Efficient large-scale single-step evaluations and indirect genomic prediction of genotyped selection candidates.

BACKGROUND: Single-step genomic best linear unbiased prediction (ssGBLUP) models allow the combination of genomic, pedigree, and phenotypic data into a single model, which is computationally challenging for large genotyped populations. In practice, genotypes of animals without their own phenotype and progeny, so-called genotyped selection candidates, can become available after genomic breeding values have been estimated by ssGBLUP. In some breeding programmes, genomic estimated breeding values (GEBV) for these animals should be known shortly after obtaining genotype information but recomputing GEBV using the full ssGBLUP takes too much time. In this study, first we compare two equivalent formulations of ssGBLUP models, i.e. one that is based on the Woodbury matrix identity applied to the inverse of the genomic relationship matrix, and one that is based on marker equations. Second, we present computationally-fast approaches to indirectly compute GEBV for genotyped selection candidates, without the need to do the full ssGBLUP evaluation. RESULTS: The indirect approaches use information from the latest ssGBLUP evaluation and rely on the decomposition of GEBV into its components. The two equivalent ssGBLUP models and indirect approaches were tested on a six-trait calving difficulty model using Irish dairy and beef cattle data that include 2.6 million genotyped animals of which about 500,000 were considered as genotyped selection candidates. When using the same computational approaches, the solving phase of the two equivalent ssGBLUP models showed similar requirements for memory and time per iteration. The computational differences between them were due to the preprocessing phase of the genomic information. Regarding the indirect approaches, compared to GEBV obtained from single-step evaluations including all genotypes, indirect GEBV had correlations higher than 0.99 for all traits while showing little dispersion and level bias. CONCLUSIONS: In conclusion, ssGBLUP predictions for the genotyped selection candidates were accurately approximated using the presented indirect approaches, which are more memory efficient and computationally fast, compared to solving a full ssGBLUP evaluation. Thus, indirect approaches can be used even on a weekly basis to estimate GEBV for newly genotyped animals, while the full single-step evaluation is done only a few times within a year.

A computationally efficient method for approximating reliabilities in large-scale single-step genomic prediction.

BACKGROUND: In this study, computationally efficient methods to approximate the reliabilities of genomic estimated breeding values (GEBV) in a single-step genomic prediction model including a residual polygenic (RPG) effect are described. In order to calculate the reliabilities of the genotyped animals, a single nucleotide polymorphism best linear unbiased prediction (SNPBLUP) or a genomic BLUP (GBLUP), was used, where two alternatives to account for the RPG effect were tested. In the direct approach, the genomic model included the RPG effect, while in the blended method, it did not but an index was used to weight the genomic and pedigree-based BLUP (PBLUP) reliabilities. In order to calculate the single-step GBLUP reliabilities for the breeding values for the non-genotyped animals, a simplified weighted-PBLUP model that included a general mean and additive genetic effects with weights accounting for the non-genomic and genomic information was used. We compared five schemes for the weights. Two datasets, i.e., a small (Data 1) one and a large (Data 2) one were used. RESULTS: For the genotyped animals in Data 1, correlations between approximate reliabilities using the blended method and exact reliabilities ranged from 0.993 to 0.996 across three lactations. The slopes observed by regressing the reliabilities of GEBV from the exact method on those from the blended method were 1.0 for all three lactations. For Data 2, the correlations and slopes ranged, respectively, from 0.980 to 0.986 and from 0.91 to 0.96, and for the non-genotyped animals in Data 1, they ranged, respectively, from 0.987 to 0.994 and from 0.987 to 1, which indicate that the approximations were in line with the exact results. The best approach achieved correlations of 0.992 to 0.994 across lactations. CONCLUSIONS: Our results demonstrate that the approximated reliabilities calculated using our proposed approach are in good agreement with the exact reliabilities. The blended method for the genotyped animals is computationally more feasible than the direct method when RPG effects are included, particularly for large-scale datasets. The approach can serve as an effective strategy to estimate the reliabilities of GEBV in large-scale single-step genomic predictions.

Invited review: Reliability computation from the animal model era to the single-step genomic model era.

