Made available in DSpace on 2014-06-11T19:32:15Z (GMT). No. of bitstreams: 0 Previous issue date: 2012-04-27Bitstream added on 2014-06-13T20:43:05Z : No. of bitstreams: 1 pereira_rj_dr_jabo.pdf: 607358 bytes, checksum: 1336bea4603875e3ea34e119f6764b26 (MD5) / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP) / Com o objetivo de desenvolver um modelo adequado para a avaliação genética da raça Gir Leiteiro (GL) utilizando-se as produções no dia do controle (PDC), foram analisados registros de controle leiteiro de primeiras lactações de vacas da raça GL e cruzadas com Holandês. Primeiramente, as PDC foram analisadas por modelos multicaracterísticas, onde a produção em cada um dos 10 meses da lactação foi considerada como uma característica diferente. Compararam-se sete modelos: multicaracterísticas padrão, três modelos de posto reduzido ajustando os primeiros 2, 3 ou 4 componentes principais genéticos e três modelos considerando-se uma estrutura de análise fatorial para a matriz de covariâncias genéticas, com 2, 3 ou 4 fatores. Entre os modelos comparados, aquele ajustando os dois primeiros componentes principais foi o melhor de acordo com os critérios de comparação, e diminuiu consideravelmente o número de parâmetros estimados e o tempo gasto para a convergência. Entretanto, ainda assim o número de parâmetros a ser estimado permaneceu grande. Em um segundo estudo, foram comparados modelos de regressão aleatória cujos efeitos genético aditivo e de ambiente permanente foram modelados por polinômios ortogonais de Legendre ou funções Splines lineares. A variância residual foi modelada por 1, 5 ou 10 classes. Cinco classes modelaram as mudanças nas variâncias residuais adequadamente e foram utilizadas na comparação dos modelos. O modelo que utilizou polinômios de Legendre de grau 3 para o efeito genético aditivo e 4 para o de ambiente permanente foi o melhor modelo de acordo com os critérios DIC e BIC. Entretanto, a alta correlação de rank (0,998) entre este modelo e aquele aplicando polinômios de Legendre de grau 3 para os efeitos genético aditivo e de ambiente... / In order to develop a suitable model for genetic evaluation of Dairy Gyr breed (DG) using test-day yields (TD), first lactation TD records from purebred DG and crossbred (DG x Holstein) cows were analyzed. First, the TD were analyzed by multivariate models, where the production in each of the 10 months was considered to be a different trait. The following seven models were compared: standard multivariate model, three reduced rank models fitting the first 2, 3 or 4 genetic principal components, and three models considering a 2-, 3- or 4-factor structure for the genetic covariance matrix. Among the models compared, that one fitting the first two principal components was the best according to the model selection criteria, and markedly reduced the number of parameters estimated and the time spent to reach convergence. However, the number of parameters to be estimated remained high. In a second study, random regression models (RRM) in which additive genetic and permanent environmental effects were modeled using orthogonal Legendre polynomials or linear Spline functions, were compared. Residual variances were modeled considering 1, 5, or 10 classes. Five classes fitted the changes in residual variances adequately and were used for model comparison. According to the DIC and BIC criteria, a model using third-degree and fourth-degree Legendre polynomials for additive genetic and permanent environmental effects, respectively, provided the best fit. However, the high rank correlation (0.998) between this model and that applying third-degree Legendre polynomials for additive genetic and permanent environmental effects, indicates that practically the same bulls would be selected by both models. Therefore, the latter model, which is less parameterized, was used in the other studies components of the thesis. In the following... (Complete abstract click electronic access below)

