Accuracy of genomic prediction using different models and response variables in the Nordic Red cattle population

Breeding animals can be accurately evaluated using appropriate genomic prediction models, based on marker data and phenotype information. In this study, direct genomic values (DGV) were estimated for 16 traits of Nordic Total Merit (NTM) Index in Nordic Red cattle population using three models and two different response variables. The three models were as follows: a linear mixed model (GBLUP), a Bayesian variable selection model similar to BayesA (BayesA*) and a Bayesian least absolute shrinkage and selection operator model (Bayesian Lasso). The response variables were deregressed proofs (DRP) and conventional estimated breeding values (EBV). The reliability of genomic predictions was measured on bulls in the validation data set as the squared correlation between DGV and DRP divided by the reliability of DRP. Using DRP as response variable, the reliabilities of DGV among the 16 traits ranged from 0.151 to 0.569 (average 0.317) for GBLUP, from 0.152 to 0.576 (average 0.318) for BayesA* and from 0.150 to 0.570 (average 0.320) for Bayesian Lasso. Using EBV as response variable, the reliabilities ranged from 0.159 to 0.580 (average 0.322) for GBLUP, from 0.157 to 0.578 (average 0.319) for BayesA* and from 0.159 to 0.582 (average 0.325) for Bayesian Lasso. In summary, Bayesian Lasso performed slightly better than the other two models, and EBV performed slightly better than DRP as response variable, with regard to prediction reliability of DGV. However, these differences were not statistically significant. Moreover, using EBV as response variable would result in problems with the scale of the resulting DGV and potential problem due to double counting.     

Accuracies of estimated breeding values from ordinary genetic evaluations do not reflect the correlation between true and estimated breeding values in selected populations

The accuracy of estimated breeding values (EBVs) is an important parameter in livestock genetic improvement. It is used to calculate response to selection and to express the credibility of individual EBVs. Although it is well-known that selection reduces accuracy, this effect is not well-studied and accuracies from genetic evaluations are not adjusted for selection. This work investigates the effect of selection on accuracy of EBVs estimated using best linear unbiased predictors. Results show that accuracies in a selected population may be considerably smaller than the ordinary accuracy from genetic evaluation. Accuracy of the parent average is dramatically reduced by selection, up to a factor of three. Expressions for equilibrium accuracies in selected populations are presented and depend only on the unselected accuracy and the intensity of selection. Thus, schemes with the same unselected accuracy and intensity of selection also have the same equilibrium accuracy and response to selection. At the same unselected accuracy, therefore, schemes based on between-family information do not show greater reduction in response and accuracy because of the Bulmer effect. An example shows that benefit of genomic selection may be underestimated when the effect of selection on accuracy is ignored. Finally, this work argues that the SE of an EBV and the correlation between true and EBVs are different things, and that accuracies should not be adjusted for selection when they primarily serve to indicate the SEs of EBVs.     

Comparison of genomic and traditional BLUP-estimated breeding value accuracy and selection response under alternative trait and genomic parameters

Accuracy of prediction of estimated breeding values based on genome-wide markers (GEBV) and selection based on GEBV as compared with traditional Best Linear Unbiased Prediction (BLUP) was examined for a number of alternatives, including low heritability, number of generations of training, marker density, initial distributions, and effective population size (Ne). Results show that the more the generations of data in which both genotypes and phenotypes were collected, termed training generations (TG), the better the accuracy and persistency of accuracy based on GEBV. GEBV excelled for traits of low heritability regardless of initial equilibrium conditions, as opposed to traditional marker-assisted selection, which is not useful for traits of low heritability. Effective population size is critical for populations starting in Hardy-Weinberg equilibrium but not for populations started from mutation-drift equilibrium. In comparison with traditional BLUP, GEBV can exceed the accuracy of BLUP provided enough TG are included. Unfortunately selection rapidly reduces the accuracy of GEBV. In all cases examined, classic BLUP selection exceeds what was possible for GEBV selection. Even still, GEBV could have an advantage over traditional BLUP in cases such as sex-limited traits, traits that are expensive to measure, or can only be measured on relatives. A combined approach, utilizing a mixed model with a second random effect to account for quantitative trait loci in linkage equilibrium (the polygenic effect) was suggested as a way to capitalize on both methodologies.     

The effects of birth weight and estimated breeding value for protein deposition on nitrogen efficiency in growing pigs

