HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) All Versions of this Article: jas.2006-093v1 85/3/610    most recent Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Speidel, S. E. Articles by Garrick, D. J. Search for Related Content PubMed PubMed Citation Articles by Speidel, S. E. Articles by Garrick, D. J.

Weaning weight inheritance in environments classified by maternal body weight change¹

S. E. Speidel^*, R. M. Enns^*,2 and D. J. Garrick^*,

^* Department of Animal Sciences, Colorado State University, Fort Collins 80523-1171; and and Institute of Veterinary, Animal and Biomedical Sciences, Massey University, Palmerston North, New Zealand

	Abstract

Top Abstract INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION IMPLICATIONS LITERATURE CITED

 
In good environments, cow intake is sufficient for their own growth and for milk production to support their calf. In poor environments, cows lose BW or may reduce milk supply to maintain themselves. Heritability for direct genetic and maternal components of weaning weight as well as the correlations between these components might be expected to vary according to these circumstances. The purpose of this study was to estimate heritability and genetic correlations for the direct genetic and maternal components of weaning weight classified in 2 environments according to maternal BW gain and to identify whether a single heritability estimate is appropriate for the differing environments experienced by cows from year to year. Data used in this analysis was obtained from the Red Angus Association of America and consisted of 96,064 cow BW observations and 27,534 calf weaning weight observations. A dam’s change in BW from one year to the next was used to classify each calf’s weaning weight into 1 of 2 environmental groups, those being good or poor. Best linear unbiased estimates of the change in cow BW with age were obtained from analysis of cow BW using a repeatability model. If the phenotypic change in cow BW exceeded this average BW change, the calf’s weaning weight associated with the end of this time frame was classified as having been observed in a good environment. If not, the calf’s corresponding weaning weight was classified as having occurred in a poorer than average environment. Heritability estimates of 0.24 ± 0.03, 0.24 ± 0.03, 0.13 ± 0.02, and 0.14 ± 0.02 were obtained for weaning weight good direct, poor direct, good maternal, and poor maternal, respectively. Correlations between direct genetic and maternal weaning weight components in the good and poor environments were –0.47 ± 0.08 and –0.20 ± 0.09, respectively. These variance components are not sufficiently distinct to warrant accounting for dam nutritional environment in national cattle evaluation.

Key Words: beef cattle • heritability • weaning weight • environment

	INTRODUCTION

Top Abstract INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION IMPLICATIONS LITERATURE CITED

 
Nutritional burdens due to climatic differences, drought conditions, and management-induced feed limitations may alter the ability of cows and their calves to express genetic differences in direct and maternal ability. Within a herd, differences in the ability of cows to secure adequate nutrition may exist due to differences in grazing behavior resulting from differences in genetics (Winder et al., 1995, 1996; Bailey et al., 2001) or learning environment (Howery et al., 1998), and in turn may influence production levels, such as with cow dominance (Phillips and Rind, 2002). These factors likely produce differences in DMI, which may then result in the varying relationships between DMI of the cow and weaning weight of the calf, as reported by Jenkins and Ferrell (1994) across breeds.

The partitioning of variation in weaning weights recorded under differing environmental conditions based on individual cow circumstances across years has not been investigated. If estimates of heritability differ for good vs. poor cow environments, and genetic correlations are not unity between environmental conditions, national cattle evaluations would not appropriately account for these differences, and as a result, may inappropriately rank sires.

The purpose of this study was to evaluate the influence of cow environment on expression of genetic differences in weaning weight. The goal was to categorize individual cows as willing to erode their own BW to support lactation for the benefit of their calf or, in contrast, to achieve above average increases in their own body reserves. Heritability and genetic correlations for weaning weights recorded in these alternative cow nutritional environments were estimated to identify whether a single value is appropriate for the differing environmental conditions experienced by cows from year to year.

	MATERIALS AND METHODS

Top Abstract INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION IMPLICATIONS LITERATURE CITED

 
Because these data were obtained from an existing historical database (Red Angus Association of America, RAAA) they were not subject to animal care and use committee approval.

The general methodology for this study was to classify weaning weights in the RAAA database as occurring in a good or poor cow environment and to subsequently estimate direct, maternal, and permanent environmental variance components from both environments and to compare those with estimates from a model typically used in national cattle evaluation. Good and bad cow environments were determined based on changes in individual cow BW from year to year. The specifics of these procedures are presented in more detail below.

Cow BW Data
Cow BW observations were collected at calf weaning and spanned the years 1981 to 2004. Weight observations of the same cows as yearlings were pooled with these numbers and used to calculate BW change from 1 to 2 yr of age as described later.