Two experiments were carried to evaluate the effect of different nutritional plans on the DM intake, ingestive and social behavior and health status of dairy cows from different racial compositions during the transition period. In the first trial we evaluated the inclusion or not (CON) of 10 g / day of oregano (OE) extract (Origanum vulgare) or 5 g / day of green tea extract (GT) (Camellia sinensis L.) in the diet provided to 24 Jersey cows with BW of 441 ± 27 kg, 3.5 ± 0.3 points of BCS and with 2.7 ± 1.8 lactations. Cows that received OE tended to consume more than CON after calving. Before calving, cows OE tended to spend less time in decubitus and to ingest the concentrate in less time than GT. After calving, OE caused faster concentrate intake in relation to CON and GT. After calving, the number of total visits to the trough was lower when OE was used in relation to CON and GT, but cows that ingested GT visited the trough with feeding activity more often and tended to interact less in relation to cows that ingested EO. The use of the extracts did not influence the occurrence of diseases and metabolic disorders. In the second trial, 36 Holstein, Gir and Gyrolando F1 (½ Holstein and ½ Gir) nulliparous cows, with BW of 715 ± 64.5 kg, 4.2 ± 0.3 points of BCS and with 4.1 ± 0.4 years of age were distributed (6 cows from each racial composition) in two nutritionl plans with allowance of 1.89% or 1.69% of BW of feed (based on dry meter) during pre calving period. Dry matter intake was higher in pre and post calving for Holstein cows than the others, and it was higher in some days near the calving for 1.69% plan compared with 1.89% plan. Before calving, cows the 1.69% plan remained longer in the trough and tended to spend longer at the trough with feeding activity in relation to 1.89% plan. Holstein cows tended to stay longer in each visit with feeding at the trough and ingested more at each visit than the others before and after calving. The occurrence of diseases and metabolic disorders was not influenced by the nutritional plans but the Holstein cows showed more dystocia and clinical diseases compared to the others. Nutritional plans change the way the animals distributed their feed activities and influenced social behavior. The adoption of different nutritional plans during the period of transition modify aspects of dairy cow’s behavior and intake without affecting the occurrence of diseases.

Two experiments were carried to evaluate the effect of different nutritional plans on the DM intake, ingestive and social behavior and health status of dairy cows from different racial compositions during the transition period. In the first trial we evaluated the inclusion or not (CON) of 10 g / day of oregano (OE) extract (Origanum vulgare) or 5 g / day of green tea extract (GT) (Camellia sinensis L.) in the diet provided to 24 Jersey cows with BW of 441 ± 27 kg, 3.5 ± 0.3 points of BCS and with 2.7 ± 1.8 lactations. Cows that received OE tended to consume more than CON after calving. Before calving, cows OE tended to spend less time in decubitus and to ingest the concentrate in less time than GT. After calving, OE caused faster concentrate intake in relation to CON and GT. After calving, the number of total visits to the trough was lower when OE was used in relation to CON and GT, but cows that ingested GT visited the trough with feeding activity more often and tended to interact less in relation to cows that ingested EO. The use of the extracts did not influence the occurrence of diseases and metabolic disorders. In the second trial, 36 Holstein, Gir and Gyrolando F1 (½ Holstein and ½ Gir) nulliparous cows, with BW of 715 ± 64.5 kg, 4.2 ± 0.3 points of BCS and with 4.1 ± 0.4 years of age were distributed (6 cows from each racial composition) in two nutritionl plans with allowance of 1.89% or 1.69% of BW of feed (based on dry meter) during pre calving period. Dry matter intake was higher in pre and post calving for Holstein cows than the others, and it was higher in some days near the calving for 1.69% plan compared with 1.89% plan. Before calving, cows the 1.69% plan remained longer in the trough and tended to spend longer at the trough with feeding activity in relation to 1.89% plan. Holstein cows tended to stay longer in each visit with feeding at the trough and ingested more at each visit than the others before and after calving. The occurrence of diseases and metabolic disorders was not influenced by the nutritional plans but the Holstein cows showed more dystocia and clinical diseases compared to the others. Nutritional plans change the way the animals distributed their feed activities and influenced social behavior. The adoption of different nutritional plans during the period of transition modify aspects of dairy cow’s behavior and intake without affecting the occurrence of diseases.

