J. Anim. Sci. 2005. 83:2319-2329
© 2005 American Society of Animal Science
Genetic parameters for a maternal breeding goal in beef production1
T. Roughsedge*,2,
P. R. Amer
,
R. Thompson
and
G. Simm*
* Scottish Agricultural College, Edinburgh EH9 3JG, United Kingdom;
and
Abacus Biotech Limited, Dunedin, New Zealand; and
and
Rothamsted Experimental Station, Institute of Arable Crop Research, Harpenden, Hertfordshire AL5 2JQ and Roslin Institute (Edinburgh), Roslin, Midlothian EH25 9PS, United Kingdom
 |
Abstract
|
|---|
New maternal breeding values have been developed for use in UK beef evaluations. To undertake multitrait BLUP evaluations, it is necessary to have a full covariance matrix. This study outlines the approach taken to construct the full covariance matrices for the four beef breeds that most widely contribute to suckler beef cows in the United Kingdom. The maternal traits investigated were age at first calving, calving interval, lifespan, mature cow weight, 200-d weight, and calving difficulty. Three terminal sire traits (weight at 400 d, ultrasonic fat depth, and muscle score) were included to estimate covariances between the new and existing traits. A sire-maternal-grandsire model was used for the estimation procedure in a series of bivariate and multivariate models. A weighted bending procedure was employed to construct positive definite covariance matrices. Parameter estimates broadly agreed with literature values, although for some traits, literature information was very scarce. Some differences between parameters for different breeds were evident.
Key Words: Genetic Parameter • Maternal Trait • Weighted Bending
|
Introduction
|
|---|
To explore alternative breeding goals and conduct routine genetic evaluations in livestock, it is essential to have good estimates of genetic parameters. Until now UK beef breeding has concentrated on terminal sire traits, and the available selection tools have been directed toward achieving optimal carcass merit at least cost, while considering ease of calving as a trait of the sire (Amer et al., 1998
). This was primarily due to a large proportion of cows in the beef suckler herd being derived from the dairy herd as beef x dairy heifers; however, over recent years, there has been a shift in the dairy industry from the use of the British Friesian cow, which retained some beef characteristics, to the specialist milk-producing Holstein cow (Roughsedge et al., 1999). Due to the less desirable carcass characteristics of the Holstein breed, attention has begun to focus on using beef breeds or crosses as replacement heifers.
A herd-level bioeconomic model (Roughsedge et al., 2003a) has identified several key beef profitability drivers, including reproductive rate, age at first calving, and cow size (Roughsedge et al., 2003b). Following the choice of the most appropriate breed or breeds to be used as beef replacement heifers, selection of parents of high genetic merit for the breeding goal traits needs to be practiced. Maternal breeding tools have been developed to facilitate this selection. These tools comprise a series of new EBV and two new sub-indexes that combine into an overall index for selecting sires to produce replacement heifers (Roughsedge et al., 2005).
When developing genetic evaluation procedures for new EBV, it is important to understand the genetic relationships between the new and existing traits. The aim of this study was to outline the traits being recorded for the maternal breeding goal and to estimate the genetic parameters for the traits that will be used in the multitrait BLUP of breeding values.
|
Materials and Methods
|
|---|
Data Description
Four breeds were chosen for the investigation of traits contributing to maternal performance. The breeds were the Aberdeen Angus, Limousin, Simmental, and South Devon. They were chosen because they represent the most numerous beef breeds contributing to the genetic makeup of commercial suckler cows in the United Kingdom. These breeds also represent both native British breeds (Aberdeen Angus and South Devon) and mainland European breeds (Limousin and Simmental). Cow weights were measured on a sample of the populations of each of these breeds in the United Kingdom. Data were collected on a sample of approximately 1,000 to 1,500 cows located within 30 spring-calving herds over a 3-yr period for each breed. Measurements were taken during the winter; approximately 2,500 to 3,500 records per breed were generated. In addition, calving information and data on calf weights, fatness, and muscularity, which were routinely measured for the breeds in this study, were extracted from the Meat and Livestock Commission BeefBreeder recording service database. For the calf information, a sample of these data were taken for calves born since 1978 that had a calving difficulty (CD) score, were managed in a contemporary group of greater than nine, and were from sires having at least 24 offspring with records. These restrictions resulted in a sample of approximately 10% of the total calf records for each breed. The calving dates used to investigate fertility and survival were restricted to contemporary group sizes of five or greater. The number of records available for parameter estimation for each trait is shown in Table 1.