Contemporary groups for cow and yearling BW were formed in a similar manner as used in the national cattle evaluation, in accordance with RAAA guidelines. These included percent Red Angus (%RA) of the cow (50 %RA < 87.5; %RA 87.5) and the weaning contemporary group of her calf (at the time at which the cow BW was recorded). Weaning contemporary groups for the calf were formed using work and management group of the calf, calf sex (female, male, and castrated male), breed group of the calf (50 %RA < 87.5; %RA 87.5), weaning date, weaning age group (130 to 159 d, 160 to 250 d, or 251 to 280 d), and Brahman influence (yes/no). Yearling contemporary groups were formed from the weaning contemporary group designation outlined above, in combination with yearling work and management groups. Weaning and yearling management groups are breeder-defined cohorts that have experienced similar management practices and opportunities to perform. Weaning and yearling workgroups identified individuals belonging to the same ranch or herd.

Cow BW observations must have been associated with a valid contemporary group designation (9,912 groups). Some 2,683 observations were removed because their contemporary group exhibited no variation. The final cow BW data set contained 96,064 observations representing 34,223 individual cows.

A 3-generation pedigree consisting of animal ID, sire ID, and dam ID was built based on animals included in the final data set. This pedigree consisted of 69,049 individuals, including 8,071 sires, 39,949 dams, and 6,043 maternal grandsires.

Weaning Weight Data
Age-adjusted weaning weight data from the RAAA national database was used in the analysis. Weaning weight observations were discarded if

1. The animal was less than 50% RA;
   
2. The animal was an embryo transfer calf;
   
3. Actual weaning weight was less than 102 or greater than 544 kg;
   
4. The animal was fostered onto another cow; or
   
5. Weaning age was less than 130 d or greater than 280 d.
   

A calf’s weaning observation, to be included, must have been associated with a corresponding BW observation on its dam. That dam must have had a BW observation taken in a previous year also so that we could determine whether the weaning weight was recorded in a good or a poor environment. These requirements resulted in a data set including 27,534 weaning weight observations in 1,418 contemporary groups.

A 3-generation pedigree was constructed for the calves in this final data file. Foundation dams (dam’s whose sire and dam were unknown) with only 1 calf and who were missing their own weaning weight observation (n = 3,703) were removed from the pedigree. This resulted in a final pedigree consisting of 47,185 individual animals, including 3,620 sires, 19,765 dams, and 2,818 maternal grandsires.

Weaning Weight Environmental Classification
Yearling and older cow BW were analyzed using a single trait, repeatability animal model to classify each weaning weight observation as occurring in a good or poor cow environment. Fixed effects included in the analysis were cow BW contemporary group and age of cow (1 to 18 yr). Random effects included direct animal and permanent environmental effects. The cow BW model used in the analysis was

where X, Z, and W were incidence matrices relating cow BW observations in y to fixed (b), random genetic (u), and permanent environmental effects (p), with e defining a vector of random residual errors. Random effects in the model were assumed to have zero means and variances, represented as follows:

where A was Wright’s numerator relationship matrix, I_D and I_N were identity matrices with order equivalent to the number of females with mature BW observations and the total number of BW observations, respectively. Additive genetic (²_u), permanent environmental (²_p), and residual (²_e) variances of cow BW were assumed to be 1,872, 533, and 427 kg², respectively (Evans, 2001). This linear model was implemented and analyzed using ASREML (Gilmour et al., 2002).

Best linear unbiased estimates of cow age solutions were then used to classify the calf’s weaning weight as occurring in a good or a poor cow environment. If a cow’s actual change in BW from one year to the next met or exceeded the expected BW change as determined by the differences between age solutions for cow BW obtained above, her calf’s weaning weight, corresponding to the end of this time frame, was classified as having occurred in a good environment. If the observed BW change failed to meet this expected BW change, her calf’s weaning weight, corresponding to the end of this time period, was classified as having occurred in a poor environment. For example, according to the best linear unbiased estimates of cow BW per year of age, cow BW was expected to increase 41.6 kg as the cow aged from 2 to 3 yr of age. If the cow’s actual BW change from 2 to 3 yr of age met or exceeded 41.6 kg, her calf weaned at 3 yr of age was classified as having a weaning weight in a good environment.

