当前位置: X-MOL 学术Genet. Sel. Evol. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Bias and accuracy of dairy sheep evaluations using BLUP and SSGBLUP with metafounders and unknown parent groups.
Genetics Selection Evolution ( IF 3.6 ) Pub Date : 2020-08-12 , DOI: 10.1186/s12711-020-00567-1
Fernando L Macedo 1, 2 , Ole F Christensen 3 , Jean-Michel Astruc 4 , Ignacio Aguilar 5 , Yutaka Masuda 6 , Andrés Legarra 1
Affiliation  

Bias has been reported in genetic or genomic evaluations of several species. Common biases are systematic differences between averages of estimated and true breeding values, and their over- or under-dispersion. In addition, comparing accuracies of pedigree versus genomic predictions is a difficult task. This work proposes to analyse biases and accuracies in the genetic evaluation of milk yield in Manech Tête Rousse dairy sheep, over several years, by testing five models and using the estimators of the linear regression method. We tested models with and without genomic information [best linear unbiased prediction (BLUP) and single-step genomic BLUP (SSGBLUP)] and using three strategies to handle missing pedigree [unknown parent groups (UPG), UPG with QP transformation in the $${\mathbf{H}}$$ matrix (EUPG) and metafounders (MF)]. We compared estimated breeding values (EBV) of selected rams at birth with the EBV of the same rams obtained each year from the first daughters with phenotypes up to 2017. We compared within and across models. Finally, we compared EBV at birth of the rams with and without genomic information. Within models, bias and over-dispersion were small (bias: 0.20 to 0.40 genetic standard deviations; slope of the dispersion: 0.95 to 0.99) except for model SSGBLUP-EUPG that presented an important over-dispersion (0.87). The estimates of accuracies confirm that the addition of genomic information increases the accuracy of EBV in young rams. The smallest bias was observed with BLUP-MF and SSGBLUP-MF. When we estimated dispersion by comparing a model with no markers to models with markers, SSGBLUP-MF showed a value close to 1, indicating that there was no problem in dispersion, whereas SSGBLUP-EUPG and SSGBLUP-UPG showed a significant under-dispersion. Another important observation was the heterogeneous behaviour of the estimates over time, which suggests that a single check could be insufficient to make a good analysis of genetic/genomic evaluations. The addition of genomic information increases the accuracy of EBV of young rams in Manech Tête Rousse. In this population that has missing pedigrees, the use of UPG and EUPG in SSGBLUP produced bias, whereas MF yielded unbiased estimates, and we recommend its use. We also recommend assessing biases and accuracies using multiple truncation points, since these statistics are subject to random variation across years.

中文翻译:

使用BLUP和SSGBLUP与metafounders和未知的父母群体进行评估的偏爱和准确性。

在几种物种的遗传或基因组评估中已经报道了偏见。常见的偏见是估计育种值和真实育种值的平均值之间的系统差异,以及它们的过度散布或散布不足。此外,比较谱系预测与基因组预测的准确性是一项艰巨的任务。这项工作建议通过测试五个模型并使用线性回归方法的估计数来分析ManechTêteRousse奶牛的产奶量遗传评估中的偏倚和准确性。我们测试了带有和不带有基因组信息的模型[最佳线性无偏预测(BLUP)和单步基因组BLUP(SSGBLUP)],并使用三种策略来处理缺失的谱系[未​​知的父组(UPG),具有QP转换的UPG和$$ {\ mathbf {H}} $$矩阵(EUPG)和元创始人(MF)]。我们将所选公羊的估计繁殖值(EBV)与每年至2017年从表型的第一个女儿每年获得的相同公羊的EBV进行了比较。我们在模型内和模型间进行了比较。最后,我们比较了有和没有基因组信息的公羊出生时的EBV。在模型中,除模型SSGBLUP-EUPG表现出重要的过度分散(0.87)外,偏差和过度分散很小(偏差:遗传标准偏差为0.20至0.40;分散斜率:0.95至0.99)。对准确性的估计证实,基因组信息的增加提高了幼年公羊EBV的准确性。使用BLUP-MF和SSGBLUP-MF观察到最小的偏差。当我们通过比较无标记的模型和有标记的模型来估计离散度时,SSGBLUP-MF的值接近1,表示分散没有问题,而SSGBLUP-EUPG和SSGBLUP-UPG表现出明显的分散不足。另一个重要的观察结果是估计值随时间变化的异质性,这表明单次检查可能不足以对遗传/基因组评估进行良好的分析。基因组信息的增加提高了ManechTêteRousse幼小公羊EBV的准确性。在缺少血统书的人群中,在SSGBLUP中使用UPG和EUPG会产生偏差,而MF会产生无偏估计,我们建议使用它。我们还建议使用多个截断点来评估偏差和准确性,因为这些统计数据可能会随年份而随机变化。另一个重要的观察结果是估计值随时间变化的异质性,这表明单次检查可能不足以对遗传/基因组评估进行良好的分析。基因组信息的增加提高了ManechTêteRousse幼小公羊EBV的准确性。在缺少血统书的人群中,在SSGBLUP中使用UPG和EUPG会产生偏差,而MF会产生无偏估计,我们建议使用它。我们还建议使用多个截断点来评估偏差和准确性,因为这些统计数据可能会随年份而随机变化。另一个重要的观察结果是估计值随时间变化的异质性,这表明单次检查可能不足以对遗传/基因组评估进行良好的分析。基因组信息的增加提高了ManechTêteRousse幼小公羊EBV的准确性。在缺少血统书的人群中,在SSGBLUP中使用UPG和EUPG会产生偏差,而MF会产生无偏估计,我们建议使用它。我们还建议使用多个截断点来评估偏差和准确性,因为这些统计数据可能会随年份而随机变化。基因组信息的增加提高了ManechTêteRousse幼小公羊EBV的准确性。在缺少血统书的人群中,在SSGBLUP中使用UPG和EUPG会产生偏差,而MF会产生无偏估计,我们建议使用它。我们还建议使用多个截断点来评估偏差和准确性,因为这些统计数据可能会随年份而随机变化。基因组信息的增加提高了ManechTêteRousse幼小公羊EBV的准确性。在缺少血统书的人群中,在SSGBLUP中使用UPG和EUPG会产生偏差,而MF会产生无偏估计,我们建议使用它。我们还建议使用多个截断点来评估偏差和准确性,因为这些统计数据可能会随年份而随机变化。
更新日期:2020-08-12
down
wechat
bug