The calculation of exact reliabilities involving the inversion of mixed model equations poses a heavy computational challenge when the system of equations is large. This has prompted the development of different approximation methods. We give an overview of the various methods and computational approaches in calculating reliability from the era before the animal model to the era of single-step genomic models. The different methods are discussed in terms of modeling, development, and applicability in large dairy cattle populations. The paper also describes the problems faced in reliability computation. Many details dispersed throughout the literature are presented in this paper. It is clear that a universal solution applicable to every model and input data may not be possible, but we point out several efficient and accurate algorithms developed recently for a variety of very large genomic evaluations.

Single-step genomic BLUP with many metafounders.

Single-step genomic BLUP (ssGBLUP) model for routine genomic prediction of breeding values is developed intensively for many dairy cattle populations. Compatibility between the genomic (G) and the pedigree (A) relationship matrices remains an important challenge required in ssGBLUP. The compatibility relates to the amount of missing pedigree information. There are two prevailing approaches to account for the incomplete pedigree information: unknown parent groups (UPG) and metafounders (MF). unknown parent groups have been used routinely in pedigree-based evaluations to account for the differences in genetic level between groups of animals with missing parents. The MF approach is an extension of the UPG approach. The MF approach defines MF which are related pseudo-individuals. The MF approach needs a Γ matrix of the size number of MF to describe relationships between MF. The UPG and MF can be the same. However, the challenge in the MF approach is the estimation of Γ having many MF, typically needed in dairy cattle. In our study, we present an approach to fit the same amount of MF as UPG in ssGBLUP with Woodbury matrix identity (ssGTBLUP). We used 305-day milk, protein, and fat yield data from the DFS (Denmark, Finland, Sweden) Red Dairy cattle population. The pedigree had more than 6 million animals of which 207,475 were genotyped. We constructed the preliminary gamma matrix (Γ pre ) with 29 MF which was expanded to 148 MF by a covariance function (Γ 148). The quality of the extrapolation of the Γ pre matrix was studied by comparing average off-diagonal elements between breed groups. On average relationships among MF in Γ 148 were 1.8% higher than in Γ pre . The use of Γ 148 increased the correlation between the G and A matrices by 0.13 and 0.11 for the diagonal and off-diagonal elements, respectively. [G]EBV were predicted using the ssGTBLUP and Pedigree-BLUP models with the MF and UPG. The prediction reliabilities were slightly higher for the ssGTBLUP model using MF than UPG. The ssGBLUP MF model showed less overprediction compared to other models.

Local breed proportions and local breed heterozygosity in genomic predictions for crossbred dairy cows.

For genomic prediction of crossbred animals, models that account for the breed origin of alleles (BOA) in marker genotypes can allow the effects of marker alleles to differ depending on their ancestral breed. Previous studies have shown that genomic estimated breeding values for crossbred cows can be calculated using the marker effects that are estimated in the contributing pure breeds and combined based on estimated BOA in the genotypes of the crossbred cows. In the presented study, we further exploit the BOA information for improving the prediction of genomic breeding values of crossbred dairy cows. We investigated 2 types of BOA-derived breed proportions: global breed proportions, defined as the proportion of marker alleles assigned to each breed across the whole genome; and local breed proportions (LBP), defined as the proportions of alleles on chromosome segments which were assigned to each breed. Further, we investigated 2 BOA-derived measures of heterozygosity for the prediction of total genetic value. First, global breed heterozygosity, defined as the proportion of marker loci that have alleles originating in 2 different breeds over the whole genome. Second, local breed heterozygosity (LBH), defined as proportions of marker loci on chromosome segments that had alleles originating in 2 different breeds. We estimated variance related to LBP and LBH on the remaining variation after accounting for prediction with solutions from the genomic evaluations of the pure breeds and validated alternative models for production traits in 5,214 Danish crossbred dairy cows. The estimated LBP variances were 0.9, 1.2, and 1.0% of phenotypic variance for milk, fat, and protein yield, respectively. We observed no clear LBH effect. Cross-validation showed that models with LBP effects had a numerically small but statistically significantly higher predictive ability than models only including global breed proportions. We observed similar improvement in accuracy by the model having an across crossbred residual additive genetic effect, accounting for the additive genetic variation that was not accounted for by the solutions from purebred. For genomic predictions of crossbred animals, estimated BOA can give useful information on breed proportions, both globally in the genome and locally in genome regions, and on breed heterozygosity.

