|
|
||||||||
ANIMAL GENETICS |
Department of Animal Breeding and Genetics, Swedish University of Agricultural Sciences, SE-75007 Uppsala, Sweden
| Abstract |
|---|
|
|
|---|
Key Words: Beef Cattle Birth Weight Carcass Composition Dystocia Genetic Parameters
| Introduction |
|---|
|
|
|---|
Calving difficulty is closely correlated with heavier birth weight, as shown in reviews by Meijering (1984)
and Koots et al. (1994b)
and in a previous study on beef breeds in Sweden (Eriksson et al., 2004
). Birth weight, in turn, is positively correlated with performance traits such as growth rate and muscling score (Mohiuddin, 1993
; Koots et al., 1994b
). There are, however, few published studies on genetic correlations between calving performance and carcass traits directly. Unfavorable genetic correlations were found between calving ease and some carcass traits by MacNeil et al. (1984)
, Tilsch (1986)
, and Splan et al. (1998)
.
The aim of the current study was to estimate genetic correlations between calving difficulty score and carcass traits in Charolais and Hereford, treating first and later parity calvings as different traits. To better understand the biology underlying such relationships, genetic correlations were also estimated between birth weight and carcass traits.
| Materials and Methods |
|---|
|
|
|---|
Records lacking information on fixed effects in the model (see model definition below) were discarded, as were observations in any contemporary group with fewer than three observations. Records on calving difficulty score and birth weight, where the calf resulted from embryo transfer, or with a dam older than 14 yr, were excluded. Records of birth weight outside the accepted range of 18 to 75 kg were discarded, as were records of calving difficulty scores for twin-born calves. Observations on carcass traits for bulls younger than 10 mo or older than 24 mo at slaughter were not used in the analyses.
The numbers of observations after these edits are shown for each trait in Table 1
. For both breeds, more than 80% of the original records of calving difficulty score, and about 90% for birth weight, were used in the analyses. For carcass traits, approximately 80 and 60% of the observations could be used for Charolais and Hereford, respectively.
|
The average age at calving was 26.4 mo for Charolais heifers and 25.8 mo for Hereford heifers. In later parities, the average age at calving was about 68 mo for both breeds. Dams whose calves had a recorded birth weight were on average 58 mo old at calving. Age at slaughter was on average 15.7 and 17.2 mo for Charolais and Hereford bulls, respectively.
The average number of observations per herd-year contemporary group varied for both breeds, between 6 and 8 for calving difficulty score at first parity and the carcass traits, and between 11 and 15 for calving difficulty score at later parities and birth weight.
Trait Definitions
Calving Performance and Birth Weight.
Calving difficulty score at first and later parities was recorded in three classes: 1 = unassisted calving, 2 = normal calving (assisted by one person), and 3 = difficult calving (assistance by more than one person or necessitating veterinary aid). Calves without records of calving difficulty score (less than 0.01%) were assumed to have had an easy birth. Birth weight was recorded in kg up to 4 d after calving.
Carcass Weight, Fleshiness, and Fatness.
Carcass traits of young bulls were determined at commercial slaughterhouses. The carcasses were trimmed in accordance with EU regulations and weighed warm shortly after classification. The official carcass weight was calculated as 98% of the warm carcass weight (Statens Jordbruksverk, 1998
). Carcass fleshiness grade and carcass fatness grade were judged subjectively by trained graders in accordance with the EU system (S)EUROP. The carcass fleshiness grade describes the meat content of the carcass (shape of muscles and s.c. fat), whereas carcass fatness grade is a measure of the content of fat in carcass. Carcass fleshiness and fatness grades are both measured on a scale from 1 to 15, where higher values denote higher meat and fat contents, respectively. The carcass classification is described in greater detail in Eriksson et al. (2003)
.
Estimation of Genetic Parameters
The two breeds were analyzed separately. Due to computational constraints, bivariate animal model analyses were used for all trait combinations. The following linear statistical models were used:
![]() |
![]() |
![]() |
![]() |
where yCD1 and yCD>1 are observed calving difficulty scores at first and later parities, respectively, yBW is observed birth weight (kg), yCARC is observed carcass weight (kg), carcass fleshiness grade, or carcass fatness grade.