Orientador: Lúcia Galvão de Albuquerque / Coorientador: Rui da Silva Verneque / Coorientador: Lenira El Faro Zadra / Banca: Henrique Nunes Oliveira / Banca: Maria Eugênia Zerlotti Mercadante / Banca: Annaiza Braga Bignardi / Banca: Anibal Eugênio Vercesi Filho / Resumo: Com o objetivo de desenvolver um modelo adequado para a avaliação genética da raça Gir Leiteiro (GL) utilizando-se as produções no dia do controle (PDC), foram analisados registros de controle leiteiro de primeiras lactações de vacas da raça GL e cruzadas com Holandês. Primeiramente, as PDC foram analisadas por modelos multicaracterísticas, onde a produção em cada um dos 10 meses da lactação foi considerada como uma característica diferente. Compararam-se sete modelos: multicaracterísticas padrão, três modelos de posto reduzido ajustando os primeiros 2, 3 ou 4 componentes principais genéticos e três modelos considerando-se uma estrutura de análise fatorial para a matriz de covariâncias genéticas, com 2, 3 ou 4 fatores. Entre os modelos comparados, aquele ajustando os dois primeiros componentes principais foi o melhor de acordo com os critérios de comparação, e diminuiu consideravelmente o número de parâmetros estimados e o tempo gasto para a convergência. Entretanto, ainda assim o número de parâmetros a ser estimado permaneceu grande. Em um segundo estudo, foram comparados modelos de regressão aleatória cujos efeitos genético aditivo e de ambiente permanente foram modelados por polinômios ortogonais de Legendre ou funções Splines lineares. A variância residual foi modelada por 1, 5 ou 10 classes. Cinco classes modelaram as mudanças nas variâncias residuais adequadamente e foram utilizadas na comparação dos modelos. O modelo que utilizou polinômios de Legendre de grau 3 para o efeito genético aditivo e 4 para o de ambiente permanente foi o melhor modelo de acordo com os critérios DIC e BIC. Entretanto, a alta correlação de rank (0,998) entre este modelo e aquele aplicando polinômios de Legendre de grau 3 para os efeitos genético aditivo e de ambiente... (Resumo completo, clicar acesso eletrônico abaixo) / Abstract: In order to develop a suitable model for genetic evaluation of Dairy Gyr breed (DG) using test-day yields (TD), first lactation TD records from purebred DG and crossbred (DG x Holstein) cows were analyzed. First, the TD were analyzed by multivariate models, where the production in each of the 10 months was considered to be a different trait. The following seven models were compared: standard multivariate model, three reduced rank models fitting the first 2, 3 or 4 genetic principal components, and three models considering a 2-, 3- or 4-factor structure for the genetic covariance matrix. Among the models compared, that one fitting the first two principal components was the best according to the model selection criteria, and markedly reduced the number of parameters estimated and the time spent to reach convergence. However, the number of parameters to be estimated remained high. In a second study, random regression models (RRM) in which additive genetic and permanent environmental effects were modeled using orthogonal Legendre polynomials or linear Spline functions, were compared. Residual variances were modeled considering 1, 5, or 10 classes. Five classes fitted the changes in residual variances adequately and were used for model comparison. According to the DIC and BIC criteria, a model using third-degree and fourth-degree Legendre polynomials for additive genetic and permanent environmental effects, respectively, provided the best fit. However, the high rank correlation (0.998) between this model and that applying third-degree Legendre polynomials for additive genetic and permanent environmental effects, indicates that practically the same bulls would be selected by both models. Therefore, the latter model, which is less parameterized, was used in the other studies components of the thesis. In the following... (Complete abstract click electronic access below) / Doutor