Single-step genomic BLUP with genetic groups and automatic adjustment for allele coding.

BACKGROUND: Genomic estimated breeding values (GEBV) by single-step genomic BLUP (ssGBLUP) are affected by the centering of marker information used. The use of a fixed effect called J factor will lead to GEBV that are unaffected by the centering used. We extended the use of a single J factor to a group of J factors. RESULTS: J factor(s) are usually included in mixed model equations (MME) as regression effects but a transformation similar to that regularly used for genetic groups can be applied to obtain a simpler MME, which is sparser than the original MME and does not need computation of the J factors. When the J factor is based on the same structure as the genetic groups, then MME can be transformed such that coefficients for the genetic groups no longer include information from the genomic relationship matrix. We illustrate the use of J factors in the analysis of a Red dairy cattle data set for fertility. CONCLUSIONS: The GEBV from these analyses confirmed the theoretical derivations that show that the resulting GEBV are allele coding independent when a J factor is used. Transformed MME led to faster computing time than the original regression-based MME.

Breeding value reliabilities for multiple-trait single-step genomic best linear unbiased predictor.

Approximate multistep methods to calculate reliabilities for estimated breeding values in large genetic evaluations were developed for single-trait (ST-R2A) and multitrait (MT-R2A) single-step genomic BLUP (ssGBLUP) models. First, a traditional animal model was used to estimate the amount of nongenomic information for the genotyped animals. Second, this information was used with genomic data in a genomic BLUP model (genomic BLUP/SNP-BLUP) to approximate the total amount of information and ssGBLUP reliabilities for the genotyped animals. Finally, reliabilities for the nongenotyped animals were calculated using a traditional animal model where the increased information due to genomic data for the genotyped animals is accounted for by including pseudo-record counts for the genotyped animals. The approaches were tested using a multiple-trait ssGBLUP model on 2 data sets. The first data set (data 1) was small enough such that exact ssGBLUP model reliabilities could be computed by inversion and compared with the approximation method reliabilities. Data 1 had 46,535 first-, 35,290 second-, and 23,780 third-lactation 305-d milk yield records from 47,124 Finnish Red dairy cows. The pedigree comprised 64,808 animals, of which 19,757 were genotyped. We examined the efficiency of the MT-R2A approximation on a large data set (data 2) derived from the joint Nordic (Danish, Finnish, and Swedish) Holstein dairy cattle data. Data 2 had 17.8 million 305-d milk records from 8.3 million cows and first 3 lactations. The pedigree had 11 million animals of which 274,145 were genotyped on 46,342 SNP markers. For data 1, correlations between the exact ssGBLUP model and the ST-R2A for the genotyped (nongenotyped) animals were 0.995 (0.987), 0.965 (0.984), and 0.950 (0.983) for first, second, and third lactation, respectively. Correspondingly, correlations between exact ssGBLUP reliabilities and MT-R2A for the genotyped (nongenotyped) animals were 0.995 (0.993), 0.992 (0.991), and 0.990 (0.990) for first, second, and third lactation, respectively. The regression coefficients (b1) of ssGBLUP reliability on ST-R2A for the genotyped (nongenotyped) animals ranged from 0.87 (0.94) for first lactation to 0.68 (0.93) for third lactation, whereas for MT-R2A they were between 0.91 (0.99) for first lactation to 0.89 (0.99) for third lactation. Correspondingly, the intercepts varied from 0.11 (0.05) to 0.3 (0.06) for ST-R2A and from 0.06 (0.01) to 0.07 (0.02) for MT-R2A. The computing time for the approximation method was approximately 12% of that required by the direct exact approach. In conclusion, the developed approximate approach allows calculating estimated breeding value reliabilities in the ssGBLUP model even for large data sets.

Single-step genomic evaluation of Russian dairy cattle using internal and external information.

