trabalho completo - 52ª Reunião Anual da Sociedade Brasileira de

52a Reunião Anual da Sociedade Brasileira de
Zootecnia
Zootecnia: Otimizando Recursos e Potencialidades
Belo Horizonte – MG, 19 a 23 de Julho de 2015
Análise de componentes principais dos fatores envolvidos na taxa de prenhez em matrizes de corte
submetidas a IATF1
Maria Dulcinéia da Costa2, Robertha Veloso Rebello3, José Reinaldo Mendes Ruas2, Julieta Maria Alencar
Chamone2, Antônia de Maria Filha Ribeiro2, Anielle Cristina Alves Meneses 3, Wemerson Fábio Gomes Ribas4
1
Parte de da dissertação Mestrado do segundo autor
Programa de Pós-Graduação em Zootecnia – UNIMONTES – Janaúba, Brasil, Bolsista FAPEMIG e:mail:[email protected]
Mestranda Programa Pós-graduação em Zootecnia - UNIMONTES, BRA. Bolsista do CAPES.
4
Curso de Zootecnia - UNIMONTES, BRA. Bolsista do FAPEMIG / PIBIC.
2
3
Resumo: Foi objetivo com o trabalho avaliar a importância relativa das características que afetam a taxa de prenhez
em vacas submetidas a IATF no Norte de Minas Gerais utilizando análise de componentes principais. Vários são os
efeitos que pode afetar a taxa de prenhez e faz-se necessário conhecer a importância e contribuição de cada uma
destas variáveis. A técnica de componentes principais é utilizada para descrever os fatores responsáveis por
diferenças entre indivíduos, de acordo com um conjunto de respostas correlacionadas observadas numa população.
Foram analisadas 8683 informações na estação de monta de dezembro de 2012 a maio de 2013. Foram avaliados os
efeitos da categoria reprodutiva, escore corporal, touro, raça touro, horário da inseminação, estímulos ovulatórios,
dia da aplicação de prostaglandina, utilização do CIDR, inseminador, estruturas de ovário e dose de ECG sobre a
presença ou ausência de prenhez em animais submetidos à inseminação artificial em tempo fixo (IATF). A taxa
média de prenhez foi de 52,79%. Verificou-se que cinco componentes principais responderam por 65,01 % da
variância total. Os autores concluíram que dose de ECG, uso do CIDR, estímulos ovulatórios, raça do touro e
inseminador são responsáveis pela maior variação na taxa de prenhez em vacas submetidas a IATF e as variáveis
categorias reprodutivas, escore corporal, estrutura do ovário, dia da aplicação da protaglandina, horário da IATF
contribuem muito pouco para a variabilidade da taxa de prenhez.
Palavras–chave: biotecnologias reprodutivas, gado de corte, índice reprodutivo, reprodução
Principal component analysis for factors involved in the pregnancy rate in beef cows subjected to IATF
Abstract: The objective of this study was to evaluate the relative importance of characteristics affecting the
pregnancy rate in cows submitted to TAI in the North of Minas Gerais using principal component analysis. There
are many effects that can affect the pregnancy rate and it is necessary to know the importance and contribution of
each of these variables. The technique of principal components is used to describe the factors responsible for the
differences between individuals, according to a set of correlated responses observed in a population. We analyzed
8683 data in the breeding station from December 2012 to May 2013. The effects of reproductive category, body
score, sire, sire breed, the insemination time, ovulatory stimulus, on the application of prostaglandin, using CIDR,
inseminator, ovarian structures and ECG dose on the presence or absence of pregnancy in animals subjected to
fixed time artificial insemination (TAI). The average pregnancy rate was 52.79%. It was found that five principal
components accounted for 65.01% of the total variance. The authors concluded that ECG dose, CIDR use,
ovulatory stimulus, bull race and inseminator are responsible for the variation in pregnancy rate in cows submitted
to IATF and the variables reproductive categories, body score, ovarian structure, application of day the
prostaglandin, the IATF time contribute very little to the variability of pregnancy rate.
Keywords:, beef cattle, reproduction, reproductive rate, reproductive biotechnologies
Introduction
The pregnancy rate is one of the largest impact variables in assessing the reproductive performance and
contributes considerably to bioeconomic efficiency. There are several factors that influence pregnancy rate in
animals as efficiency in estrus detection, insemination time, semen handling, inseminator skill. It is necessary to
know the importance and contribution of each of these variables that influence the pregnancy rate (BARUSELLI,
2012, MENEGHETTI et al, 2009). When evaluating many variables it is possible that some of these parameters are
redundant (JOLLIFFE, 1973). The technique of principal components is used to describe the factors responsible for
the differences between individuals, according to a set of correlated responses observed in a population. The
method consists in the transformation of the original variables into new variables, the principal components. The
principal components are orthogonal, so that the information represented by them are uncorrelated. Thus, this work
________________________________________________________________________________________________________________________________________________
Página - 1 - de 3
52a Reunião Anual da Sociedade Brasileira de
Zootecnia
Zootecnia: Otimizando Recursos e Potencialidades
Belo Horizonte – MG, 19 a 23 de Julho de 2015
aimed to verify the relative importance of the factors that influence pregnancy rate in cows submitted to IATF using
the technique of principal components.