Trait Descriptions
Weight Traits.
Three weight traits were used in the analysis, 200-d weight (W200), 400-d weight (W400), and mature weight (MW). Crump et al. (1997b) provided a description of the procedure for calculating W200 and W400. Animals have to have weights recorded between 170 and 300 d for W200 and between 300 and 500 d to qualify for W400. The procedure uses within-animal regression to calculate the weights using all available records. For MW, cows of 
3 yr of age were weighed during the start of the winter in November and December and were scored for condition at weighing to allow correction of the data for condition.
Reproductive and Functional Traits.
Four traits were considered in this category. They were CD, age at first calving (AF), calving interval (CI), and lifespan (LS).
Calving difficulty has been included in the U.K. beef BLUP evaluations since 1997. The trait is recorded as a score of 1 to 5 (1 = no assistance, 2 = easy pull, 3 = hard pull, 4 = veterinary assistance, and 5 = caesarian section).
Age at first calving was computed by first determining the number of heifers calving at three ages (2, 2.5, and 3 yr of age). Heifers were classified as calving at 2 yr if they first calved between 1.5 and 2.25 yr of age, at 2.5 yr if they first calved between 2.25 and 2.75 yr of age, and at 3 yr if they first calved between 2.75 and 3.5 yr of age. For each heifer calving, the total number of other heifer calvings in the same herd-year and the number of each age group of calving were determined. This information was then used to assign a binary score to each breeding female based on the following rules. A calving at an earlier age, given the potential to calve at a later age, was scored as 0, and a calving later, given the opportunity to calve earlier, was scored as 1. If there had been no opportunity for a heifer to calve at more than one age, the record was assigned as missing. This approach meant that heifers on farms that have two calving seasons rather than one within a year were not at an unfair age advantage, which could be the case if the actual AF were taken as the measure of the trait.
Calving interval was used as a predictor of cow fertility. The interval between first and second calvings was used in the analysis. Calving intervals not between 290 and 630 d were set as missing records, following Gutierrez et al. (2002).
For LS, the database was searched, following a set of rules developed for dairy cows (Brotherstone et al., 1997; Lubbers et al., 2000), for up to five parities of appearances, and LS was assigned as the parity the cow attained or was predicted if data were censored. The data could be censored either because the cow survived beyond parity five or because there was not sufficient time for the cow to have completed five parities. Calvings were assigned a number from 1 to n. If the actual final calving was known (i.e., cow calved at parity n but did not calve at parity n + 1, where n is less than five), then the LS was n. For cows that had time for calving n but not for calving n + 1 or for cows that reached parity five, the LS figure was assigned to reflect the parity that was expected to be reached using average survival probabilities from parity to parity in the population. In this way, all cows in the population that have a valid first calving recorded were assigned a LS figure.
The average survival of the four breeds (Aberdeen Angus, Limousin, Simmental, and South Devon) was calculated using information on the first five calvings of cows that first calved between November 1984 and October 1985, applying the rules in Table 2. By taking the average across-parity survival probability value from parity to parity (Table 3) the censored animals can be assigned a LS following Brotherstone et al. (1997) by
where n is the known number of parities completed, p is the average across-parity probability of survival from one parity to the next, and i is the number of extra parities survived.
The assigned LS were 1 if there was no recorded calving after first parity within the given time, n if the cow calved at parity n but not n + 1 within the given time, and n + p/(1 – p) if there was a calving at parity n and not time for the cow to calve at parity n + 1.
Carcass Traits.
Two traits were included as indicators of carcass quality, fat depth (FD), an ultrasonic measure of subcutaneous back fat at the 12th rib, and muscle score (MS), a visual appraisal of muscularity scored on a 1- to 15-point scale. Both of these traits are evaluated at approximately 400 d of age, at which age pedigree recorded bulls are at a similar weight to the slaughter weight in commercial production.
Data Analyses
The genetic and phenotypic variance and covariance parameters for all four breeds were estimated by residual maximum likelihood using ASREML (Gilmour et al., 2002). Because of computational limitations, a series of bivariate and multivariate linear sire-maternal-grandsire models were fitted. In addition to computational constraints, it is recognized that analyses of more than three or four traits can produce erratic estimates of parameters, in particular where the estimation procedure converges at the boundary of the parameter space (Hill and Thompson, 1978). The linear model approach was taken to analyze CD, which has been demonstrated to be effective compared with use of a threshold model in situations where small contemporary group size and small sire group size exist (Phocas and Laloë, 2003), a situation that applies in the UK pedigree sector (Crump et al., 1997a).
