Evaluation of the productive and reproductive performance of sexual precocity at different ages in Nellore heifers.

This study aimed to investigate the differences between productive and reproductive performance traits of sexually precocious and non-sexually precocious Nellore heifers and to evaluate the genetic correlation of sexual precocity with traits of economic importance. For this purpose, 300,000 Nellore heifers were evaluated for reproductive traits: heifer pregnancy (HP) at 14 (HP14), 18 (HP18), and 24 (HP24) months; heifer rebreeding (HR); number of progenies up to 53 months (NP53); and probability of the cow remaining in the herd until 76 months with at least 3 progenies (Stay). The growth-related traits evaluated included female yearling weight (YW); average daily gain from weaning to yearling (ADGW-Y); weight at maturity (MW); weaning weight of first progeny (WWprog); and female visual scores at yearling for conformation (Conf), precocity (Prec) and muscling (Musc). The effects of female YW and ADGW-Y in six categories on HP14, HP18, and HP24 were analyzed using Generalized linear mixed models (GLMM). Furthermore, a linear mixed model was used to evaluate the impact of HP on WWprog, MW, and reproductive performance (NP53 and Stay). Genetic correlations of HP evaluated in different months with growth and reproductive traits were estimated using a bivariate animal model. Precocious heifers (HP14) were lighter for YW and MW but had greater ADGW-Y than HP18 and HP24. The probability for HP14, HP18, and HP24 increased as the classes of YW and ADGW-Y increased. However, heifers weighing more than 326 kg had a slight reduction in the probability of becoming pregnant at HP14 and HP18. Precocious heifers (HP14 and HP18) produced their first progeny by 3 % lighter than HP24, although they had a greater NP53. Precocious heifers at 18 months (HP18) were 3 % and 6.8 % more likely to remain in the herd than HP14 and HP24 heifers, respectively. Genetic correlations between growth traits (WW, YW, ADGW-Y, and MW) and heifer pregnancy (HP14, HP18, and HP24) ranged from weak (rg = 0.27 ± 0.05) to moderate (rg = -0.47 ± 0.07). The genetic correlation between HR and HP was stronger for HP24 (0.75) against HP14 (0.58) and HP18 (0.64). Although, the genetic correlation between NP53 and Stay with HP14 was higher (rg = 0.53 and 0.45) than those observed for HP18 (rg = 0.46 and 0.38) and HP24 (rg = 0.35 and 0.39). The genetic correlation estimates between HP and visual scores were moderate and favorable for HP14. Selecting HP14 is beneficial for production systems because it increases the NP53 during the productive life without compromising heifer productivity or reproductive performance. However, attention should be given to improving the HR of heifers who become pregnant early.

Genome-wide association study considering genotype-by-environment interaction for productive and reproductive traits using whole-genome sequencing in Nellore cattle.

BACKGROUND: The genotype-by-environment interaction (GxE) in beef cattle can be investigated using reaction norm models to assess environmental sensitivity and, combined with genome-wide association studies (GWAS), to map genomic regions related to animal adaptation. Including genetic markers from whole-genome sequencing in reaction norm (RN) models allows us to identify high-resolution candidate genes across environmental gradients through GWAS. Hence, we performed a GWAS via the RN approach using whole-genome sequencing data, focusing on mapping candidate genes associated with the expression of reproductive and growth traits in Nellore cattle. For this purpose, we used phenotypic data for age at first calving (AFC), scrotal circumference (SC), post-weaning weight gain (PWG), and yearling weight (YW). A total of 20,000 males and 7,159 females genotyped with 770k were imputed to the whole sequence (29 M). After quality control and linkage disequilibrium (LD) pruning, there remained ∼ 2.41 M SNPs for SC, PWG, and YW and ∼ 5.06 M SNPs for AFC. RESULTS: Significant SNPs were identified on Bos taurus autosomes (BTA) 10, 11, 14, 18, 19, 20, 21, 24, 25 and 27 for AFC and on BTA 4, 5 and 8 for SC. For growth traits, significant SNP markers were identified on BTA 3, 5 and 20 for YW and PWG. A total of 56 positional candidate genes were identified for AFC, 9 for SC, 3 for PWG, and 24 for YW. The significant SNPs detected for the reaction norm coefficients in Nellore cattle were found to be associated with growth, adaptative, and reproductive traits. These candidate genes are involved in biological mechanisms related to lipid metabolism, immune response, mitogen-activated protein kinase (MAPK) signaling pathway, and energy and phosphate metabolism. CONCLUSIONS: GWAS results highlighted differences in the physiological processes linked to lipid metabolism, immune response, MAPK signaling pathway, and energy and phosphate metabolism, providing insights into how different environmental conditions interact with specific genes affecting animal adaptation, productivity, and reproductive performance. The shared genomic regions between the intercept and slope are directly implicated in the regulation of growth and reproductive traits in Nellore cattle raised under different environmental conditions.

Benchmarking machine learning and parametric methods for genomic prediction of feed efficiency-related traits in Nellore cattle.

Genomic selection (GS) offers a promising opportunity for selecting more efficient animals to use consumed energy for maintenance and growth functions, impacting profitability and environmental sustainability. Here, we compared the prediction accuracy of multi-layer neural network (MLNN) and support vector regression (SVR) against single-trait (STGBLUP), multi-trait genomic best linear unbiased prediction (MTGBLUP), and Bayesian regression (BayesA, BayesB, BayesC, BRR, and BLasso) for feed efficiency (FE) traits. FE-related traits were measured in 1156 Nellore cattle from an experimental breeding program genotyped for ~ 300 K markers after quality control. Prediction accuracy (Acc) was evaluated using a forward validation splitting the dataset based on birth year, considering the phenotypes adjusted for the fixed effects and covariates as pseudo-phenotypes. The MLNN and SVR approaches were trained by randomly splitting the training population into fivefold to select the best hyperparameters. The results show that the machine learning methods (MLNN and SVR) and MTGBLUP outperformed STGBLUP and the Bayesian regression approaches, increasing the Acc by approximately 8.9%, 14.6%, and 13.7% using MLNN, SVR, and MTGBLUP, respectively. Acc for SVR and MTGBLUP were slightly different, ranging from 0.62 to 0.69 and 0.62 to 0.68, respectively, with empirically unbiased for both models (0.97 and 1.09). Our results indicated that SVR and MTGBLUBP approaches were more accurate in predicting FE-related traits than Bayesian regression and STGBLUP and seemed competitive for GS of complex phenotypes with various degrees of inheritance.