Estimation of Weaning Weight Variance Components
Variance components for weaning weight were estimated using 2 models. In model 1, weaning weight was analyzed using a single trait, multiple component animal model represented by

where y was a vector of weaning weight observations; X was a known incidence matrix relating the fixed effects in b to the observations in y; and Z_w, Z_m, and Z_p were known incidence matrices relating the random animal effects u_w, u_m, u_p to the observations in y for direct, maternal, and maternal permanent environmental effects, respectively. The vector e represented random residuals unique to each weaning weight observation in y. The random effects were assumed to have means of zero and variances, shown as

where I_D and I_N were identity matrices with order equal to the number of dams with weaning weights on their calves and the number of weaning weight observations, respectively. Fixed effects included in the model were weaning contemporary group of the calf and a sex x age of dam interaction.

In model 2, weaning weight observations classified as having been observed in good or poor environments were analyzed as correlated traits using a bivariate model. The model used in the analysis is represented in matrix form as

where y was a vector of weaning weight observations classified by subscript as good (g) or below average (b), respectively; X was an incidence matrix relating good and poor weaning weight observations to their respective fixed effects in b; Z was a known incidence matrix relating good and poor weaning weight observations to their random effects in u, a vector representing random additive effects for direct, maternal, and maternal permanent environmental effects for good and poor weaning weight, respectively; and e was a vector of good and poor weaning weight random residuals unique to each observation. The random effects were assumed to have means of zero and genetic variances represented by

maternal permanent environmental variances represented by

and residual variances as

In the above equations, subscripts w_g, m_g, p_g, w_b, m_b, and p_b refer to weaning "good" direct, maternal, maternal permanent environment; and weaning "poor" direct, maternal, and maternal permanent environment, respectively. The I_Dg, I_Db, I_g, and I_b were identity matrices with order equal to the number of dams whose offspring have good weaning observations, have poor weaning observations, the total number of good observations and the total number of poor observations, respectively. The matrix E contained all zeros, except for locations corresponding to particular dams with calves in both environments, in which case that element of E was unity. Residual covariances were zero because of the nature of this analysis, in which no individual calf could have both a good and a poor observation recorded.

Variance component estimates were obtained for the above 2 weaning weight models using ASREML (Gilmour et al., 2002). The numerator relationship matrix was constructed for 47,185 individuals with an average inbreeding level of 0.026.

	RESULTS AND DISCUSSION

Top Abstract INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION IMPLICATIONS LITERATURE CITED

 
Cow BW Analysis
A total of 96,064 BW observations represented 34,223 individual cows in the data set. The mean BW was 488.8 kg with minimum and maximum BW of 226.8 and 904.9 kg, respectively. The minimum BW was a yearling BW.

The best linear unbiased estimates for cow BW changes are in Table 1. These differences were used to classify the calf’s weaning weight as coming from a good or poor cow environment. Cows were expected to gain 118.9 kg from 1 to 2 yr of age and 41.6 kg from 2 to 3 yr of age based on the fixed effects solutions from the cow BW analysis. This large discrepancy is a result of a varying amount of time elapsing between these measurements. Yearling BW was recorded at approximately 1 yr of age, and 2-yr-old cow BW was taken approximately 1.5 yr later at the time of weaning her first calf. During the first year of these 1.5 yr, the animals are still growing and likely have adequate to abundant fat stores. A trend for gaining BW was evident up to 7 yr of age at which point cows experienced BW loss until 14 yr of age. Few cows were represented above 14 yr of age, and the corresponding age change effects had large standard errors. Weaning weights of cows older than 14 yr were discarded for all subsequent results reported.

View larger version (17K): [in this window] [in a new window]   Figure 1. Percentage of the total number of good vs. poor observations per year of age classified according to the nutritional status of the dam as represented by above-or below-average BW gain from one year to the next.

Table 2 shows the number of calves attributed to cows in the data set. About one-third of cows had a single calf in the analysis. About one-sixth of the cows had 2 calves, and these were approximately equally distributed across the 4 sequences poor-poor, poor-good, good-poor, and good-good. There were a total of 13,582 dams that represented 11 different good vs. poor combinations. There were 7,288 dams with only 1 observation, whereas only 2 dams had 11 different observations. The largest percentage (58%) of good observations was represented where 10 observations were recorded. However, there were approximately equal numbers of good and poor observations across all combinations.

Weaning Weight Analysis
Weaning weight summary statistics are reported in Table 3. The mean weaning weight of calves in the poor environment (240.3 kg) was approximately 6 kg lighter than in the good environment (246.9 kg). The poor environment did not contain the lowest weaning weight observation, but the highest weaning weight (375.1 kg) observation in this class was approximately 26 kg less than the maximum weaning weight observation in the good environment (401.0 kg). These simple summary statistics show no practical difference in weaning weight observations between the overall, good, and poor weaning weight observations.

