F. Fonseca e Silva

Seleção e classificação multivariada de modelos de crescimento não lineares para bovinos Nelore

This study aimed to evaluate cluster analysis in classifying and selecting non linear models to describe Nelore beef cattle growth based on different goodness of fit criteria tests. A total of 12 non linear models were evaluated based on the following criteria: the determination coefficient (R2), error mean square (QME), Akaike information criterion (AIC), Bayesian information criterion (BIC), mean quadratic error of prediction (MEP) and predicted determination coefficient (R2p). The Brody model showed the best adjustment for the data set.

Follicular dynamics and gene expression in granulosa cells, corpora lutea and oocytes from gilts of breeds with low and high ovulation rates.

Follicular dynamics and the expression of candidate genes using real-time polymerase chain reaction (PCR) were compared during the oestrous cycle of pig breeds with high (commercial line; n=24) and low (local Brazilian Piau; n=21) ovulation rates and prolificacy. Gilts were killed on Days 0, 4, 10 and 18 of the oestrous cycle and visible ovarian follicles were classified by follicular diameter. Recovered cumulus-oocyte complexes were classified as normal or atretic and frozen in liquid nitrogen until RNA extraction. Low ovulation rates and/or prolificacy in Piau gilts was associated with a different pattern of follicle development, with lower numbers of small follicles on Day 18, fewer large follicles on Days 0 and 18 (P≤0.05) and a higher proportion of atretic follicles on Days 0 and 18 (P≤0.05). Compared with commercial line gilts, less-prolific Piau gilts exhibited higher expression of apoptotic genes during luteolysis (CASP3 and FASL; P≤0.05), decreased expression of TGFBR2 and BAX mRNA in the corpus luteum (P≤0.05), higher expression of apoptotic genes (FAS, BCL2 and CASP8; P≤0.05) in granulosa cells and a greater abundance (P≤0.05) of genes controlling oocyte-secreted factors (GDF9, BMP15 and BMP6), suggesting underlying mechanisms controlling differences in follicular development, ovulation rate and inherent prolificacy in this pig breed.

Modelo hierárquico bayesiano aplicado na avaliação genética de curvas de crescimento de bovinos de corte

The estimation of the (co)variance components for the parameters of the growth models can be evaluated by many methods. The Bayesian approach is an alternative method of the estimation. A study was performed using simulated and real data from Nelore cattle for estimation of the (co)variance components for the parameters of Von Bertalanffy growth curve, using a bayesian hierarchical model. From the estimated components, the heritabilities for each parameter and genetic and environmental correlations between these parameters were determined. The samples of posterior marginal distributions for the parameters a, R, μ , u, G, and σ2e were obtained by using Gibbs Sampler algorithm and for the parameters b e k by using the Metropolis-Hastings algorithm. The efficiency of the bayesian inference methodology was verified since estimated parameters were quite close to the simulated ones. The parameters a and k from real data showed heritabilities compatible with the reality indicating they could be used in selection programs.

Caracterização de colágenos tipos I e III no estroma do carcinoma de células escamosas cutâneo em cães

O carcinoma de celulas escamosas (CCE) e uma neoplasia epitelial maligna que acomete caes e diversas outras especies, incluindo a humana. O CCE afeta varios sitios anatomicos e pode desenvolver metastase. O objetivo deste estudo foi a caracterizacao das fibras de colagenos tipos I e III no estroma do CCE cutâneo de caes. Para este trabalho, utilizaram-se 44 amostras de pele incluidas em parafina e que tiveram previo diagnostico de CCE. As amostras foram processadas histologicamente e coradas com hematoxilina/eosina para confirmacao do diagnostico e classificacao do grau de diferenciacao tumoral e com a coloracao histoquimica de picrosirius para observacao dos colagenos tipos I e III. O colageno tipo III mostrou maior expressao nos CCEs cutâneos bem diferenciados. O papel do colageno do tipo III nas neoplasias nao esta bem esclarecido, e outros fatores alem do grau de diferenciacao celular podem estar envolvidos em sua expressao e determinar sua importância na biologia tumoral.