Fixed Effects.
The fixed effects are combination of herd and year of birth (hy), birth season (bseas), combination of age group of dam with sex of calf (dam age-sex), age group of dam (dam age), sex of calf (sex), type (single or twin) of birth (type), age in days at weighing birth weight (bwage), slaughter season (sseas), and age class at slaughter (sage). The ages at which dams calved at first parity were grouped as 20 to 23, 24, 25 and 26, 27 to 30, 31 to 35, and 36 to 44 mo; for later parities the classes were less than 42, 43 to 54, 55 to 66, 67 to 78, and 79 to 168 mo of age at calving. For birth weight and carcass traits, three classes of dams age were used: less than 3 yr, 3 to 4 yr, and more than 4 yr at calving. Four birth seasons (November to February, March and April, May and June, and July to October) were used in the analyses.
The slaughter seasons used were January to March, April and May, June and July, August and September, and October to December. There were 14 age classes at slaughter. The records of carcass traits used in the analyses came from 18 different slaughterhouses. The effect of slaughterhouse was found not to be significant (Eriksson et al., 2003
) and was therefore excluded from the model.
Random Effects.
Random effects in the model are the permanent environmental effect of the dam (pe), additive maternal genetic effect of the dam (m), and additive direct genetic effect of the animal (a) and residual (e). For random effects, zero means and variances
, G0
A, and R0
I were assumed for permanent environmental effect of dam, genetic effects and residuals, respectively, where G0 is the covariance matrix between direct and maternal additive genetic effects, A is the numerator relationship matrix,
is the Kronecker product, and R0 is the covariance matrix between residuals for both traits. Covariances between genetic and environmental effects were assumed to be zero and variances due to dominance or epistatic effects were assumed not to exist.
(Co)variance Estimation.
(Co)variances were estimated using the average information algorithm (Jensen et al., 1997
) for restricted maximum likelihood included in the DMU package (Jensen and Madsen, 1994
). The convergence criterion was chosen so that the norm of update vector for the (co)variance components was less than 104. Asymptotic standard errors of (co)variance components were computed from the inverse average information matrix. Standard errors of genetic correlations were obtained by Taylor series expansions (Madsen and Jensen, 2000
).
Heritabilities.
Direct and maternal heritabilities on the observable scale were calculated as
and
, respectively, where
for calving difficulty score at first parity,
for calving difficulty score at later parities and for birth weight, and
for the carcass traits. To make comparisons possible with other studies, heritabilities of calving difficulty score on the underlying continuous scale were approximated from those on the observable scale, using a transformation described by Gianola (1982)
.
| Results |
|---|
|
|
|---|
|
|
|
|
|
Genetic Correlations with Calving Difficulty Score.
Carcass weight was positively genetically correlated with direct effects on calving difficulty score (0.12 to 0.40), but weakly or negatively correlated with maternal effects on calving difficulty score (0.70 to 0.07).
Estimated genetic correlations between carcass fatness grade and direct effects on calving difficulty score at first parity were zero to moderate and positive (0.42 for Charolais and 0.04 for Hereford), whereas the corresponding correlations with calving difficulty score at later parities were negative (0.16 for Charolais and 0.39 for Hereford). All estimated genetic correlations between carcass fatness grade and maternal effects on calving difficulty score were negative (0.17 to 0.21), except for later parities in Charolais where a low and positive correlation (0.20) was estimated.
Moderate and positive unfavorable genetic correlations (0.42 for Charolais and 0.54 for Hereford) between carcass fleshiness grade and maternal effects on calving difficulty score at first parity were found for both breeds. At later parities, this correlation was close to zero for Charolais and negative for Hereford (0.60). The genetic correlation between carcass fleshiness grade and direct effect on calving difficulty score was slightly negative for Charolais (0.12) but moderate and positive for Hereford (0.54) at first parity, and weak at later parities.