Maternal heritability estimates in the poor environment were marginally higher than in the good environment (0.13 ± 0.02 in the good vs. 0.14 ± 0.02 in the poor). The difference between the maternal estimates of heritability in the good and poor environments was not significant when considering the standard error of the estimates. These results conflict with those reported by Brown et al. (1993), where it was found that maternal effects for growth tend to vary with nutritional environment.

Correlations between the direct effects (0.96) in good and poor cow environments and between maternal effects (0.89) in good and poor environments were extremely high. On a wider scale, De Mattos et al. (2000) found genetic correlations in the Hereford breed for direct and maternal effects across Canada, Brazil, and the United States to be greater than 0.80, indicating sires ranked similarly in all environments. The high genetic correlations found here for direct and maternal effects across good and poor environments and the findings reported by De Mattos et al., (2000) above indicate weaning weight observations in these 2 environments may not be different traits.

The negative genetic correlations between the direct and maternal components found in model 1 were confirmed in the results from fitting model 2 (–0.47 in the good environment and –0.20 in the poor environment). However, weaning weight maternal was more strongly correlated to its direct counterpart in the good environment than in the poor environment. This agrees with our interpretation of the results reported by Snelling et al. (1996) where the direct maternal genetic correlation decreased from an average of 0.03 to –0.21 with improved environment. Direct comparisons of results are difficult because that study did not attempt to identify cow environment within herd, but appear to support the finding reported herein. There was little difference in estimated genetic correlations between direct genetic effects in one environment and maternal genetic effects in the other environment (–0.25 and –0.29).

The maternal permanent environmental effects for these 2 traits are in Table 5. Maternal permanent environment effects accounted for a similar proportion of variation in weaning weight in good (0.17) and in poor environments (0.15). The correlation between these 2 effects was 0.72.

It might be argued that classification of calves according to the BW gain of their dams imposes a form of indirect selection on weaning weight. Should that be the case, estimates of genetic and perhaps phenotypic variation from fitting model 2 (²_PGood: 539 kg², ²_PBad: 478 kg²) would be eroded in relation to the estimates obtained from model 1 (²_P: 421 kg²). The results do not provide any evidence of reduced direct, maternal genetic, or phenotypic variation.

It was hypothesized that anabolic and catabolic systems are under different genetic control, and therefore the inheritance of maternal effects may differ in cows that have above- vs. below-average BW gains during lactation. However, unexpected findings were the near-perfect genetic correlation and high permanent environmental correlation between maternal effects in the 2 environments. These results may argue that the classification of cows was poorly achieved using the single mature BW observed at weaning each year. Further studies with cows classified using a better measure of cow energy status (e.g., monthly BW, condition scores, or ultrasonic measures) would be desirable.

Additional analysis prompted by the comments of an anonymous referee were undertaken using model 2 but with stricter classification of cows into the poor environment. A cow’s yearly BW change had to be in the bottom 10% of BW gain for their age group for their calf’s corresponding weaning weight to be classified into the poor environment. All other BW were classified as having occurred in a good environment. Heritabilities and genetic correlations for this good/poor division are in Table 6. No differences were seen between heritability estimates in both good and poor environments for both weaning weight direct (0.24 ± 0.03 in the good environment vs. 0.25 ± 0.09 in the poor environment) and weaning weight maternal (0.13 ± 0.02 in the good environment vs. 0.13 ± 0.08 in the poor environment).

Alternative methods for classifying individual, yearly, cow environment were considered, but all introduced confounding of genetic potential and environmental classification or altered the question addressed by classifying entire contemporary groups as performing in a single environment. For instance, grouping calves based on cow milk EPD would reduce genetic variance in milk production.

In an effort to assure that there was no confounding of growth and milk production potential in this study, an additional analysis was performed. Cows with a single calf weaned were eliminated from the data because these seemed the most likely candidates for confounding because cows with multiple observations readily reclassified as previously discussed. All heritability and genetic correlation estimates were within 1 standard error as reported in Table 5 with the exception of maternal heritability in the good environment which increased to 0.30 ± 0.03 (as did maternal heritability in the poor environment) and C², which was 0.14 ± 0.02 and 0.11 ± 0.02 in the good and poor environments, respectively.

