Genomic selection in Nellore cattle: new approaches for reproductive traits and polled character

• Ano de defesa: 2025
• Autor(a) principal: Temp, Larissa Bordin [UNESP]
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Instituição de defesa: Universidade Estadual Paulista (Unesp)
• Palavras-chave em Português: Bos indicus; Pedigree; Metafounders; Bovinos de corte
• Link de acesso: https://hdl.handle.net/11449/259850; http://lattes.cnpq.br/3041618469950729; https://orcid.org/0000-0003-1003-7977

Genetic improvement programs focused on the Nellore breed are essential for promoting productivity gains and ensuring the sustainability of the livestock sector. Using data from the National Association of Breeders and Researchers (ANCP), this study had two main objectives. The first was to estimate genetic parameters and assess the influence of the phenotypic classification of horn development, animal sex, and the inclusion of non-autosomal SNP markers on the genomic prediction of the polled trait in Nellore cattle, using the single-step genomic best linear unbiased prediction (ssGBLUP) method. Two, three, and four phenotypic categories for horn development were analyzed, in a total of 12 statistical models. Heritability estimates ranged from 0.44 to 0.83. The inclusion of non-autosomal SNPs in models with four phenotypic categories resulted in a 5.26% improvement in prediction accuracy, a 37% reduction in bias, and a 4.55% improvement in dispersion compared to models that only considered autosomal SNPs. These results suggest that the use of genomic information in selection for the polled trait is a feasible strategy, representing a non-invasive and low-cost approach to increasing the frequency of polled animals. The second objective of the study was to evaluate the use of metafounders (MF) and unknown parent groups (UPGs) in the genomic evaluation of traits related to sexual precocity, fertility, longevity, and productivity. Records for scrotal circumference at 365 days (SC365), age at first calving (AFC), and accumulated cow productivity (ACP) were used. Four models were implemented, taking into account UPGs and MF based on different criteria: commercial and registered herds, uncertain paternity, and patriarchs (animals born in Brazil from imported sires). Heritability estimates ranged from 0.07 to 0.40. The accuracy and bias of predictions were assessed using the linear regression method. Although the models with MF showed small improvements in accuracy, the results suggest that increasing the number of genotypes and phenotypes available may help improve MF estimates and, consequently, predictive accuracy.