Direct-Maternal Correlations.
Estimated correlations between direct and maternal genetic effects were generally negative (Table 3
). The only exception was for calving difficulty score at later parities in Hereford, where a positive direct-maternal correlation was estimated (though with a standard error higher than 1.0).
Residual Correlations.
Estimated residual correlations were generally weak and nonsignificant, ranging from 0.11 to 0.23, with the closest relationship found between birth weight and carcass weight in Hereford (Table 7
).
|
| Discussion |
|---|
|
|
|---|
Carcass Traits.
The heritability estimates for carcass fleshiness and fatness grades, for both breeds, were close to earlier estimates based on a slightly larger Swedish dataset (Eriksson et al., 2003
). The heritabilities for carcass weight were, however, somewhat lower than the estimates from trivariate analyses of carcass data in the aforementioned study. Moderate to high heritabilities of carcass traits related to weight, fatness, and muscularity have been found in several studies (see review by Koots et al., 1994a
), and in more recent studies by Splan et al. (1998)
and Crews et al. (2003)
.
Genetic Correlations
Correlations with Carcass Weight.
Favorable negative genetic correlations between calving difficulty as a trait of heifer dams and carcass weight of their male half-sibs have previously been reported for beef cattle by MacNeil et al. (1984)
and Splan et al. (1998)
. MacNeil et al. (1984)
argued that this might be explained by an increased mature size of dams due to selection for higher carcass weight. Crews and Kemp (1999)
estimated moderate and positive (0.44 to 0.52) genetic correlations between carcass weight and direct and maternal effects on birth weight. Our results were in agreement with these findings.
The positive genetic correlation between carcass weight and direct effect on birth weight in our study probably explains the unfavorable positive correlation between carcass weight and direct effect on calving difficulty score. Carcass weight was also positively correlated with maternal effect on birth weight. However, carcass weight was weakly, or even negatively, correlated with maternal calving difficulty score. This suggests that an increased birth weight due to maternal effects following selection for carcass weight would be compensated for by an increased size of the dam. The latter is supported by the moderately positive genetic correlation estimated between dam weight and pelvic area (Naazie et al., 1991
) and the negative correlation between pelvic area and maternal calving difficulty (Bennett and Gregory, 2001
).
Correlations with Carcass Fatness.
Some authors have previously suggested that selection for a leaner carcass tends to increase calving difficulty, at least as a dam trait, as was also indicated in our study. MacNeil et al. (1984)
concluded that selection for decreased fat trim at constant age would lengthen maternal gestation, and increase maternal effects on birth weight and maternal calving difficulty. Also Splan et al. (1998)
found negative genetic correlations (0.14 to 0.29) between calving difficulty of heifers and carcass traits involving fatness of steers. Philipsson (1976)
reported a phenotypic tendency for very fat or very lean dairy heifers to have more difficult calvings, and Paputungan et al. (1994)
, found that there was an optimal body condition score for calving performance of beef heifers. Differences in average fatness of heifers and older cows in relation to the optimum may influence the correlations estimated between fatness and maternal calving difficulty score at first and later parities.
The results of our analyses suggest that a higher fatness grade at slaughter of young beef cattle is genetically correlated with lower birth weight. This was also found by Crews and Kemp (1999)
, who estimated genetic correlations with carcass fatness of 0.44 for direct (and 0.09 for maternal) birth weight. Earlier-maturing heifers, which put on fat at a lower weight and younger age, might also be more physiologically mature at the time of first calving. Bennett and Gregory (2001)
found that lighter birth weight and reduced calving difficulty were associated with younger heifer age at puberty.
Correlations with Carcass Fleshiness.
No obvious trends across breeds or parities were found for correlations between carcass fleshiness and calving difficulty score or birth weight. The unfavorable genetic correlation estimated between carcass fleshiness grade and maternal effect on calving difficulty score at first parity in Charolais could not be explained by higher birth weights due to maternal effects. Rather, this could have been the result of a more developed musculature around the birth canal, causing difficulties at calving.
