Genetic correlations between production and semen traits

S.H. Oh, M.T. See,

North Carolina State University, Department of Animal Science

T.E. Long, and J.M. Galvin

Smithfield Premium Genetics, Roanoke Rapids, NC

Summary

Currently, boars selected for commercial use as AI sires are evaluated on grow-finish performance and carcass characteristics. If AI sires were also evaluated and selected on semen production, it may be possible to reduce the number of boars required to service sows, thereby improving the productivity and profitability of the boar stud. The objective of this study was to estimate genetic correlations between production and semen traits in the boar: average daily gain (ADG), back fat thickness (BF) and muscle depth (MD) as production traits, and total sperm cells (TSC), total concentration (TC), volume collected (SV), number of extended doses (ND), and acceptance rate of ejaculates (AR) as semen traits.Semen collection records and performance data for 843 boars and two generations of pedigree data were provided by NPD USA. Back fat thickness and MD were measured by real-time ultrasound. Genetic parameters were estimated from five four-trait andone five-trait animal models using MTDFREML. Average heritability estimates were 0.39 for ADG, 0.32 for BF, 0.15 for MD, and repeatability estimates were 0.38 for SV, 0.37 for TSC,0.09 for TC, 0.39 for ND, and 0.16 for AR. Semen traits showed a strong negative genetic correlation with MD and positive genetic correlation with BF. Genetic correlations between semen traits and ADG were low. Therefore, current AI boar selection practices may be having a detrimental effect on semen production.

Introduction

The production of a large quantity of high quality semen is important to pork producers since most sows are artificially inseminated (Singleton, 2001). The adoption of artificial insemination (AI) has had a significant impact on the structure of the swine genetics industry. It has been reported that AI now accounts for more than 60 percent of the total swine mating in the United States (Singleton, 2001). This effectively reduces the number of boars required in the U.S. swine breeding herd and at the same time increases the importance of high fertility and genetic merit for each boar. While genetic evaluation procedures (BLUP) to select the top boars for AI are commonplace, the genetic control of semen traits has not been extensively studied. Currently, boars selected for commercial use as AI sires are evaluated on grow-finish performance and carcass characteristics. If AI sires were also evaluated and selected on semen production, it may be possible to reduce the number of boars required to service sows, thereby improving the productivity and profitability of the boar stud. In the past, male fertility traits were not analyzed due to loss of data during natural mating (Brandt and Grandjot, 1998). However, a larger data set can be obtained due to adoption of artificial insemination techniques. Correlation analysis between semen traits and grow-finish performance can be based on these data. The objective of this study was to estimate genetic correlations between production traits such as average daily gain (ADG), back fat thickness (BF) and muscle depth (MD), and boar reproductive traits such as semen volume collected (SV), total sperm cells (x 10⁹) (TSC), total concentration of sperm per mL (x 10⁶) (TC), number of extended doses (ND), and acceptance rate of ejaculates (AR).

Materials and Methods

data source

Semen collection records and performance data for 843 boars selected for artificial insemination were provided by NPD USA (Roanoke Rapids, NC). A total of 1,736 individuals were included in the pedigree file. Boars represented three breeds and were housed in two farms. Each farm was similar in numbers of boars of each breed.

Traits were ADG, BF, MD, SV, TSC, TC, ND, and AR. Backfat thickness and MD were measured longitudinally by real-time ultrasound using Aloka 500 (Corometrics; Ithaca, NY). Semen traits were recorded as repeated records. Semen volume was measured as the weight of the ejaculate volume. Total concentration was measured using a self-calibrating photometer. Total sperm cells were determined by multiplying SV and TC. Acceptance rate of ejaculates is based on the subjective evaluation of technicians and for an individual collection is binomial. Acceptance rate was calculated over the lifetime of the boar as the number of accepted collections divided by the total collections placing this data on a more normal scale. Technicians discarded ejaculates when blood or urine was present in the collection, when an evaluation of semen morphology presented a large number of abnormal sperm cells or when motility of sperm cells was low. Number of extended doses was calculated using total sperm cells divided by desired number of sperm cells per dose. For these data, each dose averaged 2.7 billion sperm with 100 ml fluid.

For these analyses the arithmetic mean of each semen trait for each individual was calculated to perform the multiple trait analyses with production traits. Therefore, our estimates for semen production traits are repeatabilities since permanent environmental effects were not separated in the model. This also resulted in averaging out the effects of collector, year-season and age of boar.

Genetic parameters were estimated from five four-trait and one five-trait animal models. Five different combinations of four multiple traits were (1) ADG, BF, MD and SV (2) ADG, BF, MD and TSC (3) ADG, BF, MD and TC (4) ADG, BF, MD and ND (5) ADG, BF, MD and AR, respectively. The five-trait analysis consisted of all semen traits, SV, TSC, TC, ND, and AR.

statistical analysis

Models for single and multiple trait evaluations were as follows:

Y_ijklm = m + A_i + F_j + T_k + B_l + e_ijklm

where, m is overall mean, A_i is the random additive genetic effect of i^th animal, F_j is the fixed effect of j^th farm, T_k is the fixed effect of k^th test batch, B_l is the fixed effect of l^th breed and e_ijklm is measurement error. Initial analyses of each trait were conducted using a single trait, animal model. The vector presentation of this model is: Y = Xb + Zu + e where, Y is the vector of observations for all traits, b is a vector of common fixed effects due to farm, test batch and breed, u is a vector of random genetic effects and e is a vector of residuals and X and Z are incidence matrices relating observations to the fixed and animal effects, and E [y¢ u¢ e¢]¢ = [b¢X¢ 0¢ 0¢]¢. Variances of the random variables were:

where  denotes a direct product operation,  and  are genetic and residual covariance matrices, with order equal to the number of traits in the analysis, and A is the numerator relationship matrix.

Variance and covariance components were estimated by a derivative-free REML algorithm (Graser et al., 1987) using the MTDFREML computer programs developed by Boldman et al. (1995). This set of programs minimizes -2 times the restricted log likelihood function (), that is, -2 = constant + log|R| + log|G| + log|C| + y¢Py where C is a full-rank coefficient matrix for the mixed model equations and y¢Py is the weighted sum of squares for the residuals, with P a projection matrix. Stopping criterion was set as 10^-10 for the simplex variance. Convergence was achieved after stopping criterion was obtained at the same or larger -2, after a minimum of two cold restarts with the parameter estimates as new starting values.

Using information acquired from univariate analyses of each trait as starting values the multi-trait models were applied to estimate the (co)variance structure. To aid convergence and complete this analysis with available computing resources, the (co)variance structure was estimated from separate four-trait and five-trait analyses. For a four-trait model

where = additive genetic variance of trait i,  is the genetic covariance between two. Estimation of the genetic and environmental correlations () from the REML (co)variance estimates is straightforward.

Results and Discussion

Table 1 presents the pooled results across all analyses. Estimates of heritability were, on average, 0.39, 0.32, and 0.15 for ADG, BF, and MD, and repeatability estimates were 0.38, 0.37, 0.09, 0.39, and 0.16 for SV, TSC, TC, ND and AR, respectively. Genetic parameters were very consistent across the four trait combinations. Heritabilities of production traits have been well documented, and the estimates found for ADG and BF in this study are similar to literature averages (McPhee et al., 1979; Lutaaya et al., 2001), but lower than that reported by Smith et al. (1962) and Mrode et al. (1993).

Table 1. Heritabilities, Repeatabilities (diagonal), genetic and phenotypic correlations below and above diagonal between production (ADG, BF, and MD) and semen (SV, TSC, TC, ND, and AR) traits⁵

Back Page