The significant fixed effects were determined using ASREML and are summarized in Table 4. Fixed effects included stage of pregnancy for cows with categories of "calved and not gestating," "gestating but 50 d or more before term," and "50 d or less before calving." For all traits, sire was fitted as a random effect. For W200 and CD models, maternal grandsire was fitted as a random effect, and covariance between sire and maternal grandsire was fitted. Dam permanent environment was fitted as a random effect to models of calf traits (W400, W200, and CD). Animal permanent environment was fitted for the cow MW model. The cow reproductive and functional traits AF, RS, and LS were analyzed using a trivariate model. The calf traits W400, W200, and CD were analyzed using a trivariate model. All other combinations for the rest of the traits were analyzed as bivariate models. The data sets used for these analyses were subsets of the previously described data set and were selected to maximize the number of sires with offspring with records of both traits. In the case of traits with maternal components, the number of maternal grandsires in common was also considered.
The post-processing capability of ASREML was used to partition the sire and maternal grandsire (co)variance components into animal and maternal (co)variance components following Willham (1963). The estimates were combined into full (co)variance matrices for each breed, taking average within-breed parameter estimates, weighted by the inverse estimate variance, where multiple estimates existed between data sub-samples. To make the resulting covariance matrices positive definite, a bending procedure was required. A great deal of variation existed in the amount of information available for estimating the different (co)variances, and to take account of this, an iterative bending procedure was followed (Jorjani et al., 2003), which allowed more bend to those parameters with higher SE. The squared SE of the heritability and correlation estimates were used as weighting factors. Where a covariance estimate was not obtained because of convergence problems, a parameter of zero with a SE of 1 was used in the matrix before bending.
|
Results
|
|---|
Raw Trait Means
The raw trait means and SD (corrected for fixed effects) are presented in Table 1
. These breeds were not recorded in the same environment, and as such, the raw trait means do not provide an accurate measure of between-breed differences.
Variance Components
Tables 5 to 8 show the variance components following transformation of the estimated components from the sire and sire-maternal-grandsire models, along with their SE, for the four breeds. These tables are referred to in the following sections.
View this table:
[in this window]
[in a new window]
|
Table 5. Aberdeen Angus estimates of phenotypic ( 2P) and error (2E) variance and genetic components (h2 = direct heritability, m2 = maternal heritability, c2 = maternal permanent environment, p2 = direct permanent environment; all x 100) together with SE (x 100)
|
|
View this table:
[in this window]
[in a new window]
|
Table 6. South Devon estimates of phenotypic (2P) and error (2E) variance and genetic components (h2 = direct heritability, m2 = maternal heritability, c2 = maternal permanent environment, p2 = direct permanent environment; all x 100) together with SE (x 100)
|
|
View this table:
[in this window]
[in a new window]
|
Table 7. Limousin estimates of phenotypic (2P) and error (2E) variance and genetic components (h2 = direct heritability, m2 = maternal heritability, c2 = maternal permanent environment, p2 = direct permanent environment; all x 100) together with SE (x 100)
|
|
View this table:
[in this window]
[in a new window]
|
Table 8. Simmental estimates of phenotypic (2P) and error (2E) variance and genetic components (h2 = direct heritability, m2 = maternal heritability, c2 = maternal permanent environment, p2 = direct permanent environment; all x 100) together with SE (x 100)
|
|
Live Weights.
Cow weight (MW) heritability was estimated to be approximately 0.4 in the two continental breeds, Limousin and Simmental, and approximately 0.2 in the two British breeds, Aberdeen Angus and South Devon. The repeatability was quite similar across breeds, ranging from 0.71 in Simmental to 0.79 in Limousin. The estimates for the British breeds were lower than previously reported MW heritabilities, but the repeatability estimates were very similar to those found previously (Arango et al., 2002; Meyer et al., 2004; Nephawe et al., 2004). Estimates of direct heritability for the other weight traits, W200 direct (W200d) and W400, were again largely consistent across breeds. The maternal W200 (W200m) component of variance at 0.02 to 0.03 was less than the 0.11 estimate of Meyer et al. (2004).