Material e Methods
We analyzed information from the 8683 livestock file assigned for the provision of veterinary services
company of 32 farms in 18 municipalities in the north of Minas Gerais in the breeding station from December 2012
to May 2013. The effects of parity cow, body condition score, bull breed bull time of insemination, reproductive
categories ovulatory stimulus, on the application of prostaglandin re-use the CIDR inseminator, ovarian structures
and dose of ECG on the presence or absence of pregnancy in animals submitted to artificial insemination at fixed
time (TAI). Reproductive categories of matrices were nulliparous (heifers), primiparous (one calving), multiparous
(more than one calving) and single cows (without suckling calves). The body score was based on the subjective
analysis of the nutritional status of animals by means of visual assessment of veterinarians who were beginning the
protocol depending on the muscle coverage and fat mass, on a scale from 1.0 to 5.0 (1 = very Thin, 5 = very fat)
according Houghton et al. (1990), adapted to intervals of 0.25. The structures assessed by ovarian ultrasound were
classified according to the size of the follicle P: 4 to 5 mm, L: 5 to 8 mm, G +, 8 mm in diameter and up CL:
Presence of corpus luteum in the ovary. The time of insemination was in the morning and in the afternoon. The
reuse of CIDR possible to use 1, 2, and 3 to 4 times. Ovulatory stimuli were calf removal (shang) application of
equine chorionic gonadotropin and the implementation of both at once 3 (RB + ECG). We used the Chi-square test
at 5% significance (P <0.05) by the PROC FREQ procedure and for the principal component analysis, the
PRINCOMP the Statistical Analysis System (SAS, 2000).
Results and Discussion
The average pregnancy rate was 52.79%. Significant differences in calving order, body condition score, time
insemination, bull, bull breed, CIDR reuse, ovulatory stimulus, ovarian structures, ECG dosage, day of applying
lutalize and inseminators. The principal component analysis explains the structure of variance and covariance of the
original variables, building a smaller set of linear combinations (principal components) that preserve most of the
information provided by the variables in question. There are five principal components accounted for 65.01% of the
total variance (Table 1).
Table 1. Principal components (PC), eigenvalues (λ), percentage of the total variance explained by
the principal components (VPC) and accumulated variance (AV).
Principal components
λ
VPC
AV
CP1
1.98398165
0.1804
0.1804
CP2
1.77301318
0.1612
0.3415
CP3
1.26521954
0.1150
0.4566
CP4
1.08478187
0.0986
0.5552
CP5
1.04383288
0.0949
0.6501
CP6
0.95420195
0.0867
0.7368
CP7
0.78322490
0.0712
0.8080
CP8
0.65832382
0.0598
0.8679
CP9
0.62031451
0.0564
0.9243
CP10
0.50634099
0.0460
0.9703
CP11
0.32676470
0.0297
1.0000
Considering the five principal components (Table 2) accounted for 65% of the variation in pregnancy
rate in cows submitted to TAI, the original variables dose of ECG, CIDR use, ovulatory stimulus,
bull breed and inseminator make this change as the variables reproductive categories, body score,
ovarian structure, day of application of prostaglandin, the IATF time does not make it up.
Table 2. Estimate Eigenvectors obtained from the correlation matrix between the
variables in cows submitted to TAI in the North of Minas Gerais.
Original variables
CP1
CP2
CP3
CP4
Reproductive categories
-.313057 0.388817
-.239453
0.106842
Body Score
0.386179 -.084929
-.385421
-.167425
Ovarian structure
0.411697 -.419349
0.029733
0.184015
pregnancy rate
CP5
-.037137
-.174550
0.038281
________________________________________________________________________________________________________________________________________________
Página - 2 - de 3
52a Reunião Anual da Sociedade Brasileira de
Zootecnia
Zootecnia: Otimizando Recursos e Potencialidades
Belo Horizonte – MG, 19 a 23 de Julho de 2015
Application prostaglandin day
Ovulatory stimulus
ECG Dose
CIDR use
TAI
Sire
Sire breed
inseminator
-.211244
0.315827
0.488729
-.132067
0.194303
0.366405
0.068682
0.077300
0.358207
0.244824
0.373650
-.511237
0.251694
0.044626
0.094929
0.046614
0.351367
0.556392
0.151928
0.278133
-.407646
0.045090
0.105292
-.283660
0.340691
-.058072
-.048923
0.087094
0.047435
0.458399
-.718553
0.261861
0.062938
-.137290
-.132016
-.143340
-.359055
0.350730
0.507111
0.629929
Conclusions
The variables dose of ECG, Ovarian structure, CIDR use, ovulatory stimulus, bull breed and inseminator
are responsible for the variation in pregnancy rate in cows submitted to IATF;
The variables reproductive categories, body score, ovarian structure, application day of prostaglandin, the
TAI contribute very little to the variability of pregnancy rate.
Acknowledgements
The authors thank to FAPEMIG for financial support.
References
BARUSELLI, P. S.; SALES, J. N. S.; SALA, R. V. et al. History, evolution and perspectives of timed artificial
insemination programs in Brazil. Animal Reproduction, v. 9, n. 3, p. 139-152, 2012.
JOLLIFFE, I.T. Discarding variables in a principal component analysis. II: Real data. Appl. Stat., v.22, p.21-31,
1973.
MENEGHETTI, M.; SÁ FILHO, O. J.; PERES, R. et al. Fixed-time artificial insemination with estradiol and
progesterone for Bos indicus cows I: Basis for development of protocols. Theriogenology, v. 72, n. 2, p. 179–189,
2009.
VASCONCELOS, J. L. M.; VILELA, E. R.; SÁ FILHO, O. G. Remoção temporária de bezerros em dois momentos
do protocolo de sincronização da ovulação GnRH-PGF2α-BE em vacas Nelore pós-parto Remoção temporária de
bezerros em dois momentos do protocolo de sincronização da ovulação GnRH-PGF2α-BE em vacas Nelore pósparto. Arquivo Brasileiro de Medicina Veterinária e Zootecnia, v. 61, p. 95-103, 2009.
SAS Statistical analysis system SAS/STAT users guide release 9.1. ed. Cary, NC, SAS Institute Inc. 2002–2008.
________________________________________________________________________________________________________________________________________________
Página - 3 - de 3