	IMPLICATIONS

Top Abstract INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION IMPLICATIONS LITERATURE CITED

 
No differences were found between the heritability estimates for the direct weaning weight effects according to nutritional environments. Maternal heritability was found to be elevated in the classified environments. Genetic correlations support the direct and maternal expressions as representing the same traits in these diverse nutritional environments. Accounting for heterogeneous genetic variances in national cattle evaluations increases computational effort and adds complexity to the interpretation of the resulting genetic predictions. The results presented here provide no support for altering the current approach for the analysis of weaning weights across environments where nutritional resources are diverse.

	Footnotes

 
¹ The authors wish to thank the Red Angus Association of America for providing the data used in this study.

² Corresponding author: mark.enns{at}colostate.edu

Received for publication February 16, 2006. Accepted for publication October 26, 2006.

	LITERATURE CITED

Top Abstract INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION IMPLICATIONS LITERATURE CITED

 


Bailey, D. W., D. D. Kress, D. C. Anderson, D. L. Boss, and E. T. Miller. 2001. Relationship between terrain use and performance of beef cows grazing in foothill rangeland. J. Anim. Sci. 79:1883–1891.[Abstract/Free Full Text]

Brown, M. A., A. H. Brown, Jr., W. G. Jackson, and J. R. Miesner. 1993. Genotype x environment interactions in postweaning performance to yearling in Angus, Brahman, and reciprocal-cross calves. J. Anim. Sci. 71:3273–3279.[Abstract]

De Mattos, D., J. K. Bertrand, and I. Misztal. 2000. Investigation of genotype x environment interactions for weaning weight for Herefords in three countries. J. Anim. Sci. 78:2121–2126.[Abstract/Free Full Text]

Evans, J. L. 2001. Prediction of Mature Weight and Maintenance Energy in Cattle. Ph.D. Diss., Colorado State Univ., Fort Collins.

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, United Kingdom.

Howery, L. D., F. D. Provenza, R. E. Banner, and C. B. Scott. 1998. Social and environmental factors influence cattle distribution on rangelands. Appl. Anim. Behav. Sci. 55:231–244.[CrossRef]

Jenkins, T. G., and C. L. Ferrell. 1994. Productivity through weaning of nine breeds of cattle under varying feed availabilities: I. Initial evaluation. J. Anim. Sci. 72:2787–2797.[Abstract]

Koots, K. R., J. P. Gibson, C. Smith, and J. W. Wilton. 1994a. Analyses of published genetic parameter estimates for beef production traits. 1. Heritability. Anim. Breed. Abstr. 62:310–338.

Koots, K. R., J. P. Gibson, and J. W. Wilton. 1994b. Analyses of published genetic parameter estimates for beef production traits. 2. Phenotypic and genetic correlations. Anim. Breed. Abstr. 62:825–853.

Phillips, C. J. C., and M. I. Rind. 2002. The effects of social dominance on the production and behavior of grazing dairy cows offered forage supplements. J. Dairy Sci. 85:51–59.[Abstract]

Robinson, D. L. 1996a. Estimation and interpretation of direct and maternal genetic parameters for weights of Australian Angus cattle. Livest. Prod. Sci. 45:1–11.

Robinson, D. L. 1996b. Models which might explain negative correlations between direct and maternal genetic effects. Livest. Prod. Sci. 45:111–122.[CrossRef]

Snelling, W. M., M. D. MacNeil, D. D. Kress, D. C. Anderson, and M. W. Tess. 1996. Factors influencing genetic evaluations of linebred Hereford cattle in diverse environments. J. Anim. Sci. 74:1499–1510.[Abstract]

Winder, J. A., J. S. Brinks, R. M. Bourdon, and B. L. Golden. 1990. Genetic analyses of absolute growth measurements, relative growth rate and restricted selection indices in Red Angus cattle. J. Anim. Sci. 68:330–336.[Abstract]

Winder, J. A., D. A. Walker, and C. C. Bailey. 1995. Genetic aspects of diet selection in the Chihuahuan desert. J. Range Manage. 48:549–553.

Winder, J. A., D. A. Walker, and C. C. Bailey. 1996. Effect of breed on botanical composition of cattle diets on Chihuahuan desert range. J. Range Manage. 49:209–214.

This article has been cited by other articles:

				K. L. Shirley, D. W. Beckman, and D. J. Garrick Inheritance of pulmonary arterial pressure in Angus cattle and its correlation with growth J Anim Sci, April 1, 2008; 86(4): 815 - 819. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: jas.2006-093v1 85/3/610    most recent Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Speidel, S. E. Articles by Garrick, D. J. Search for Related Content PubMed PubMed Citation Articles by Speidel, S. E. Articles by Garrick, D. J.