Reproductive and Functional Traits.
The heritability estimate for CI was low at around 0.04 for Limousin, but ranged from 0.09 to 0.13 for the other three breeds. Many other studies have found similarly low heritabilities for fertility traits in cattle. Age at first calving was moderately heritable in three of the breeds with estimates of 0.17, 0.26, and 0.22 for Simmental, Limousin, and Aberdeen Angus, respectively. A lower value of 0.05 was found for the South Devon. Heritability of LS ranged from 0.03 to 0.13, which agrees with estimates reported in dairy cattle (Brotherstone et al., 1997; Lubbers et al., 2000). Direct heritabilities of CD (CDd) varied widely among breeds from 0.13 for Limousin to 0.35 for Simmental, with Aberdeen Angus and South Devon having estimates of 0.26 and 0.19, respectively. These values are consistent with the estimates for Aberdeen Angus, Limousin, and Simmental of Bennett and Gregory (2001) and are within the range of estimates of Trus and Wilton (1988). Less variation was observed in the maternal heritability of CD (CDm), with estimates from 0.07 to 0.11. The estimates of maternal heritability were consistently lower than those of Bennett and Gregory (2001).
Carcass Traits.
Fat depth and MS had moderate heritability estimates for all breeds. South Devon had the highest values for both traits (0.48 and 0.31 for FD and MS, respectively). The lowest estimate for FD heritability was observed in Aberdeen Angus at 0.28, and the lowest estimate for MS was noted in Simmental at 0.24. The phenotypic variation of MS was greatest in the two British breeds, Aberdeen Angus and South Devon.
Correlation Estimates.
Tables 9 to 12 show the correlation matrices derived following the weighted bending procedure for the four breeds. These tables are referred to in the following sections. A number of correlations could not be estimated because of a combination of insufficient data and convergence problems. These are indicated in Tables 9 to 12 as parameters having SE = 1. The bending process resulted in nonzero reported values for those parameters. Following the bending process, < 9% of the parameters were changed >1 SE unit from their original estimate, and where this was the case, the new parameter allocated was not significantly different from zero.
View this table:
[in this window]
[in a new window]
|
Table 9. Genetic correlation matrix following weighted bending procedure for Aberdeen Angus (heritabilities on diagonal)
|
|
View this table:
[in this window]
[in a new window]
|
Table 10. Genetic correlation matrix following weighted bending procedure for South Devon (heritabilities on diagonal)
|
|
View this table:
[in this window]
[in a new window]
|
Table 11. Genetic correlation matrix following weighted bending procedure for Limousin (heritabilities on diagonal)
|
|
View this table:
[in this window]
[in a new window]
|
Table 12. Genetic correlation matrix following weighted bending procedure for Simmental (heritabilities on diagonal)
|
|
Genetic Correlation Matrix Following Weighted Bending.
A high correlation was observed among the weight traits, MW, W400, and W200, in all breeds. In the Limousin breed, the highest correlations were between adjacent weights. In Aberdeen Angus, correlations were consistent in size across all three weight traits; however, in South Devon and Simmental, higher correlations were estimated between MW and W200 than between MW and W400. The estimate of the genetic correlation between W200 and MW across all breeds is similar to the value of 0.85 between weaning weight and MW found by Kaps et al. (1999) in Aberdeen Angus. In general, the highest correlation estimates are typically observed between adjacent weights (Koots et al., 1994; Meyer et al., 2004); however, Northcutt and Wilson (1993) found a higher genetic correlation between MW and weaning weight (0.62) than between MW and yearling weight (0.45).
Weight traits showed positive and negative correlations with FD. Mature weight was positively correlated with FD in Limousin and Simmental (0.43 and 0.28), but was negatively correlated with FD in Aberdeen Angus and South Devon (–0.16 and –0.37). All breeds except Aberdeen Angus showed a positive correlation between W400 and FD, ranging from 0.28 for Limousin to 0.46 for Simmental. The Aberdeen Angus had a negative correlation of –0.2 between W400 and FD. Literature estimates of correlations between weight traits and carcass fatness also are variable. Meyer et al. (2004) observed negative genetic correlations between MW and several measurements of carcass fat. A genetic correlation of 0.25 between rib FD and slaughter weight was reported from an analysis of a pooled multibreed sample of castrate males (Gregory et al., 1995a), and an average correlation across many studies of 0.32 between yearling weight and back fat adjusted to a constant age was reported in the parameter review of Koots et al. (1994).