Modelos lineares generalizados mistos na avaliação genética da prenhez precoce na raça Nelore

This study aimed to use the generalized linear models with probit and logit link function to evaluate early pregnancy, and to observe the effects on genetic variability and on sire selection when different ages are adopted in the definition of this trait. Early pregnancy was studied at 15 (EP15), and 21 (EP21) months. The analysis was done in R software. Pearson correlations (PC), between genetic predicted values and percentage of bulls in common considering only 10% of bulls with higher genetic values (TOP 10), between classification by logit and probit models and in each model among EP15 and EP21, were calculated. The heritability for EP15 and EP21 were close between models, except for EP15 using probit link function. PC and TOP10 among models were high. The Akaike and Bayesian criteria reported was similar between models. TOP10, considering the same model, among EP15-EP21 were moderated between EP15-EP21.

Parâmetros e ganhos genéticos em características de crescimento de bovinos Tabapuã da Bahia

We compared two models (with or without maternal effect) in the estimation of genetic parameters through Bayes Factor (BF) and the Deviance Information Criterion (DIC). Additionally, we evaluated the genetic, maternal and phenotypic trends in Tabapua bovine growth characteristics in the state of Bahia. The model that included the maternal effect provided smaller BF values (167,629.2; 117,341.2 and 124,804.8) and DIC (174,550.0; 120,242.7 and 128,037.2) for weights at 205 (W205), 365 (W365) and 550 (W550) days of age, respectively. The average estimates, a posteriori, the direct and maternal heritability (0.33, 0.43 and 0.44) and (0.15, 0.14 and 0.16) for the three characteristics, respectively. Genetic trends for direct effect were 0.4415 and 0.3613kg/year for W205 and W365 and presented increases of 0.25 and 0.15% in the average characteristics of the year. The phenotypic trend for W205 was 0.7039kg/year. Maternal genetic trends for the three characteristics showed losses and indicate absence of selection matrices for good maternal ability. Despite the magnitude of the estimated direct and maternal heritability, they indicate genetic gain opportunities, genetic and phenotypic trends indicated few direct gains and no gains for maternal ability.

Seleção genômica ampla para curvas de crescimento

A methodology was proposed for the genetic evaluation of growth curves considering SNP (Single Nucleotide Polymorphisms) markers. At the first step, nonlinear regression growth models (Logistic) were fitted to the weight-age of each animal, and on second step the parameter estimates of the Logistic model were used as phenotype in a regression model (Bayesian LASSO - BL) which covariates were given by SNP genotypes. This approach allows the estimation of GBV (Genomic Breeding Values) for weight at either time of growth trajectory, allowing also the production of genomic growth curves, which selected groups of individuals with larger growth efficiency. The simulated data set was constituted of 2,000 individuals (being 1,000 in the training and 1,000 in the validation population) each one with 453 SNP markers distributed along 5 chromosomes. The results indicated high efficiency of the BL method to predict GBV in the validation population using information from the training population (correlation coefficients varying between 0.79 and 0.93). The BL also presented high efficiency to detect QTL, once the most expressive estimated SNP effects were located at positions closed to true QTL position fixed in the simulation.

Proposta de análise genética de curvas de crescimento de bovinos por meio do algoritmo SAEM

Two methodologies in genetic evaluation of growth curves of Nellore cattle were compared: the SAEM algorithm and the Two Step method. To implement these methodologies the Brody modified growth curve and the sire model were used. The difference between the SAEM and the Two Step is that SAEM estimates jointly the parameters of the model and genetics and environmental effects and the Two Step method does this process in two independent steps. Estimates of the fixed effects and genetics parameters, and prediction breeding values for the sires were obtained from the methodologies. From the breeding values genetic curves were obtained for the sires. The SAEM algorithm proved consistent in the estimation of fixed effects and prediction of random effects.

Efeito da autocorrelação residual na avaliação genética de cabras para a produção de leite e para o formato da curva de lactação

The objective of this study was to evaluate the residual autocorrelation effect on the estimates of genetic parameters for total milk yield (MY) and the coefficients a, b and c of Woods model and, consequently, the animals classification for these traits. Woods model was fitted to the lactation of goats considering three cases of residual structure: (IE) - independent errors, (AR1) - first order autoregressive errors, and (IE - AR1) - AR1 errors only for lactations that had significant residual autocorrelation and IE for others. Estimates of the coefficients a, b and c and PL were used as dependent variables in a multivariate animal model, which included the random effects of animal and permanent environmental and fixed effects of contemporary group, number of foals per birth, genetic groups and order of calving, besides the covariates of lactation length (linear) and age at parturition (linear and quadratic). The different residual structures affected the estimates of genetic parameters and the classification of animals for the traits studied. Therefore, the selection of animals can be modified with the presence of residual autocorrelation and the structure IE-AR1 is presented as an alternative to solve this problem.