Genetic Trends for Reproductive Traits in Hampshire Swine

S. J. Moeller¹, J. W. Mabry²,T. J. Baas³, K. J. Stalder⁴ and M. T. See⁵

¹The Ohio State University, Columbus; ²University of Georgia, Athens; ³Iowa State University, Ames; ⁴The University of Tennessee, Jackson; ⁵North Carolina State University, Raleigh

Introduction

Reproductive efficiency plays a significant role in the economic viability of a pork production system. Genetic improvement in reproductive efficiency has, in the past, been relatively slow due to the low heritability estimates for these traits. Improvements in computational speed and the incorporation of enhanced statistical models have vastly improved the ability to measure genetic variation and track genetic change for reproductive traits due to improved estimation capabilities and higher accuracy of breeding value estimates.

Materials and Methods

Genetic trends for purebred Hampshire swine were calculated using data submitted by Hampshire breeders over an 11-year period (1988 to 1998). Breeding values (BV) were computed using PEST genetic evaluation software. The model for estimating BV’s for reproductive traits (number born alive (NBA), number weaned (NW) and 21-day litter weight (LWT)) included the fixed effect of contemporary group and random genetic effect of the animal, with permanent environmental and mate within contemporary group as uncorrelated random effects. Contemporary groups were females bred, gestated, farrowed, and weaned together. Data were pre-adjusted for known sources of variation as described in (1). NBA was adjusted for parity of the female and age at farrowing. LWT was adjusted for parity of the female, age at farrowing, number allowed to nurse, and age at weaning. Two, multi-trait bio-economic indexes were computed including: 1) Sow Productivity Index (SPI) combining NBA, NW and LWT and 2) Maternal Line Index combining NBA, NW, LWT, Days to 114 kg (DAYS), and Backfat adjusted to 114 kg (BF). Breeding values for DAYS and BF were computed in a model including the fixed effects of contemporary group and sex within herd, the random genetic effect of the pig, and an uncorrelated random effect of the litter a pig was born in. Contemporary groups for DAYS and BF were pigs farrowed, housed and fed together. Components of (co)variance used for BV estimation were specific to the Hampshire breed as described in (2). Genetic trends were calculated as the change in average breeding value or index unit per year across the eleven years.

Results and Discussion

Average breeding value estimates for the Hampshire population were very small and genetic trends were nearly zero for NBA, LWT, NW, and SPI. Calculated progress over 11 years for NBA, NW, and LWT were 0.0039 pigs/year, 0.0007 pigs/year, and 0.0198 kg/year, respectively. The SPI index showed no genetic improvement over the period evaluated (0.04 index units/year). Small gains in the MLI index (.47 index units/year), which combines breeding values for maternal and performance traits were observed. The gain in MLI may have been a function of intense selection for reduced BF (-.28 mm/year) in the Hampshire breed. The focus on BF by the Hampshire breed and the antagonistic correlation between reproductive and carcass traits may be the reason for no improvement in BV’s for reproductive traits in the Hampshire population.

References

(4) (1)Culbertson, M. S., J. W. Mabry, J. K. Bertrand, & A. H. Nelson. 1998. Breed specific adjustment factors for reproductive traits in Duroc, Hampshire, Landrace, and Yorkshire swine. J. Anim. Sci. 75:2362:2367.

(4) (2) The new STAGES program. How will the numbers change. Seedstock Edge 5:45-59 (No. 4).

Previously published in the Proceedings of the 14^th International Congress on Animal Reproduction (ICAR), Vol 1:P 304. July, 2000. Stockholm, Sweden.