Two experiments were carried to evaluate the effect of different nutritional plans on the DM intake, ingestive and social behavior and health status of dairy cows from different racial compositions during the transition period. In the first trial we evaluated the inclusion or not (CON) of 10 g / day of oregano (OE) extract (Origanum vulgare) or 5 g / day of green tea extract (GT) (Camellia sinensis L.) in the diet provided to 24 Jersey cows with BW of 441 ± 27 kg, 3.5 ± 0.3 points of BCS and with 2.7 ± 1.8 lactations. Cows that received OE tended to consume more than CON after calving. Before calving, cows OE tended to spend less time in decubitus and to ingest the concentrate in less time than GT. After calving, OE caused faster concentrate intake in relation to CON and GT. After calving, the number of total visits to the trough was lower when OE was used in relation to CON and GT, but cows that ingested GT visited the trough with feeding activity more often and tended to interact less in relation to cows that ingested EO. The use of the extracts did not influence the occurrence of diseases and metabolic disorders. In the second trial, 36 Holstein, Gir and Gyrolando F1 (½ Holstein and ½ Gir) nulliparous cows, with BW of 715 ± 64.5 kg, 4.2 ± 0.3 points of BCS and with 4.1 ± 0.4 years of age were distributed (6 cows from each racial composition) in two nutritionl plans with allowance of 1.89% or 1.69% of BW of feed (based on dry meter) during pre calving period. Dry matter intake was higher in pre and post calving for Holstein cows than the others, and it was higher in some days near the calving for 1.69% plan compared with 1.89% plan. Before calving, cows the 1.69% plan remained longer in the trough and tended to spend longer at the trough with feeding activity in relation to 1.89% plan. Holstein cows tended to stay longer in each visit with feeding at the trough and ingested more at each visit than the others before and after calving. The occurrence of diseases and metabolic disorders was not influenced by the nutritional plans but the Holstein cows showed more dystocia and clinical diseases compared to the others. Nutritional plans change the way the animals distributed their feed activities and influenced social behavior. The adoption of different nutritional plans during the period of transition modify aspects of dairy cow’s behavior and intake without affecting the occurrence of diseases.

Made available in DSpace on 2015-03-26T13:42:09Z (GMT). No. of bitstreams: 1 texto completo.pdf: 205468 bytes, checksum: 88389728ebe8628dfada4221324a292d (MD5) Previous issue date: 2006-02-10 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / The pedigree and production database of Gyr cattle selected for milk production, stored at Embrapa had been employed in this research work. There were studied population s genetic structure, inbreeding effects on productive and reproductive traits and a comparative genetic evaluation regarding a complete and selected genetic basis. The ENDOG program has used 27,610 animals to obtain the individual coefficients of inbreeding (F) and average relatedness (CR), the effective number of animals (Ne), founders (fe) and ancestors (fa) and generation intervals (IG). The average F and AR coefficients of population were 2.82 % and 2.10 %, respectively. The Ne ranged from 16.1 to 125.3 for complete generations. The estimated fe and fa were, respectively, 146 and 75, with only 28 ancestors being the source of 50 % of the populations genes. The total IG was 8.41 years, being higher for males than for females. Data from 24,045 lactations ended between 1960 and 2004, from 8,879 cows distributed in 26 herds, were used for verifying the effects of inbreeding over the productive traits: productions of milk (PL), fat (PGOR), protein (PPRO), lactosis (PLAC), total solids (PSOL) and length of lactation (DLAC); and reproductive traits: age at first calving (ID1P) and calving interval (INTP). The variance and regression analysis used SAS procedures. The variance analysis for productive traits and calving intervals considered as fixed effects the herd, year-season of calving, generation class (CG) nested in herd, inbreeding class (CF) and age of the cow at calving, as covariate, with linear and quadratic terms. For age at first calving there were considered in the model the fixed effects of herd-year of birth, birth season, CF and CG-herd. The productive traits were affected (p<0.05) by all sources of variation. The age of the cow at calving and birth season did not affect (p>0.05) the calving interval and age at first calving, respectively. A regression study of the traits, adjusted for others variation sources, was carried out considering the average inbreeding coefficient of each class, weighted by the number of observations of the class. The linear and quadratic