Genomic data are widely used in predicting the breeding values of dairy cattle. The accuracy of genomic prediction depends on the size of the reference population and how related the candidate animals are to it. For populations with limited numbers of progeny-tested bulls, the reference populations must include cows and data from external populations. The aim of this study was to implement state-of-the-art single-step genomic evaluations for milk and fat yield in Holstein and Russian Black & White cattle in the Leningrad region (LR, Russia), using only a limited number of genotyped animals. We complemented internal information with external pseudo-phenotypic and genotypic data of bulls from the neighbouring Danish, Finnish and Swedish Holstein (DFS) population. Three data scenarios were used to perform single-step GBLUP predictions in the LR dairy cattle population. The first scenario was based on the original LR reference population, which constituted 1,080 genotyped cows and 427 genotyped bulls. In the second scenario, the genotypes of 414 bulls related to the LR from the DFS population were added to the reference population. In the third scenario, LR data were further augmented with pseudo-phenotypic data from the DFS population. The inclusion of foreign information increased the validation reliability of the milk yield by up to 30%. Suboptimal data recording practices hindered the improvement of fat yield. We confirmed that the single-step model is suitable for populations with a low number of genotyped animals, especially when external information is integrated into the evaluations. Genomic prediction in populations with a low number of progeny-tested bulls can be based on data from genotyped cows and on the inclusion of genotypes and pseudo-phenotypes from the external population. This approach increased the validation reliability of the implemented single-step model in the milk yield, but shortcomings in the LR data recording scheme prevented improvements in fat yield.

Genetic parameters for female fertility in Nordic Holstein and Red Cattle dairy breeds.

Genetic evaluation of female fertility in Danish, Finnish, and Swedish dairy cows was updated in 2015 to multiple-trait animal model evaluation, where heifer and cow fertility up to third parity are considered as separate traits. A model for conception rate was also developed, which required variance component estimation for Nordic Holstein and Nordic Red Dairy Cattle. We used a multiple-trait multiple-lactation sire model to determine variance components for interval from calving to first insemination, length of service period, and conception rate. Monte Carlo Expectation Maximization REML allowed estimation of all 11 traits simultaneously. Study data were sampled from Swedish Holstein (n = 140,040) and Red Dairy Cattle (n = 101,315) heifers and cows. Conception rate observations are binomial observations with various numbers of failures preceding an observation of success. Using a simulation study, we confirmed that including a service number effect into the conception rate model allowed us to model the change in expectation of successful AI with increasing number of services. Heifers outperformed cows in all fertility traits according to the phenotypic means in the records. Heritabilities for the traits varied from 3 to 7% for interval from calving to first insemination, from 1 to 5% for length of service period, and from 1 to 3% for conception rate. Genetic correlations within traits (i.e., between parities) were favorable, ranging from moderate to high; genetic correlations between heifer and cow traits were lower than between cow traits in different parities. Lowest genetic correlations between traits were for interval from calving to first insemination and conception rate, intermediate for interval from calving to first insemination and length of service period, and highest for length of service period and conception rate. The variance components estimated in this study have been used in Nordic fertility breeding value evaluations since 2016.

Efficient Monte Carlo algorithm for restricted maximum likelihood estimation of genetic parameters.

Monte Carlo (MC) methods have been found useful in estimation of variance parameters for large data and complex models with many variance components (VC), with respect to both computer memory and computing time. A disadvantage has been a fluctuation in round-to-round values of estimates that makes the estimation of convergence challenging. Furthermore, with Newton-type algorithms, the approximate Hessian matrix might have sufficient accuracy, but the inaccuracy in the gradient vector exaggerates the round-to-round fluctuation to intolerable. In this study, the reuse of the same random numbers within each MC sample was used to remove the MC fluctuation. Simulated data with six VC parameters were analysed by four different MC REML methods: expectation-maximization (EM), Newton-Raphson (NR), average information (AI) and Broyden's method (BM). In addition, field data with 96 VC parameters were analysed by MC EM REML. In all the analyses with reused samples, the MC fluctuations disappeared, but the final estimates by the MC REML methods differed from the analytically calculated values more than expected especially when the number of MC samples was small. The difference depended on the random numbers generated, and based on repeated MC AI REML analyses, the VC estimates were on average non-biased. The advantage of reusing MC samples is more apparent in the NR-type algorithms. Smooth convergence opens the possibility to use the fast converging Newton-type algorithms. However, a disadvantage from reusing MC samples is a possible "bias" in the estimates. To attain acceptable accuracy, sufficient number of MC samples need to be generated.

