There is an increasing need to account for genotype-by-environment (G × E) interactions in livestock breeding programs to improve productivity and animal welfare across environmental and management conditions. This is even more relevant for pigs because selection occurs in high-health nucleus farms, while commercial pigs are raised in more challenging environments. In this study, we used single-step homoscedastic and heteroscedastic genomic reaction norm models (RNM) to evaluate G × E interactions in Large White pigs, including 8686 genotyped animals, for reproduction (total number of piglets born, TNB; total number of piglets born alive, NBA; total number of piglets weaned, NW), growth (weaning weight, WW; off-test weight, OW), and body composition (ultrasound muscle depth, MD; ultrasound backfat thickness, BF) traits. Genetic parameter estimation and single-step genome-wide association studies (ssGWAS) were performed for each trait. The average performance of contemporary groups (CG) was estimated and used as environmental gradient in the reaction norm analyses. We found that the need to consider heterogeneous residual variance in RNM models was trait dependent. Based on estimates of variance components of the RNM slope and of genetic correlations across environmental gradients, G × E interactions clearly existed for TNB and NBA, existed for WW but were of smaller magnitude, and were not detected for NW, OW, MD, and BF. Based on estimates of the genetic variance explained by the markers in sliding genomic windows in ssGWAS, several genomic regions were associated with the RNM slope for TNB, NBA, and WW, indicating specific biological mechanisms underlying environmental sensitivity, and dozens of novel candidate genes were identified. Our results also provided strong evidence that the X chromosome contributed to the intercept and slope of RNM for litter size traits in pigs. We provide a comprehensive description of G × E interactions in Large White pigs for economically-relevant traits and identified important genomic regions and candidate genes associated with GxE interactions on several autosomes and the X chromosome. Implementation of these findings will contribute to more accurate genomic estimates of breeding values by considering G × E interactions, in order to genetically improve the environmental robustness of maternal-line pigs.

中文翻译：

---

基于单步基因组反应规范的母系猪繁殖、身体成分和生长性状的基因型-环境相互作用

在牲畜育种计划中，越来越需要考虑基因型与环境 (G × E) 的相互作用，以提高环境和管理条件下的生产力和动物福利。这与猪更相关，因为选择发生在高健康的核心农场，而商品猪则在更具挑战性的环境中饲养。在本研究中，我们使用单步同方差和异方差基因组反应范数模型 (RNM) 来评估大白猪的 G × E 相互作用，包括 8686 只基因型动物，用于繁殖（出生仔猪总数，TNB；仔猪总数出生活体，NBA；断奶仔猪总数，NW），生长（断奶体重，WW；非测试体重，OW）和身体成分（超声肌肉深度，MD；超声背膘厚度，BF）特征。对每个性状进行遗传参数估计和单步全基因组关联研究（ssGWAS）。当代群体 (CG) 的平均表现被估计并用作反应规范分析中的环境梯度。我们发现在 RNM 模型中需要考虑异质残差是依赖于特征的。基于对 RNM 斜率的方差分量和跨环境梯度的遗传相关性的估计，TNB 和 NBA 显然存在 G × E 相互作用，WW 存在但幅度较小，并且未检测到 NW、OW、MD 和BF。根据 ssGWAS 中滑动基因组窗口中的标记解释的遗传方差估计，几个基因组区域与 TNB、NBA 和 WW 的 RNM 斜率相关，表明环境敏感性背后的特定生物学机制，并确定了数十种新的候选基因。我们的结果还提供了强有力的证据，表明 X 染色体有助于 RNM 对猪窝产仔数性状的截距和斜率。我们全面描述了大白猪中​​经济相关性状的 G × E 相互作用，并确定了与几个常染色体和 X 染色体上 GxE 相互作用相关的重要基因组区域和候选基因。通过考虑 G × E 相互作用，这些发现的实施将有助于更准确地对育种值进行基因组估计，从而在遗传上改善母系猪的环境稳健性。我们的结果还提供了强有力的证据，表明 X 染色体有助于 RNM 对猪窝产仔数性状的截距和斜率。我们全面描述了大白猪中​​经济相关性状的 G × E 相互作用，并确定了与几个常染色体和 X 染色体上 GxE 相互作用相关的重要基因组区域和候选基因。通过考虑 G × E 相互作用，这些发现的实施将有助于更准确地对育种值进行基因组估计，从而在遗传上改善母系猪的环境稳健性。我们的结果还提供了强有力的证据，表明 X 染色体有助于 RNM 对猪窝产仔数性状的截距和斜率。我们全面描述了大白猪中​​经济相关性状的 G × E 相互作用，并确定了重要的基因组区域和与几个常染色体和 X 染色体上 GxE 相互作用相关的候选基因。通过考虑 G × E 相互作用，这些发现的实施将有助于更准确地对育种值进行基因组估计，从而在遗传上改善母系猪的环境稳健性。我们全面描述了大白猪中​​经济相关性状的 G × E 相互作用，并确定了重要的基因组区域和与几个常染色体和 X 染色体上 GxE 相互作用相关的候选基因。通过考虑 G × E 相互作用，这些发现的实施将有助于更准确地对育种值进行基因组估计，从而在遗传上改善母系猪的环境稳健性。我们全面描述了大白猪中​​经济相关性状的 G × E 相互作用，并确定了重要的基因组区域和与几个常染色体和 X 染色体上 GxE 相互作用相关的候选基因。通过考虑 G × E 相互作用，这些发现的实施将有助于更准确地对育种值进行基因组估计，从而在遗传上改善母系猪的环境稳健性。