Genetic evaluation of the probability of pregnancy at
14 months for Nellore heifers1
J. P. Eler*2, J. A. II V. Silva*, J. B. S. Ferraz*, F. Dias†, H. N. Oliveira‡,
J. L. Evans§, and B. L. Golden¶
*Universidade de São Paulo, Cx. P. 23, 13.630 970, Pirassununga, SP, Brazil;
†Agropecuária CFM, Ltda., Av. Feliciano Sales Cunha, 1330, 15035-900, S.J. do Rio Preto, Brazil;
‡Universidade Estadual Paulista, Cx. P. 560, 18618-970, Botucatu, Brazil; §Oklahoma State University,
Stillwater 74078; and ¶Colorado State University, Fort Collins 80523
heritability, obtained by Method ℜ, was 0.57 with standard error of 0.01. The EPD was predicted using a maximum a posteriori threshold method and was expressed
as deviations from 50% probability. The range in EPD
was −24.50 to 24.55%, with a mean of 0.78% and a
SD of 7.46%. We conclude that EPD for probability of
pregnancy can be used to select heifers with a higher
probability of being fertile. However, it is mainly recommended for the selection of bulls for the production of
precocious daughters because the accuracy of prediction
is higher for bulls, depending on their number of
daughters.
ABSTRACT: To estimate the heritability for the
probability that yearling heifers would become pregnant, we analyzed the records of 11,487 Nellore animals
that participated in breeding seasons at three farms in
the Brazilian states of São Paulo and Mato Grosso do
Sul. All heifers were exposed to a bull at the age of
about 14 mo. The probability of pregnancy was analyzed
as a categorical trait, with a value of 1 (success) assigned to heifers that were diagnosed pregnant by rectal
palpation about 60 d after the end of the breeding season of 90 d and a value of 0 (failure) assigned to those
that were not pregnant at that time. The estimate of
Key Words: Beef Cattle, Genetic Parameters, Heifers, Pregnancy
2002 American Society of Animal Science. All rights reserved.
Introduction
J. Anim. Sci. 2002. 80:951–954
defined as the probability that a heifer will be pregnant
after the end of the breeding season when she is expose
to a bull or inseminated. It is a binary trait, with a
value of 1 for pregnant heifers and a value of 0 for
nonpregnant heifers.
Few estimates of heritability (h2) for probability of
pregnancy have been published, especially with respect
to the concept established here. Koots et al. (1994),
Doyle et al. (1996), and Snelling et al. (1996) reported h2
values of 0.05 ± 0.01; 0.21 ± 0.11, and 0.30, respectively.
Evans et al. (1999), in a single-trait analysis, obtained
a value of 0.14 ± 0.09 and, in a two-trait analysis with
scrotal circumference, obtained a value of 0.24 ± 0.12
for the Hereford breed. Doyle et al. (2000) reported an
h2 of 0.21 ± 0.01 for the Angus breed. Higher h2 values
recently reported for probability of pregnancy may be
attributed to the analytical procedures adopted, which
may be more appropriate for the analysis of categorical
data (Snelling et al., 1995).
The objective of this study was to analyze the pregnancy records for Nellore heifers exposed to a bull at
14 mo of age and to evaluate the viability of using
probability of pregnancy at 14 mo as a selection criterion.
In beef cattle, observed fertility is an inherent trait
(i.e., the underlying genetic potential that expresses
the endocrine and physiological functions of a heifer)
that cannot be fully defined by phenotypic measurements made directly on the animals. Thus, many studies have been conducted to find reliable indicators of
fertility, and a wide gamut of measurement methods
has been used, such as scrotal circumference in male
relatives, age at first calving, and calving interval.
In order to find a trait indicative of sexual precocity
that might be included in the objectives of selection,
recent research has evaluated heifer pregnancy (Doyle
et al., 1996; Evans et al., 1999), which is designated in
the present study as the probability of pregnancy and is
1
The authors wish to acknowledge financial support from CAPES,
CNPq, FAPESP, and Agro-Pecuária CFM, Ltda.
