Estimates of Genetic Parameters for Number Born Alive using a Random Regression Model with Considered Parities

DOI JASI Web Site Web Site 11 References

Bibliographic Information

Other Title
  • 産次変量回帰モデルを用いた生存産子数の遺伝的パラメータの推定
  • サンジ ヘンリョウ カイキ モデル オ モチイタ セイゾン サンコスウ ノ イデンテキ パラメータ ノ スイテイ

Search this article

Abstract

<p>Number born alive (NBA) is one of the most important traits to improve the sustainability of reproductive performance of sows. The aim of this study was to estimate genetic parameters for NBA using a random regression model (RRM) with different order Legendre polynomials of parities, and to compare the parameters with those from the other models [single trait model (STM), repeatability model (RPM) and multi trait model (MTM)], which are generally used for genetic evaluation of NBA. The NBA data of Landrace and Large White, which includes from the first to the sixth parity, were collected from the database on the National Swine Genetic Evaluation scheme. The number of litter records of Landrace and Large White were 55,857 and 34,024, respectively, and their pedigree records were 11,035 and 6,722, respectively. Two types of RRMs were fitted in this study: one included farm effect (RRM1), and the other included farm-parity (RRM2) as a fixed effect. Both the RRMs included additive genetic effects and permanent environmental effects as random effects, and were fitted up to the second order (i.e., cubic) Legendre polynomials of parities for the random effects. Estimated variance components across parities by RRM2 showed smooth trajectories to compare with RRM1 in both breeds. In addition, the genetic parameters estimated by RRM2 were similar to MTM than RRM1. As a result of the trajectories of the variance components and the contribution of the eigenvalues, the most appropriate order to Legendre polynomials for additive genetic effects could be up to the second order in both breeds and for permanent environmental effects could be the first order in Landrace and up to the second order in Large White. In addition, the contribution of the eigenvalues showed that the intercept explained ranged from approximately 90 to 92% of genetic variability for NBA and genetic curve with the first- and second-order terms of Legendre polynomials explained ranged from approximately 8 to 10% of genetic variability in RRM2.</p>

Journal

  • Nihon Yoton Gakkaishi

    Nihon Yoton Gakkaishi 56 (4), 139-148, 2019-12-16

    The Japanese Society of Swine Science

References(11)*help

See more

Keywords

Details 詳細情報について

Report a problem

Back to top