Of the reproductive traits, AF, where a higher value indicates later AF, was negatively correlated with MW, except in the Simmental breed. All breeds except South Devon had a very modest negative correlation between W400 and AF, ranging from –0.06 in South Devon to –0.18 in Limousin, suggesting that selecting for higher W400 results in lower AF. This result is in line with a correlation of –0.11 between yearling weight and age at puberty reported by Gregory et al. (1995b). Lifespan was also negatively correlated with MW, with the exception of Aberdeen Angus, where it was 0.50. In Aberdeen Angus, the correlation between LS and W400 also was moderate and positive (0.32), a trend again not seen in the other breeds, where it was low or negative. Calving interval was negatively correlated with CDd in Aberdeen Angus and Limousin; Simmental and South Devon breeds had zero and low positive correlations with CDd. The reverse was observed in CDm. Aberdeen Angus and Limousin breeds both had moderate positive genetic correlations between CI and CDm. Direct CD was positively correlated with W200d and W400 in all of the breeds, except Aberdeen Angus, where a sign reversal was noted. In a previous study, Bennett and Gregory (2001) found the same magnitude of correlation in a weighted average correlation from estimates across a number of breeds.
|
Discussion
|
|---|
The results of this study help to fill a void in the information required to implement multitrait BLUP genetic evaluations incorporating maternal traits of high economic importance to beef production systems. There are, however, a number of gaps in the correlation matrices presented in this study, where insufficient data were available to estimate the correlation. These gaps are indicted in Tables 9 to 12 by SE = 1.
Fertility was analyzed using CI as a correlated trait. It was argued by MacGregor and Casey (1999) following Bourdon and Brinks (1983) that calving date is a more suitable measure of female fertility than CI unless previous calving date is used to correct CI. Those researchers suggest that shorter CI are the result of late calving dates in the previous year and, hence, when the CI from first to second calving is used, there is selection for cows that calved late as heifers. In this study, it was found that a negative genetic correlation existed between AF and CI, indicating that older first-calvers had shorter CI; however, the data in this study were from farms with spring and autumn calving, many with both. Approximately 60% of calvings were recorded in the first 6 mo (January to June) of the year, and 40% were recorded in the last 6 mo. Nonetheless, many calvings fell in the period between peak spring and peak autumn calving, suggesting that restricted mating practices are not strictly adhered to in the United Kingdom at present. It is not possible to distinguish late spring-calving cows from early autumn-calving cows; hence, assigning date of calving becomes problematic. The relatively small UK herd size also makes the approach of Rege and Famula (1993) to score calving date within herd-year-season of calving untenable. Until records of joining periods are recorded on UK farms, the CI trait within multitrait BLUP provides us with the only useful selection tool for fertility.
The LS trait investigated in this study was adapted from dairy breeding, and the trait was shown to be heritable in beef populations. Lubbers et al. (2000) made a comparison between survival analysis using the proportional hazard model and the linear model approach to the LS trait applied in this study using dairy cow data. It was found that breeding values resulting from the two approaches were highly correlated (0.93 to 0.97 across model variations). The drawback of the approach, as highlighted by Brotherstone et al. (1997), is that proportionally more emphasis is placed on young cows in the analysis; therefore, it is recommended that only cows with two or more calvings be included in the analysis. The LS approach used in this study also uses more information than the binary approach to productive life such as that used in stayability studies (Snelling et al., 1995), where a binary score is allocated to a cow reaching a specific target number of calvings by a given age and the number of calvings achieved by those not reaching the target is not differentiated.
The data sample sizes used to evaluate genetic correlations were low between some traits in this analysis, which is reflected in the high SE associated with many of the parameters. As a result, caution should be exercised in considering the between-breed parameter differences to be truly attributable to breed and not to sampling issues and management differences between breeds. A dilemma is faced when using the information in the construction of selection indices over the use of estimated breed-specific parameter sets vs. a pooled set of parameters. By combining the data, some of the sampling problems can be accounted for. There is extensive discussion in the literature of the negative effects of errors in parameter estimates on the effectiveness of selection indices, most notably Sales and Hill (1976), who drew attention to the reduction in index accuracy between that predicted and that realized when inaccurate parameter estimates are used. Robustness of commercial evaluation systems is the key to their effective industrial application; hence, the way in which parameter sets are constructed needs to reflect this.
