Department of Animal Science, National Chung Hsing University, Taichung 402, Taiwan; Ray Hsing Agricultural Biotechnology Co. Ltd., Yunlin 633, Taiwan.
Department of Animal Science, National Chung Hsing University, Taichung 402, Taiwan.
Poult Sci. 2024 Oct;103(10):104063. doi: 10.1016/j.psj.2024.104063. Epub 2024 Jul 6.
In local chickens targeted for niche markets, genotyping costs are relatively high due to the small population size and diverse breeding goals. The single-step genomic best linear unbiased prediction (ssGBLUP) model, which combines pedigree and genomic information, has been introduced to increase the accuracy of genomic estimated breeding value (GEBV). Therefore, this model may be more beneficial than the genomic BLUP (GBLUP) model for genomic selection in local chickens. Additionally, the single-step genome-wide association study (ssGWAS) can be used to extend the ssGBLUP model results to animals with available phenotypic information but without genotypic data. In this study, we compared the accuracy of (G)EBVs using the pedigree-based BLUP (PBLUP), GBLUP, and ssGBLUP models. Moreover, we conducted single-SNP GWAS (SNP-GWAS), GBLUP-GWAS, and ssGWAS methods to identify genes associated with egg production traits in the NCHU-G101 chicken to understand the feasibility of using genomic selection in a small population. The average prediction accuracy of (G)EBV for egg production traits using the PBLUP, GBLUP, and ssGBLUP models is 0.536, 0.531, and 0.555, respectively. In total, 22 suggestive- and 5% Bonferroni genome-wide significant-level SNPs for total egg number (EN), average laying rate (LR), average clutch length, and total clutch number are detected using 3 GWAS methods. These SNPs are mapped onto Gallus gallus chromosomes (GGA) 4, 6, 10, 18, and 25 in NCHU-G101 chicken. Furthermore, through SNP-GWAS and ssGWAS methods, we identify 2 genes on GGA4 associated with EN and LR: ENSGALG00000023172 and PPARGC1A. In conclusion, the ssGBLUP model demonstrates superior prediction accuracy, performing on average 3.41% than the PBLUP model. The implications of our gene results may guide future selection strategies for Taiwan Country chickens. Our results highlight the applicability of the ssGBLUP model for egg production traits selection in a small population, specifically NCHU-G101 chicken in Taiwan.
在针对利基市场的本地鸡中,由于种群规模较小和多样化的繁殖目标,基因分型成本相对较高。单步基因组最佳线性无偏预测(ssGBLUP)模型结合了系谱和基因组信息,已被引入以提高基因组估计育种值(GEBV)的准确性。因此,与基因组 BLUP(GBLUP)模型相比,该模型可能更有利于本地鸡的基因组选择。此外,单步全基因组关联研究(ssGWAS)可用于将 ssGBLUP 模型结果扩展到具有可用表型信息但无基因型数据的动物。在这项研究中,我们比较了基于系谱的 BLUP(PBLUP)、GBLUP 和 ssGBLUP 模型使用(G)EBV 的准确性。此外,我们进行了单 SNP 全基因组关联研究(SNP-GWAS)、GBLUP-GWAS 和 ssGWAS 方法,以鉴定与 NCHU-G101 鸡产蛋性状相关的基因,以了解在小种群中使用基因组选择的可行性。使用 PBLUP、GBLUP 和 ssGBLUP 模型对产蛋性状的(G)EBV 的平均预测准确性分别为 0.536、0.531 和 0.555。总共使用 3 种 GWAS 方法检测到总蛋数(EN)、平均产蛋率(LR)、平均产蛋间隔和总产蛋数的 22 个提示性和 5%Bonferroni 全基因组显著水平 SNP。这些 SNP 映射到 NCHU-G101 鸡的鸡染色体(GGA)4、6、10、18 和 25 上。此外,通过 SNP-GWAS 和 ssGWAS 方法,我们在 GGA4 上鉴定出与 EN 和 LR 相关的 2 个基因:ENSGALG00000023172 和 PPARGC1A。总之,ssGBLUP 模型的预测准确性平均比 PBLUP 模型高 3.41%。我们的基因结果的意义可能指导未来对台湾土鸡的选择策略。我们的结果强调了 ssGBLUP 模型在小种群(特别是台湾 NCHU-G101 鸡)产蛋性状选择中的适用性。