No previous studies were available that could be used to compare the relationship between calving performance and carcass fleshiness grade according to the (S)EUROP system. In studies on other traits related to the content of lean meat in carcass, unfavorable relationships with calving ease or birth weight have often been found. Crews and Kemp (1999)
estimated positive genetic correlations (0.51 to 0.54) between direct effect on birth weight and both rib eye area and percentage lean yield. They found weak and nonsignificant correlations between the same carcass traits and maternal birth weight, whereas MacNeil et al. (1984)
estimated a genetic correlation of 0.30 between retail product of males and birth weight as a trait of heifer dams. Bennett and Gregory (2001)
reported that the direct effect on percentage of retail product was associated with a direct effect on increased calving difficulty score. Mohiuddin (1993)
presented zero to moderate positive genetic correlations between muscling score and birth weight. However, for French beef bulls, Renand (1985a
,b)
estimated weak negative genetic correlations between calving difficulty and both live final fleshiness and carcass fleshiness grade.
Breed Differences.
Certain breed differences in correlations were indicated by our results. The correlations were estimated with high standard errors, but it is likely that breed differences exist in relationships between carcass and calving traits, as the two breeds in our study are of somewhat different types. In a comparison between beef breeds by Roughsedge et al. (2001)
, Hereford was characterized as an earlier-maturing breed compared with Charolais, reaching puberty at an earlier age and lighter weight; also, Hereford calves had lighter birth weights. Hereford was one of the breeds with thickest fat cover at slaughter, whereas in a review by Marshall (1994)
, Charolais ranked highest among several sire breeds for carcass weight. Similar phenotypic differences in birth weight and carcass measurements were also found between Charolais and Hereford in our study (Table 1
).
Differences Between Parities.
For Charolais, the estimated genetic correlations between calving difficulty score and carcass traits were generally weaker at later parities than at first parity. This could be expected for maternal traits, as the heifers were more similar in age to the bulls for which carcass traits were recorded. For Hereford, closer correlations were estimated between carcass traits and calving difficulty score at later parities, although with high standard errors.
Some of the estimated genetic correlations with calving difficulty score differed between parities regarding sign and size, such as correlations with carcass fatness for Charolais and with carcass fleshiness for Hereford. Heifers have not yet reached their full mature size and young beef heifers may have stored less fat than older cows. This might influence the relationships between traits at first and later parities. In our study, however, few genetic correlations were significant, and thus we could not draw any firm conclusions about differences between parities regarding relationships between carcass traits and calving performance.
Choice of Data and Models for Analyses
In our study, carcass trait records were available for young bulls only. Heifers generally have thicker backfat and are less heavily muscled than young bulls (Crews and Kemp, 2001
). Moderate to high positive genetic correlations between live measurements of fat depth, live weights, and muscularity of replacement heifers on the one hand, and carcass fat depth, weight, and LM area of steers on the other, were estimated by Crews and Kemp (2001)
. Estimated correlations between maternal effects on calving traits and direct effects on carcass traits reflect the relationship between maternal ability and size and body composition of dams. The use of linear analyses for the categorical trait calving difficulty score is theoretically suboptimal, but has been shown to work for practical purposes (Meijering, 1985
; Hagger and Hofer, 1990
; Ramirez-Valverde et al., 2001
). Initial attempts to analyze calving difficulty score using threshold models failed due to the large number of contemporary groups in which all recorded calving difficulty scores fell into the same category (so-called "extreme category problems") (Eriksson et al., 2004
). Even though estimated heritabilities on the observable scale and residual correlations are biased due to the assumption of linearity, genetic correlations should be unaffected (Gianola, 1982
), and these were the main focus of interest in this study.
| Implications |
|---|
|
|
|---|
| Footnotes |
|---|
2 Correspondence: P.O. Box 7023 (phone: +46-18-672007; fax: +46-18-672648; e-mail: susanne.eriksson{at}hgen.slu.se).
Received for publication September 11, 2003. Accepted for publication April 14, 2004.
| Literature Cited |
|---|
|
|
|---|
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] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |