Lucia Galvão Albuquerque

Reaction norms for the study of genotype-environment interaction for growth and indicator traits of sexual precocity in Nellore cattle

The objective of this study was to quantify the magnitude of genotype-environment interaction (GxE) effects on age at first calving (AFC), scrotal circumference (SC), and yearling weight (YW) in Nellore cattle using reaction norms. For the study, 89,152 weight records of female and male Nellore animals obtained at yearling age were used. Genetic parameters were estimated with a single-trait random-regression model using Legendre polynomials as base functions. The heritability estimates were of low to medium magnitude for AFC (0.05 to 0.47) and of medium to high magnitude for SC (0.32 to 0.51) and YW (0.13 to 0.72), and increased as the environmental gradient became more favorable. The genetic correlation estimates ranged from 0.25 to 1.0 for AFC, from 0.71 to 1.0 for SC, and from 0.42 to 1.0 for YW. High Spearman correlation coefficients were obtained for the three traits, ranging from 0.97 to 0.99. The reaction norms along the environmental gradient of 10 sires each with the highest or lowest breeding value for YW predicted by single-trait analysis demonstrated more plastic phenotypes for YW and more robust phenotypes for SC. The effect of GxE was most important for YW and AFC with respect to SC. When animals are selected for higher SC or YW or lower AFC, considering or not the GxE effect, it is expected that the same animals will be selected. The reaction norms obtained based on sire breeding values along the environmental gradient showed that animals with extreme breeding values respond differently as environmental conditions improve.

Study of lipid metabolism-related genes as candidate genes of sexual precocity in Nellore cattle.

The objective of this study was to identify associations between known polymorphisms in genes related to adipose tissue and sexual precocity in Nellore cattle. A total of 1689 precocious and non-precocious heifers belonging to farms participating in Conexão Delta G breeding program were studied. SNPs from the Illumina High-Density Bovine SNP BeadChip were used. This chip contains 777,000 SNPs located within the region of the candidate genes at a distance of up to 5 kb, considering that linkage disequilibrium (LD) exists at this distance. Linear models were used for statistical analysis. The fastPHASE and GenomeStudio programs were used for haplotype reconstruction and LD analysis based on r2 statistics. Fifty-seven candidate genes and 443 SNPs were analyzed: among the latter, 370 SNPs formed 83 haplotypes, while the remaining SNPs were studied separately. Statistical analysis showed that only three haplotypes, one haplotype consisting of two SNPs located in the FABP4 gene and two haplotypes consisting of four and two SNPs located in the PPP3CA gene, had a significant effect on sexual precocity at P < 0.05. It can be concluded that the FABP4 and PPP3CA genes influence sexual precocity and may therefore be used in selection programs designed to improve sexual precocity in Nellore cattle.

Multivariate analysis of test-day and total milk yield in goats

The objective of this study was to estimate genetic parameters for 305-day cumulative milk yield (MY305) and its association with test-day milk yield (TDMY) in Saanen and Alpine goats in order to provide information that allows the use of TDMY as selection criteria. This was done using standard multi-trait and reduced rank models. Data from 1157 lactations, including the first three kiddings, and 5435 test-day records from 683 Saanen and 449 Alpine goats were used. MY305 was analyzed together with TDMY by multi-trait analysis, from the first to tenth test-day, using records of the first three lactations as repeated measures. Three multi-trait models were used: a standard (SM) and two reduced rank models that fitted the first two (PC2) and three (PC3) genetic principal components. Akaike and Schwarz Bayesian information criteria were used to compare models. Heritability for TDMY estimated with the SM ranged from 0.20 to 0.66, whereas the range calculated from the PC2 model was 0.16 to 0.63. Genetic correlations between TDMY and MY305 were positive and moderate to high, ranging from 0.56 to 0.98 when estimated with the SM, and 0.91 to 1.00 when estimated with the PC2. The standard multi-trait model produced estimates that were more accurate than the reduced rank models. Although the SM provided the worst fit according to the two model selection criteria, it was the best in this dataset.