2
Correspondence and present address: Faculdade de Zootecnia e
Engenharia de Alimentos/USP-CP 23. CEP 13630-970 Pirassununga
(SP), Brazil (E-mail: [email protected]).
Received January 23, 2001.
Accepted November 9, 2001.
951
952
Eler et al.
Material and Methods
We analyzed the pregnancy records for Nellore heifers exposed to bulls at about 14 mo of age during the
breeding seasons of 1995 to 2001 in three herds. The
data came from Agro-Pecuária CFM, Ltda., and the
herds were located northeast of São Paulo and in Mato
Grosso do Sul. Agro-Pecuária CFM was a purebred Nellore operation consisting of approximately 17,000 cows
and selling an average of 2,000 young bulls per year
out of about 7,000 males weaned.
The heifers were placed randomly in lots with a group
of bulls or in some cases in lots with a single bull, for
a breeding season of 90 d. About 60 d after the end of the
breeding season, the heifers were evaluated by rectal
palpation for the diagnosis of pregnancy. Heifers with
a positive diagnosis (pregnant) were scored as 1 and
heifers with a negative diagnosis were scored as 0.
The records of the seven breeding seasons cited above
were used to prepare a data file with 12,575 pregnancy
records divided into 274 contemporary groups. Of these
records, 2,284 (18.2%) were scored 1 and 10,291 (81.8%)
were scored 0. The contemporary groups consisted of
the contemporary heifer group plus the reproduction
lot. The latter was determined on the basis of the herd
plus the service sire(s). The contemporary heifer group
consisted of herd and year of birth + weaning management group + postweaning management group.
The data set analyzed after eliminating of contemporary groups with no variation, consisted of the pregnancy records for 11,487 heifers born from 1993 to 1999
and divided into 234 contemporary groups with 2,266
records (19.8%) being scored 1 and 9,221 (80.2%) being
scored 0. Of the 11,487 heifers studied, all of them with
a known dam, 8,957 were daughters of a known sire
and 2,530 were produced from multiple-sire mating and
were considered to be from an unknown sire. The data
set included 210 sires and 8.564 dams. The pedigree
data included all animals with an observation plus their
parents and grandparents a total of 26,030 animals,
including 561 sires and 15,926 dams.
The mathematical model included fixed contemporary group effects (234 contemporary groups) and the
effects of dam’s age class at calving (seven classes: 1,
up to 27 mo; 2, 28 to 36; 3, 37 to 48; 4, 49 to 72; 5, 73
to 120; 6, 121 to 144; and 7 older than 144 mo).
In a preliminary analysis, the model included the
effect of heifer’s age at the beginning of the breeding
season as a covariate. The regression (0,01) was not
significant and the covariate was excluded from the
final analysis. The mean age at the beginning of the
breeding season was 410 d, ranging from 365 to 532 d.
Only 231 heifers enter the breeding season with more
than 500 d and for the last three breeding seasons, most
the heifers were exposed at about the same age of 12 mo.
The random effects included were animal additive
genetic effect and residual effect. Variance components
were estimated by the method ℜ (Reverter et al., 1994),
and genetic values were predicted using an a posteriori
Table 1. Dam’s age class at calving (DAC), number of
observation (N) and solution for 14-mo pregnancy
probability, expressed on percent probability
DAC
1
2
3
4
5
6
7
N
Solutiona
258
1,970
2,024
2,812
3,024
747
652
36.4
4.9
1.2
0.5
0.0
1.5
2.7
a
Compared with DAC 5.
maximum likelihood threshold model (MAP; Gianola
and Foulley, 1983; Harville and Mee, 1984) on an underlying genetic scale. The ABTK2.0, Animal Breeder’s
ToolKit, software (Golden et al., 1992) was used both
for the estimate of variance components and for the
solution of mixed models.