Single step genomic evaluation for female fertility in Nordic Red dairy cattle.

Joint Nordic (Denmark, Finland, Sweden) genetic evaluation of female fertility is currently based on the multiple trait multilactation animal model (BLUP). Here, single step genomic model (ssGBLUP) was applied for the Nordic Red dairy cattle fertility evaluation. The 11 traits comprised of nonreturn rate and days from first to last insemination in heifers and first three parities, and days from calving to first insemination in the first three parities. Traits had low heritabilities (0.015-0.04), but moderately high genetic correlations between the parities (0.60-0.88). Phenotypic data included 4,226,715 animals with records and pedigree 5,445,392 animals. Unknown parents were assigned into 332 phantom parent groups (PPG). In mixed model equations animals were associated with PPG effects through the pedigree or both the pedigree and genomic information. Genotype information of 46,914 SNPs was available for 33,969 animals in the pedigree. When PPG used pedigree information only, BLUP converged after 2,420 iterations whereas the ssGBLUP evaluation needed over ten thousand iterations. When the PPG effects were solved accounting both the pedigree and the genomic information, the ssGBLUP model converged after 2,406 iterations. Also, with the latter model breeding values by ssGBLUP and BLUP became more consistent and genetic trends followed each other well. Models were validated using forward prediction of the young bulls. Reliabilities and variance inflation of predicted genomic breeding values (values for parent averages in brackets) for the 11 traits ranged 0.22-0.31 (0.10-0.27) and 0.81-0.95 (0.83-1.06), respectively. The ssGBLUP model gave always higher validation reliabilities than BLUP, but largest increases were for the cow fertility traits.

Single-step SNP-BLUP with on-the-fly imputed genotypes and residual polygenic effects.

BACKGROUND: Single-step genomic best linear unbiased prediction (BLUP) evaluation combines relationship information from pedigree and genomic marker data. The inclusion of the genomic information into mixed model equations requires the inverse of the combined relationship matrix [Formula: see text], which has a dense matrix block for genotyped animals. METHODS: To avoid inversion of dense matrices, single-step genomic BLUP can be transformed to single-step single nucleotide polymorphism BLUP (SNP-BLUP) which have observed and imputed marker coefficients. Simple block LDL type decompositions of the single-step relationship matrix [Formula: see text] were derived to obtain different types of linearly equivalent single-step genomic mixed model equations with different sets of reparametrized random effects. For non-genotyped animals, the imputed marker coefficient terms in the single-step SNP-BLUP were calculated on-the-fly during the iterative solution using sparse matrix decompositions without storing the imputed genotypes. Residual polygenic effects were added to genotyped animals and transmitted to non-genotyped animals using relationship coefficients that are similar to imputed genotypes. The relationships were further orthogonalized to improve convergence of iterative methods. RESULTS: All presented single-step SNP-BLUP models can be solved efficiently using iterative methods that rely on iteration on data and sparse matrix approaches. The efficiency, accuracy and iteration convergence of the derived mixed model equations were tested with a small dataset that included 73,579 animals of which 2885 were genotyped with 37,526 SNPs. CONCLUSIONS: Inversion of the large and dense genomic relationship matrix was avoided in single-step evaluation by using fully orthogonalized single-step SNP-BLUP formulations. The number of iterations until convergence was smaller in single-step SNP-BLUP formulations than in the original single-step GBLUP when heritability was low, but increased above that of the original single-step when heritability was high.

Across-country test-day model evaluations for Holstein, Nordic Red Cattle, and Jersey.