A common aim in developing genetic correlation matrices among large numbers of traits is to help facilitate multitrait BLUP evaluation procedures. This requires knowledge of the correlations between all traits considered, which is difficult to achieve as data sets recorded on different types of traits do not often share common individuals, especially in the case of sex-limited traits. For example, postweaning traits (growth and carcass) are often measured on male offspring, whereas reproductive traits are measured in the female offspring. There also is great variation in the amount of information available on different traits; however, developing knowledge of these genetic correlation matrices identifies early predictors of traits for which information is scarce until later in an animal’s life. These early predictors enable more informed selection decisions to be taken sooner.
|
Implications
|
|---|
In this study, direct and maternal genetic correlation matrices and variance components were estimated for nine traits contributing to aspects of the performance of cows in beef herds. The matrices were estimated for four breeds (Aberdeen Angus, Limousin, Simmental, South Devon), representing the most significant beef breeds contributing to the UK suckler beef industry. A weighted bending procedure was used to adjust the matrices to be positive definite following their construction from estimates of correlations between sub-groups of the traits. The resulting correlation matrices add to the information required in the undertaking of multitrait breeding value evaluations to assist selection of bulls to produce replacement heifers for the herd.
|
Footnotes
|
|---|
1 We are grateful to the Meat and Livestock Commission; MLC Signet Breeding Services; the Dept. of Environment, Food, and Rural Affairs; and the Scottish Executive Environment and Rural Affairs Dept. for funding this work through the LINK Sustainable Livestock Production Program. 
2 Correspondence: West Mains Road (phone: 44-0131-535-3225; fax: 44-0131-535-3121; e-mail: tim.roughsedge{at}sac.ac.uk).
Received for publication January 31, 2005.
Accepted for publication June 14, 2005.
|
Literature Cited
|
|---|
Amer, P. R., R. Crump, and G. Simm. 1998. A terminal sire selection index for UK beef cattle. Anim. Sci. 67:445–454.
Arango, J. A., L. V. Cundiff, and L. D. Van Vleck. 2002. Genetic parameters for weight, weight adjusted for condition score, height and body condition score in beef cows. J. Anim. Sci. 80:3112–3122.[Abstract/Free Full Text]
Bennett, G. L., and K. E. Gregory. 2001. Genetic (co)variances for calving difficulty score in composite and parental populations of beef cattle: I. Calving difficulty score, birth weight, weaning weight, and post weaning gain. J. Anim. Sci. 79:45–51.[Abstract/Free Full Text]
Bourdon, R. M., and J. S. Brinks. 1983. Genetic, environmental and phenotypic relationships among gestation length, birth weight, growth traits and age at first calving in beef cattle. J. Anim. Sci. 55:543–553.
Brotherstone, S., R. F. Veerkamp, and W. G. Hill 1997. Genetic parameters for a simple predictor of lifespan of Holstein-Friesian dairy cattle and its relationship to production. Anim. Sci. 65:1–142.
Crump, R. E., G. Simm, D. Nicholson, R. H. Findlay, J. G. E. Bryan, and R. Thompson. 1997a. Results of multivariate individual animal model genetic evaluations of British pedigree beef cattle. Anim. Sci. 65:199–207.
Crump, R. E., N. R. Wray, R. Thompson, and G. Simm. 1997b. Assigning pedigree performance records to contemporary groups taking account of within-herd calving patterns. Anim. Sci. 65:193–198.
Gilmour, A. R., B. J. Gogel, B. R. Cullis, S. J. Welham, and R. Thompson. 2002. ASReml User Guide Release 1.0. VSN International Ltd., Hemel Hempstead, U.K.
Gregory, K. E., L. V. Cundiff, and R. M. Koch. 1995a. Genetic and phenotypic (co)variances for growth and carcass traits of pure-bred and composite populations of beef cattle. J. Anim. Sci. 73:1920–1926.[Abstract]
Gregory, K. E., L. V. Cundiff, and R. M. Koch. 1995b. Genetic and phenotypic (co)variances for production traits of female populations of purebred and composite beef cattle. J. Anim. Sci. 73:2235–2242.[Abstract]
Gutierrez, J. P., I. Alvarez, I. Fernandez, L. J. Royo, J. Diez, and F. Goyache. 2002. Genetic relationships between calving date, calving interval, age at first calving and type traits in beef cattle. Livest. Prod. Sci. 78:215–222.