Results and Discussion
The solutions for fixed effects are presented in Table
1. It is interesting to note that the heifers produced
by dams that became pregnant at 14 mo were 36.4%
superior to the daughters of mature cows, whereas the
heifers produced by dams 28 to 36 mo of age at calving
were 4.9% superior. In principle, this superiority cannot
be explained by nongenetic dam effects on daughter
pregnancy rate. There is no reason for females calving
at a young age to have an effect of such magnitude. In
reality, confounding of genetic and nongenetic effects
must be occurring in these classes of dam’s age at calving. The records analyzed involve only the last seven
seasons. Most of the cows that calved when still young
had not had a second parturition or parturition during
a mature age. Since precocious mothers only have the
daughters of the first parturition, the effect of age and
the genetic effect would be confounded. The heifers that
became pregnant at 14 mo received 10 kg of good-quality silage and a small mineral supplementation during
the last 60 d of pregnancy, whereas those that became
pregnant at the “normal” age of 26 mo or more were
left exclusively on pasture until the time for parturition.
This, however, would not explain such a significant
difference, also because daughters of young cows are
lighter at weaning. For the other classes, the differences
were not important.
The heritability estimate for probability of pregnancy
at 14 mo was 0.57 ± 0.01 and was obtained from 133
repeated random subsamples of 50% of the data (Figure
1). If we take the median as a criterion for heritability
estimation, the value would be 0.56. The 95% confidence
interval obtained for h2 by Box-Cox transformation was
0.41 to 0.74.
The heritability estimate from this analysis was
higher than those published thus far in the literature
953
Probability of Heifer Pregnancy
Table 2. Number of animals (N), mean, standard
deviation (SD), and minimum and maximum EPD
for heifer probability of pregnancy at 14 mo
according to animal’s year of birth
Year of birth
Figure 1. Heritability estimates for probability of pregnancy at 14 mo for Nellore heifers.
and may result from the adoption of appropriate analytical procedures for categorical data and from a higher
genetic variability of probability of pregnancy at 14 mo
for Nellore cows. The explanation for this difference
probably is that all the results reported in the literature
were obtained for populations of Bos taurus origin, for
which reproduction at 14 mo is considered normal. The
Nellore breed (Bos indicus) has not yet been selected
for precocity, and, therefore, its genetic variability appears to be much higher.
Analysis of the records shows, for example, a bull
siring daughters with 0% pregnancy and a bull siring
daughters with 55% pregnancy. This observation of
pregnancy refers to approximately 100 daughters of
each bull. The heritability obtained in the present study
shows that probability of pregnancy at 14 mo is a trait
highly heritable for the Nellore breed.
Based on the genetic parameters obtained in the present study, we predicted the EPD for probability of pregnancy at 14 mo of age. Solutions were obtained for all
animals in the pedigree file, and EPD were computed
by transforming MAP solutions for deviations from 50%
probability according to the formula EPDi = [Φ (MAPi
× 0.5) − 0.5] × 100 (Snelling et al., 1995), where EPDi
= expected progeny difference for the ith animal on the
probability scale, Φ = the standard accumulated distribution function, and MAPi = the solution for the ith
animal in the underlying scale.
Thus, the EPD for probability of pregnancy at 14 mo
are presented as the probability of an animal siring
daughters that will become pregnant when exposed at
14 mo of age during a given breeding season. It is deviate from 50% probability. On this basis, if the overall
heifer pregnancy rate for the data is 50%, then an EPD
equal to 20.0 for a bull, for example, means that this
bull has a 20-percentage point higher probability of
siring daughters that will calve at 2 yr of age than a
bull with a 0.0 EPD. In this study, the overall pregnancy
rate is about 20.0%, so that a bull with 20% of his
daughters getting pregnant has a 0.0 EPD and a bull
with a +20.0 EPD is expected to have more than 40%
probability of siring daughters that will calve at 2 yr
of age, which means more than 20 percentage points
higher than a bull with a 0.0 EPD.