Three random regression models were developed for routine genetic evaluation of Danish, Finnish, and Swedish dairy cattle. Data included over 169 million test-day records with milk, protein, and fat yield observations from over 8.7 million dairy cows of all breeds. Variance component analyses showed significant differences in estimates between Holstein, Nordic Red Cattle, and Jersey, but only small to moderate differences within a breed across countries. The obtained variance component estimates were used to build, for each breed, their own set of covariance functions. The covariance functions describe the animal effects on milk, protein, and fat yields of the first 3 lactations as 9 different traits, assuming the same heritabilities and a genetic correlation of unity across countries. Only 15, 27, and 7 eigenfunctions with the largest eigenvalues were used to describe additive genetic animal effects and nonhereditary animal effects across lactations and within later lactations, respectively. These reduced-rank covariance functions explained 99.0 to 99.9% of the original variances but reduced the number of animal equations to be solved by 44%. Moderate rank reduction for nonhereditary animal effects and use of one-third-smaller measurement error correlations than obtained from variance component estimation made the models more robust against extreme observations. Estimation of the genetic levels of the countries' subpopulations within a breed was found sensitive to the way the breed effects were modeled, especially for the genetically heterogeneous Nordic Red Cattle. Means to ensure that only additive genetic effects entered the estimated breeding values were to describe the crossbreeding effects by fixed and random cofactors and the calving age effect by an age × breed proportion interaction, and to model phantom parent groups as random effects. To ensure that genetic variances were the same across the 3 countries in breeding value estimation, as suggested by the variance component estimates, the applied multiplicative heterogeneous variance adjustment method had to be tailored using country-specific reference measurement error variances. Results showed the feasibility of across-country genetic evaluation of cows and sires based on original test-day phenotypes. Nevertheless, applying a thorough model validation procedure is essential throughout the model building process to obtain reliable breeding values.

Employing a Monte Carlo algorithm in Newton-type methods for restricted maximum likelihood estimation of genetic parameters.

Estimation of variance components by Monte Carlo (MC) expectation maximization (EM) restricted maximum likelihood (REML) is computationally efficient for large data sets and complex linear mixed effects models. However, efficiency may be lost due to the need for a large number of iterations of the EM algorithm. To decrease the computing time we explored the use of faster converging Newton-type algorithms within MC REML implementations. The implemented algorithms were: MC Newton-Raphson (NR), where the information matrix was generated via sampling; MC average information(AI), where the information was computed as an average of observed and expected information; and MC Broyden's method, where the zero of the gradient was searched using a quasi-Newton-type algorithm. Performance of these algorithms was evaluated using simulated data. The final estimates were in good agreement with corresponding analytical ones. MC NR REML and MC AI REML enhanced convergence compared to MC EM REML and gave standard errors for the estimates as a by-product. MC NR REML required a larger number of MC samples, while each MC AI REML iteration demanded extra solving of mixed model equations by the number of parameters to be estimated. MC Broyden's method required the largest number of MC samples with our small data and did not give standard errors for the parameters directly. We studied the performance of three different convergence criteria for the MC AI REML algorithm. Our results indicate the importance of defining a suitable convergence criterion and critical value in order to obtain an efficient Newton-type method utilizing a MC algorithm. Overall, use of a MC algorithm with Newton-type methods proved feasible and the results encourage testing of these methods with different kinds of large-scale problem settings.

Principal component and factor analytic models in international sire evaluation.

BACKGROUND: Interbull is a non-profit organization that provides internationally comparable breeding values for globalized dairy cattle breeding programmes. Due to different trait definitions and models for genetic evaluation between countries, each biological trait is treated as a different trait in each of the participating countries. This yields a genetic covariance matrix of dimension equal to the number of countries which typically involves high genetic correlations between countries. This gives rise to several problems such as over-parameterized models and increased sampling variances, if genetic (co)variance matrices are considered to be unstructured. METHODS: Principal component (PC) and factor analytic (FA) models allow highly parsimonious representations of the (co)variance matrix compared to the standard multi-trait model and have, therefore, attracted considerable interest for their potential to ease the burden of the estimation process for multiple-trait across country evaluation (MACE). This study evaluated the utility of PC and FA models to estimate variance components and to predict breeding values for MACE for protein yield. This was tested using a dataset comprising Holstein bull evaluations obtained in 2007 from 25 countries. RESULTS: In total, 19 principal components or nine factors were needed to explain the genetic variation in the test dataset. Estimates of the genetic parameters under the optimal fit were almost identical for the two approaches. Furthermore, the results were in a good agreement with those obtained from the full rank model and with those provided by Interbull. The estimation time was shortest for models fitting the optimal number of parameters and prolonged when under- or over-parameterized models were applied. Correlations between estimated breeding values (EBV) from the PC19 and PC25 were unity. With few exceptions, correlations between EBV obtained using FA and PC approaches under the optimal fit were ≥ 0.99. For both approaches, EBV correlations decreased when the optimal model and models fitting too few parameters were compared. CONCLUSIONS: Genetic parameters from the PC and FA approaches were very similar when the optimal number of principal components or factors was fitted. Over-fitting increased estimation time and standard errors of the estimates but did not affect the estimates of genetic correlations or the predictions of breeding values, whereas fitting too few parameters affected bull rankings in different countries.