regression terms were significant (p<0.01) for productive traits, in which the higher production levels were associated to inbreeding coefficients between 10 and 12 %. Only the linear regression term was significant (p<0.05) for ID1P. The proposed regression model did not explain INTP. The genetic evaluation for milk production was performed by MTDFREML system, applying the production database plus information of cow s parents and two different pedigree structures. The complete genetic basis (BGC) considered all available information and the selected genetic basis (BGS) used a subset with only pedigree information of progeny test bulls. The employment of BGS, on average, underestimated the breeding values. High Pearson and Spearman correlations between predicted breeding values by the utilization of both genetics basis were obtained. However, the selection of different animals based on of the genetic basis could reduce the genetic gains, mainly in males, which are more influenced by relative s information. / Os arquivos de pedigree e produção da raça Gir selecionada para produção de leite, armazenados na Embrapa Gado de Leite, foram utilizados para análise da estrutura genética da população, estudo do efeito da endogamia sobre características produtivas e reprodutivas, e avaliação genética comparativa entre base genética completa e selecionada. O programa ENDOG utilizou 27.610 animais no cálculo dos coeficientes individuais de endogamia (F) e de relação médio (CR), número efetivo de animais (Ne), de fundadores (fe) e de ancestrais(fa), e do intervalo de gerações (IG). Os coeficientes F e CR médios da população foram 2,82% e 2,10%, respectivamente. O Ne variou de 16,1 a 125,3 entre as gerações completas. O fe estimado foi 146 e o fa foi 75, sendo que apenas 28 ancestrais foram responsáveis pela origem de 50% dos genes da população. O IG total foi de 8,41 anos, sendo maior para machos e menor para fêmeas. Registros de 24.045 lactações ocorridas entre 1960 e 2004, de 8.879 vacas distribuídas em 26 rebanhos, foram utilizados para verificação do efeito da endogamia sobre as características produtivas: produções de leite (PL), de gordura (PGOR), de proteína (PPRO), de lactose (PLAC) e de sólidos totais (PSOL) e duração da lactação (DLAC); e reprodutivas: idade ao primeiro parto (ID1P) e intervalo de partos (INTP). As análises de variância e de regressão utilizaram procedimentos do SAS. A análise de variância para as características produtivas e intervalo de partos considerou os efeitos fixos de rebanho, ano-estação do parto, classe de geração (CG) aninhado em rebanho, classe de endogamia (CF) e idade da vaca ao parto, como covariável, nos termos linear e quadrático). Para idade ao primeiro parto, o modelo incluiu os efeitos fixos de rebanho-ano do nascimento, época de nascimento, CF e CG-rebanho. As características produtivas foram afetadas por todas as fontes de variação (p<0,05). A idade da vaca ao parto e estação de nascimento não afetaram (p>0,05) o intervalo de partos e idade ao primeiro parto, respectivamente. Realizou-se um estudo de regressão das características, ajustadas para as outras fontes de variação, em função do coeficiente médio de endogamia de cada classe, ponderando-se pelo número de observações de cada classe. Os termos linear e quadrático do modelo de regressão foram significativos (p<0,01) para as características produtivas, cujas maiores produções estiveram associadas a coeficientes de endogamia entre 10 e 12%. Somente o termo linear foi significativo (p<0,05) para ID1P. O modelo de regressão proposto não explicou (p>0,05) a característica INTP. As avaliações genéticas para produção de leite foram realizadas pelo sistema MTDFREML e utilizaram como base o arquivo de produção, acrescido de pai e mãe da vaca. Variou-se somente a informação de pedigree, ora utilizando toda informação disponível, ou base genética completa (BGC), ora utilizando apenas as informações de pedigree dos touros submetidos ao teste de progênie, ou base genética selecionada (BGS). A utilização da BGS, em média, subestimou os valores genéticos dos animais. Altas correlações de Pearson e Spearman entre os valores genéticos preditos pela utilização das duas bases foram obtidas. Contudo, a seleção de animais diferentes quando se utiliza uma base genética ou outra poderia incorrer em redução nos ganhos genéticos, sobretudo nos machos, mais influenciados pela informação de parentes.