Multi-omics analysis

Variant selection strategies to enhance genomic prediction of enteric methane emission and feed efficiency-related traits in Nellore cattle

---

Abstract

Genomic selection (GS) provides an opportunity to identify more feed-efficient animals, reflecting better profitability and sustainability in beef cattle farming. However, limited phenotyping capacity and a high number of genetic markers, such as sequence markers, represent a challenge, as they can yield to statistical difficulties and impact the accuracy of estimates. Thus, pre-selection of variants at the sequence level based on their functional effect may provide a strategy to improve GS accuracy using commercial SNP panels. In this study, the predictive capacity of multivariate ssGBLUP models for enteric methane emission (ME: methane emission-g/day; and RME: residual methane production-g/day) and feed efficiency-related traits (RFI: residual feed intake-kg/day; and DMI: dry matter intake-kg/day) were evaluated in Nellore cattle using different SNP variant selection criteria. For this, imputed whole-genome sequence genotypes were used to select markers via either GWAS or Fst index analyses. Phenotypic records from 2,516 animals born from 2004 to 2022 were used, and 4,293 animals genotyped with the Illumina Bovine HD-BeadChip were imputed to sequence genotypes to perform the GWAS and Fst analyses. A total of 997 (GWAS) and 4,721 (Fst) SNPs identified as significantly associated (not present in the HD panel) were included in the HD panel, resulting, after quality control, in two customized panels containing 374,948 (HD-GWAS) and 378,783 (HD-Fst) SNPs. Genomic breeding values (GEBVs) were obtained using BLUPF90+ software and were assessed by forward validation, with animals divided by year of birth (training: 2004-2020, N=2,177; and validation: 2021-2022, N=339). Accuracy of early prediction of young animals' GEBVs was evaluated as the correlation between GEBVs in the training and in the validation sets, while dispersion of GEBVs was assessed by the linear regression of GEBVs in the training on the GEBVs in the validation set. The prediction accuracies with the HD-Fst panel were 0.82, 0.72, 0.72, and 0.40 for DMI, RFI, ME, and RME, while the corresponding dispersions were 0.87, 1.08, 0.92, and 0.92, respectively. For the HD-GWAS panel, the prediction accuracies were 0.82, 0.71, 0.75, and 0.42 for DMI, RFI, ME, and RME, with corresponding dispersions of 0.87, 1.04, 0.86, and 0.83, respectively. In general, similar accuracies of prediction of GEBVs of young animals were obtained with the two strategies to customize the HD panel. In both cases, the predictions were less accurate for the residual traits (RFI/RME) compared to the original traits (DMI/ME), especially for RME. Regarding dispersion, GEBVs of young animals were over-dispersed for all traits and the two customized panels, except for RFI, for which an under-dispersion was observed. Next, the impact of incorporating the identified SNPs into medium-density SNP panels will be investigated, aiming to support the development of selection strategies that improve the efficiency and sustainability of Nellore cattle production.

Keywords: 2026

How to Cite:

Arikawa, L., Schenkel, F., Mota, L., Fonseca, L., Fernandes Júnior, G., Mercadante, M. & Albuquerque, L., (2026) “Variant selection strategies to enhance genomic prediction of enteric methane emission and feed efficiency-related traits in Nellore cattle”, World Congress on Genetics Applied to Livestock Production Digital Archive 2026(1): 2286999. doi: https://doi.org/10.31274/wcgalp.24192

Rights: 1