up to 89
90
91
92
93
94
95
96
97
98
99
Total
N
Mean
SD
Minimum
Maximum
8,900
1,203
1,174
977
1,802
1,557
1,811
1,907
1,074
2,384
3,241
−0.17
0.08
−0.55
−0.43
−0.12
1.32
1.69
1.51
1.26
1.79
2.88
3.82
5.27
5.62
5.98
6.52
8.46
8.28
8.36
10.65
10.49
10.39
−24.50
−23.18
−19.66
−23.05
−20.23
−20.00
−21.39
−23.63
−19.62
−21.46
−21.31
20.60
22.81
20.98
20.17
24.02
23.78
24.13
23.73
23.49
24.55
23.86
26,030
0.78
7.46
−24.50
24.55
The mean, minimum, and maximum EPD per year
of animal’s birth are presented in Table 2. Mean EPD
and SD was 0.78 and 7.46%, with the range of −24.50
to 24.55%. For this population, 153 sires had 10 or more
progeny, and for those sires the mean and SD were,
respectively, 0.58 and 12.54, with the range from −24.50
to 23.50. This shows that probability of pregnancy at
14 mo has sufficient genetic variation to be included as
a selection criterion for the Nellore breed.
Figure 2 shows the genetic trend of the trait from
1989 to 1999. It should be pointed out, however, that
only in 1995 did the heifers start to be exposed to bulls
at 14 mo (heifers born in 1993), and thus the EPD
for the animals born before 1993 were based only on
pedigree information, except for bulls that could have
the EPD based on progeny. Some cows born before 1993
could also have progeny with records. The use of probability of pregnancy at 14 mo EPD as a selection criterion
started just in the year 2000, but there is a positive
trend (0.60%/year) probably due to exposure of all heifers and selection of those calving at 2 yr old. We expect
that this trend will increase with the use of probability
Figure 2. Genetic trend of probability of pregnancy at
14 mo in Nellore heifers during the period from 1989 to
1999. Mean trend = 0.60%/year.
954
Eler et al.
of pregnancy at 14 mo as a selection criterion in the
next breeding seasons.
Implications
The results of this study are encouraging for the use
of an expected progeny difference for probability of pregnancy at 14 mo in the Nellore breed. Besides being
highly heritable, 14-mo pregnancy data are simple, inexpensive, and practical to record. An expected progeny
difference for probability of pregnancy at 14 mo would
provide breeders useful information for improving female sexual precocity.
Literature Cited
Doyle, S. P., B. L. Golden, R. D. Green, and J. S. Brinks. 2000.
Additive genetic parameter estimates for heifer pregnancy and
subsequent reproduction in Angus females. J. Anim. Sci.
78:2091–2098.
Doyle, S. P., R. D. Green, B. L. Golden, G. L. Mathiews, C. R. Comstock, and D. G. LeFever. 1996. Genetic parameter estimates
for heifer pregnancy rate and subsequent rebreeding rate in
Angus cattle. J. Anim. Sci. 74(Suppl.1):117 (Abstr.).
Evans, J. L., B. L. Golden, R. M. Bourdon, and K. L. Long. 1999.
Additive genetic relationship between heifer pregnancy and scrotal circumference in Hereford cattle. J. Anim. Sci. 77:2621–2628.
Gianola, D., and J. L. Foulley. 1983. Sire evaluation for ordered
categorical data with a threshold model. Genet. Sel. Evol.
15:201–224.
Golden, B. L., W. M. Snelling, and C. H. Mallinckrodt. 1992. Animal
breeder’s tool kit user’s guide and reference manual. Colorado
State Univ. Agric. Exp. Sta., Ft. Collins. Tech. Bull., LTB92-2.
Harville, D. A., and R. W. Mee. 1984. A mixed-model procedure for
analyzing ordered categorical data. Biometrics 40:393–408.
Koots, K. R., J. P. Gibson, C. Smith, and J. W. Wilton. 1994. Analyses
of published genetic parameter estimates for beef production
traits. 1. Heritability. Anim. Breed. Abstr. 62:309–338.
Reverter, A., B. L. Golden, R. M. Bourdon, and J. S. Brinks. 1994.
Method R variance components procedure: Application of the
simple breeding value model. J. Anim. Sci. 72:2247–2253.
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.
Snelling, W. M., M. D. MacNeil, and B. L. Golden. 1996. Application
of continuous and binary trait methods to reproductive measures
of Hereford cattle. J. Anim. Sci. 74(Suppl.1):115 (Abstr.).