Hill, W. G., and R. Thompson. 1978. Probabilities of non-positive definite between-group or genetic covariance matrices. Biometrics 34:429–439.
Jorjani, H., L. Klei, and U. Emanuelson. 2003. A simple method for weighted bending of genetic (co)variance matrices. J. Dairy Sci. 86:677–679.[Abstract/Free Full Text]
Kaps, M., W. O. Herring, and W. R. Lamberson. 1999. Genetic and environmental parameters for mature weight in Angus cattle. J. Anim. Sci. 77:569–574.[Abstract/Free Full Text]
Koots, K. R., J. P. Gibson, and J. W. Wilton. 1994. Analyses of published genetic parameter estimates for beef production traits. 2. Phenotypic and genetic correlations. Anim. Breed. Abstr. 62:826–853.
Lubbers, R., S. Brotherstone, V. P. Ducrocq, and P. M. Visscher. 2000. A comparison of a linear and proportional hazard approach to analyse discrete longevity data in dairy cows. Anim. Sci. 70:197–206.
MacGregor, R. G., and N. H. Casey. 1999. Evaluation of calving interval and calving date as measures of reproductive performance in a beef herd. Livest. Prod. Sci. 57:181–191.
Meyer, K., D. J. Johnston, and H. U. Graser. 2004. Estimates of the complete genetic covariance matrix for traits in multi-trait genetic evaluation of Australian Hereford cattle. Aust. J. Agric. Res. 55:195–210.
Nephawe, K. A., L. V. Cundiff, M. E. Dikeman, J. D. Crouse, and L. D. Van Vleck. 2004. Genetic relationships between sex-specific traits in beef cattle: Mature weight, weight adjusted for body condition score, height and body condition score of cows, carcass traits of their steer relatives. J. Anim. Sci. 82:647–653.[Abstract/Free Full Text]
Northcutt, S. L., and D. E. Wilson. 1993. Genetic parameter estimates and expected progeny differences for mature size in Angus cattle. J. Anim. Sci. 71:1148–1153.[Abstract]
Phocas, F., and D. Laloë. 2003. Evaluation models and genetic parameters for calving difficulty in beef cattle. J. Anim. Sci. 81:933–938.[Abstract/Free Full Text]
Rege, J. E. O., and T. R. Famula. 1993. Factors affecting calving date and its relationship with production traits of Hereford dams. Anim. Prod. 57:385–395.
Roughsedge, T., P. R. Amer, and G. Simm. 2003a. A bioeconomic model for the evaluation of breeds and mating systems in beef production enterprises. Anim. Sci. 77:403–416.
Roughsedge, T., P. R. Amer, R. Thompson, and G. Simm. 2005. Development of a maternal breeding goal and tools to select for this goal in UK beef production. Anim. Sci. (in press).
Roughsedge, T., S. Brotherstone, and P. M. Visscher. 1999. Quantifying genetic contributions to a dairy cattle population using pedigree analysis. Livest. Prod. Sci. 60:359–369.
Roughsedge, T., B. Lowman, P. R. Amer, and G. Simm. 2003b. Impacts of alternative replacement breeding systems on biological and economic performance in beef suckler production using a herd level bioeconomic model. Anim. Sci. 77:403–416.
Sales, J., and W. G. Hill. 1976. Effect of sampling errors on efficiency of selection indices. 2. Use of information on associated traits for improvement of a single important trait. Anim. Prod. 23:1–14.
Snelling, W. M., B. L. Golden, and R. M. Bourdon. 1995. Within-herd genetic analyses of stayability of beef females. J. Anim. Sci. 73:993–1001.[Abstract]
Trus, D., and J. W. Wilton. 1988. Genetic parameters for maternal traits in beef cattle. Can. J. Anim. Sci. 68:119–128.
Willham, R. L. 1963. The covariance between relatives for characters composed of components contributed by related individuals. Biometrics 19:18–27.
This article has been cited by other articles:
|
|
|
|
 
J. P. Gutierrez, F. Goyache, I. Fernandez, I. Alvarez, and L. J. Royo
Genetic relationships among calving ease, calving interval, birth weight, and weaning weight in the Asturiana de los Valles beef cattle breed
J Anim Sci,
January 1, 2007;
85(1):
69 - 75.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
