductive BV, Maternal Line Index (MLI) including all BV, and Terminal Sire Index (TSI) including only post-weaning BV. Maternal BVs were estimated for 115,647 pigs and post-weaning BVs were estimated for 357,349 pigs. As there was little change in BVs prior to 1988 when substantial usage and availability of BVs occurred, genetic trends were estimated by regressing mean BV on year of birth using a spline model (3) with a knot at 1988. The spline model results in separate genetic trends for animals born prior to and after 1988.
Results and Discussion
Since 1988, the average BV for NBA, NWN and LWT has improved .36 pigs, .10 pigs and 3.6 kg, respectively. The SPI and MLI have increased 5 and 17 index points, respectively. Regression coefficients indicating annual genetic improvement for NBA, NWN, LWT, SPI and MLI between 1988 and 1998 were .036 pigs, .01 pigs, .38 kg, .52 index points, and 1.73 index points, respectively. Prior to 1988 annual genetic improvement for NBA, SPI, and MLI was .02 pigs, .17 index points, and -.32 index points, respectively. Regressions for NWN and LWT prior to 1988 were not significantly different from zero. Regression coefficients for the post-weaning traits of BF, LMA, Days, and Lean were -.06 cm, .43 cm2, -.45 days, and .21 kg, respectively. These results would indicate that genetic improvement has been achieved simultaneously for both reproductive and post-weaning traits. American Yorkshire breeders have made significant genetic improvement for economically important reproductive traits by implementing effective across-herd selection programs. Genetic improvement of reproductive traits should help improve the overall efficiency and profitability of pork production.
(1) Culbertson, M. S., J. W. Mabry, J. K. Bertrand and A. H. Nelson. 1998. J. Anim. Sci. 75:2362-2367.
(2) Mabry, J.W. 1998. Seedstock Edge 5:45-49 (no.4). West Lafayette, IN U.S.A.
(3) Freund, R. J., and R. C. Littell. (1991) SAS System for Regression 2nd Edn, SAS Institute, NC.
Previously published in the Proceedings of the 14th International Congress on Animal Reproduction (ICAR), Vol 1:P 304. July, 2000. Stockholm, Sweden