The effects of birth weight (BiW; low BiW [LBW] vs. high BiW [HBW]) and estimated breeding value (EBV) for protein deposition (low EBV [LBV] vs. high EBV [HBV]) on N retention, N efficiency, and concentrations of metabolites in plasma and urine related to N efficiency in growing pigs were studied. At an age of 14 wk, 10 LBW–LBV (BiW: 1.07 ± 0.09 [SD] kg; EBV: −2.52 ± 3.97 g/d, compared with an average crossbred pig with a protein deposition of 165 g/d), 10 LBW–HBV (BiW: 1.02 ± 0.13 kg; EBV: 10.47 ± 4.26 g/d), 10 HBW–LBV (BiW: 1.80 ± 0.13 kg; EBV: −2.15 ± 2.28 g/d), and 10 HBW–HBV (BiW: 1.80 ± 0.15 kg; EBV: 11.18 ± 3.68 g/d) male growing pigs were allotted to the experiment. The pigs were individually housed in metabolism cages and were subjected to an N balance study in two sequential periods of 5 d, after an 11-d dietary adaptation period. Pigs were assigned to a protein adequate (A) or protein restricted (R, 70% of A) regime in a change-over design. Pigs were fed 2.8 times the energy requirements for maintenance. Nontargeted metabolomics analyses were performed in urine and blood plasma samples. The N retention (in g/d) was higher in the HBW than in the LBW pigs (P < 0.001). The N retention (in g/[kg metabolic body weight (BW^0.75) · d]) and N efficiency, however, were not affected by the BiW of the pigs. The N retention (P = 0.04) and N efficiency (P = 0.04) were higher in HBV than in LVB pigs on the A regime but were not affected by EBV in pigs on the R regime. Restricting the dietary protein supply with 30% decreased the N retention (P < 0.001) but increased the N efficiency (P = 0.003). Nontargeted metabolomics showed that a hexose, free amino acids (AA), and lysophosphatidylcholines were the most important metabolites in plasma for the discrimination between HBV and LBV pigs, whereas metabolites of microbial origin contributed to the discrimination between HBV and LBV pigs in urine. This study shows that BiW does not affect N efficiency in the later life of pigs. Nitrogen efficiency and N retention were higher in HBV than in LBV pigs on the A regime but similar in HBV and LBV pigs on the R regime. In precision feeding concepts aiming to further optimize protein and AA efficiency in pigs, the variation in EBV for protein deposition of pigs should be considered as a factor determining N retention, growth performance, and N efficiency.     

基于简化基因组测序技术和基因芯片技术比较研究黄羽肉鸡基因组选择

旨在比较简化基因组测序技术和基因芯片技术实施基因组选择的基因组估计育种值（GEBV）准确性。本研究在AH肉鸡资源群体F₂代中随机选取395个个体（其中公鸡212只,母鸡183只,来自8个半同胞家系）,同时采用10×SLAF测序技术和Illumina Chicken 60K SNP芯片进行基因标记分型。采用基因组最佳无偏估计法（GBLUP）和BayesCπ对6周体重、12周体重、日均增重、日均采食量、饲料转化率和剩余采食量等6个性状进行GEBV准确性比较研究,并采用5折交叉验证法验证。结果表明,采用同一基因标记分型平台,两种育种值估计方法所得GEBV准确性差异不显著（P>0.05）;不同的性状对基因标记分型平台的选择存在差异,对于6周体重,使用基因芯片可获得更高的GEBV准确性（P<0.05）,对于剩余采食量,则使用简化基因组测序可获得更高的GEBV准确性（P<0.05）。综合6个性状GEBV均值比较,两个基因标记分型平台之间差异不到0.01,高通量测序技术和基因芯片技术都可以用于黄羽肉鸡基因组选择。     

Comparison between Reduced-Representation Genome Sequencing and SNP Chip for Genomic Selection in Yellow-feathered Broiler

This study aimed to compare the accuracy of the genomic estimated breeding value (GEBV) using reduced-representation genome sequencing technology and SNP chip technology to implement genomic selection. A total of 395 individuals (212♂+ 183♀, from 8 half-sib families) were randomly selected from F₂ generation of AH broiler resource population, and genotyped with 10×specific-locus amplified fragment sequencing (SLAF-seq) and Illumina Chicken 60K SNP BeadChip. Genomic best linear unbiased prediction (GBLUP) and BayesCπ were used to compare the accuracy of genomic estimated breeding values (GEBV) for 6 traits: body weight at the 6^th week, body weight at the 12^th week, average daily gain (ADG), average daily feed intake (ADFI), feed conversion ratio (FCR) and residual feed intake (RFI). A 5-fold cross validation procedure was used to verify the accuracies of GEBV between prediction models and between genotyping platforms. The results showed that there was no significant difference between accuracies of GEBV predicted by GBLUP and BayesCπ using the same genotyping platform(P>0.05). The superiority of the two genotyping platforms was different for different traits. For body weight at the 6^th week, the accuracy of GEBV was higher using chip SNPs (P<0.05). On the contrary, the accuracy was higher using SLAF-seq for residual feed intake (P<0.05). Comprehensive comparison of the means of GEBV for 6 traits, the difference between the two genotyping platforms was less than 0.01, therefore, both high throughput sequencing and chip SNPs can be used for genomic selection in yellow-feathered broiler.     

A robust method for simultaneous estimation of single gene and polygenic effects in dairy cows using externally estimated breeding values as prior information

The aim of this study was to develop a robust method to estimate single gene and random polygenic animal effects simultaneously in a small field dataset with limited pedigree information. The new method was based on a Bayesian approach using additional prior information on the distribution of externally estimated breeding values. The field dataset consisted of 40 269 test‐day records for milk performance traits for 1455 genotyped dairy cows for the 11 bp‐deletion in the coding sequence of the myostatin gene. For all traits, estimated additive effects of the favoured wild‐type allele (‘+’ allele) were smaller when applying the new method in comparison with the application of a conventional mixed inheritance test‐day model. Dominance effects of the myostatin gene showed the same behaviour but were generally lower than additive effects. Robustness of methods was tested using a data‐splitting technique, based on the correlation of estimated breeding values from two samples, with one‐half of the data eliminated randomly from the first sample and the remaining data eliminated from the second sample. Results for 100 replicates showed that the correlation between split datasets when prior information included was higher than the conventional method. The new method led to more robust estimations for genetic effects and therefore has potential for use when only a small number of genotyped animals with field data and limited pedigree information are available.