Principal component approach in variance component estimation for international sire evaluation.

BACKGROUND: The dairy cattle breeding industry is a highly globalized business, which needs internationally comparable and reliable breeding values of sires. The international Bull Evaluation Service, Interbull, was established in 1983 to respond to this need. Currently, Interbull performs multiple-trait across country evaluations (MACE) for several traits and breeds in dairy cattle and provides international breeding values to its member countries. Estimating parameters for MACE is challenging since the structure of datasets and conventional use of multiple-trait models easily result in over-parameterized genetic covariance matrices. The number of parameters to be estimated can be reduced by taking into account only the leading principal components of the traits considered. For MACE, this is readily implemented in a random regression model. METHODS: This article compares two principal component approaches to estimate variance components for MACE using real datasets. The methods tested were a REML approach that directly estimates the genetic principal components (direct PC) and the so-called bottom-up REML approach (bottom-up PC), in which traits are sequentially added to the analysis and the statistically significant genetic principal components are retained. Furthermore, this article evaluates the utility of the bottom-up PC approach to determine the appropriate rank of the (co)variance matrix. RESULTS: Our study demonstrates the usefulness of both approaches and shows that they can be applied to large multi-country models considering all concerned countries simultaneously. These strategies can thus replace the current practice of estimating the covariance components required through a series of analyses involving selected subsets of traits. Our results support the importance of using the appropriate rank in the genetic (co)variance matrix. Using too low a rank resulted in biased parameter estimates, whereas too high a rank did not result in bias, but increased standard errors of the estimates and notably the computing time. CONCLUSIONS: In terms of estimation's accuracy, both principal component approaches performed equally well and permitted the use of more parsimonious models through random regression MACE. The advantage of the bottom-up PC approach is that it does not need any previous knowledge on the rank. However, with a predetermined rank, the direct PC approach needs less computing time than the bottom-up PC.

Estimation of non-linear growth models by linearization: a simulation study using a Gompertz function.

A method based on Taylor series expansion for estimation of location parameters and variance components of non-linear mixed effects models was considered. An attractive property of the method is the opportunity for an easily implemented algorithm. Estimation of non-linear mixed effects models can be done by common methods for linear mixed effects models, and thus existing programs can be used after small modifications. The applicability of this algorithm in animal breeding was studied with simulation using a Gompertz function growth model in pigs. Two growth data sets were analyzed: a full set containing observations from the entire growing period, and a truncated time trajectory set containing animals slaughtered prematurely, which is common in pig breeding. The results from the 50 simulation replicates with full data set indicate that the linearization approach was capable of estimating the original parameters satisfactorily. However, estimation of the parameters related to adult weight becomes unstable in the case of a truncated data set.

Recording strategies and selection potential of feed intake measured using the X-ray method in rainbow trout.

This study examines the way long-term feed intake should be recorded accurately for selective breeding purposes, and estimates selection potential in feed intake using the X-ray method to record individual daily feed intake in rainbow trout (Oncorhynchus mykiss). The analysis showed that the point estimates of daily feed intake displayed low repeatabilities (r = 0.09-0.32). This indicates that a minimum of three repeated records were needed to accurately record average feed intake at a fixed age. To effectively breed for feed intake over the whole growing period, it is necessary to determine average feed intake at different ages, since there were only moderate phenotypic and genetic correlations between average daily feed intake recorded at 140 g, 750 g and 2000 g wet mass. Heritability for average daily feed intake was low (average h(2) = 0.10), indicating that modest genetic changes can be obtained in response to selection. It was concluded that selection to genetically change long-term feed intake can be successful, yet repeated observations at several life stages are needed to ensure the accuracy of feed intake estimates and the efficiency of selection.