Pedigree-based Relationship Matrix Research Articles

Genetic and genomic evaluation of age at first calving in Italian Holsteins

Age at first calving (AFC) represents the non-productive period of around 2 yr in Holstein cows and thus it has a relevant impact on the cost of rearing replacements in the dairy herd. In the present study we aimed at evaluating genetic and genomic aspects of AFC in the Italian Holstein population. Data of 4,206,218 heifers with first calving between 1996 and 2020 were used. Age at first calving averaged 26.09 ± 3.07 mo and decreased across years. Heritability was estimated using a linear animal model which included the fixed effects of herd-year-season of birth and classes of gestation length, and the random animal additive genetic effect fitted to a pedigree-based relationship matrix. Estimated breeding values (EBV) and genomically enhanced breeding values (GEBV) were obtained, and they were standardized to mean 100 and standard deviation 5, where animals above the mean are those contributing to reduce AFC. Heritability estimates of AFC ranged from 0.031 to 0.045. The trend of sires' GEBV was favorable and indicated a reduced AFC across years. Approximate genetic correlations between GEBV of AFC and GEBV of other economically important traits were calculated on a subset of genotyped females born after 2015. Moderate favorable associations of AFC with production traits (0.39 to 0.51), udder depth (0.40), interval from first to last insemination in heifer (-0.43), and longevity (0.34) were assessed. Overall, the greatest lifetime productive performances and most favorable days open in first lactation were observed when heifers calved at 22-23 mo. On the other hand, progeny of sires with GEBV of AFC above the mean yielded more milk, fat, and protein in first lactation, and had shorter days open than progeny of sires with GEBV of AFC below the mean. Results suggested that breeding strategies to improve AFC should be pursued, also considering genetic correlations between AFC and traits which are already part of the Italian Holstein breeding objective. The inclusion of AFC in an aggregate index is expected to contribute to enhance farm income.

Genomic evaluation for breeding and genetic management in Cordia africana, a multipurpose tropical tree species

BackgroundPlanting tested forest reproductive material is crucial to ensure the increased resilience of intensively managed productive stands for timber and wood product markets under climate change scenarios. Single-step Genomic Best Linear Unbiased Prediction (ssGBLUP) analysis is a cost-effective option for using genomic tools to enhance the accuracy of predicted breeding values and genetic parameter estimation in forest tree species. Here, we tested the efficiency of ssGBLUP in a tropical multipurpose tree species, Cordia africana, by partial population genotyping. A total of 8070 trees from three breeding seedling orchards (BSOs) were phenotyped for height. We genotyped 6.1% of the phenotyped individuals with 4373 single nucleotide polymorphisms. The results of ssGBLUP were compared with pedigree-based best linear unbiased prediction (ABLUP) and genomic best linear unbiased prediction (GBLUP), based on genetic parameters, theoretical accuracy of breeding values, selection candidate ranking, genetic gain, and predictive accuracy and prediction bias.ResultsGenotyping a subset of the study population provided insights into the level of relatedness in BSOs, allowing better genetic management. Due to the inbreeding detected within the genotyped provenances, we estimated genetic parameters both with and without accounting for inbreeding. The ssGBLUP model showed improved performance in terms of additive genetic variance and theoretical breeding value accuracy. Similarly, ssGBLUP showed improved predictive accuracy and lower bias than the pedigree-based relationship matrix (ABLUP).ConclusionsThis study of C. africana, a species in decline due to deforestation and selective logging, revealed inbreeding depression. The provenance exhibiting the highest level of inbreeding had the poorest overall performance. The use of different relationship matrices and accounting for inbreeding did not substantially affect the ranking of candidate individuals. This is the first study of this approach in a tropical multipurpose tree species, and the analysed BSOs represent the primary effort to breed C. africana.

Genomic prediction using the lmekin function from the coxme R package

The increasing use of genomic selection (GS) in plant and animal breeding programs has led to the development of software that fits models based on unique scenarios. Accordingly, several R packages have been developed for GS. The lmekin function from the coxme R package was one of the first functions implemented in R to fit models with random family effects using the pedigree–based relationship matrix. The function allows the user to provide the covariance structures for the random effects; thus, the GBLUP model can be fitted. This fitting process consists of replacing, in the traditional BLUP model, the additive relationship matrix derived from a pedigree by the additive relationship matrix derived from markers. Thus, the objective of this study was to employ the lmekin function in the context of genomic prediction by comparing the results of this function with those obtained using five R packages for GS: rrBLUP, BGLR, sommer, lme4qtl, and lme4GS. The comparisons were performed considering the computational times and predicted values for a wheat dataset and simulated big data. In addition, we implemented a 5-fold cross-validation scheme through considering the values predicted by the lmekin function for the wheat dataset. The results indicated that the lmekin function was effective in predicting genomic breeding values considering multiple random effects and relatively small sample sizes. The rrBLUP package processed the fastest for the scenario with only one genetic random effect, and the high temporal efficiency of the sommer package was confirmed for the scenario with more than one genetic random effect. Differences in computational times occurred because of the different algorithms implemented in the packages to estimate the variance components.

Estimation of the genetic parameters for temperament in Haflinger horses based on the results of a character test using a pedigree-based relationship matrix

Literature provides indications that the personality and temperament of a horse are sufficiently genetically determined. This background was the impetus to continue collecting phenotypic information on the temperament in the South Tyrolean Haflinger horses, not only to have marketing tools but also to be able to estimate genetic parameters for a future genetic evaluation. The latter should open the possibility to include temperament traits in the breeding goal. Therefore, the aim of the present study was to consider for the very first-time phenotypic information from character testing for estimating genetic parameters for temperament in Haflinger horses. For the present study, data from routine character testing of 210 South Tyrolean Haflinger mares between 2019 and 2021 were considered. A multivariate estimation of heritability (h2) and genetic correlations (r) was carried out for the criteria ‘Interest’, ‘Activity’ and ‘Excitement’. In addition, univariate and bivariate estimation runs were also carried out. The pedigree-based heritability for the behavioural trait Interest was lowest with 0.07 and a standard error of ±0.04. For the other two criteria, Activity and Excitement, h2 was 0.19 ± 0.05 and 0.21 ± 0.04, respectively. The genetic correlation between Interest and Activity and between Interest and Excitement was in both cases negative with −0.52 and −0.61, respectively while the genetic correlation between Excitement and Activity was positive (0.99). Results of the present study highlight the potential of considering temperament traits in breeding of Haflinger horses.

Impact of kinship matrices on genetic gain and inbreeding with optimum contribution selection in a genomic dairy cattle breeding program

BackgroundGenomic selection has increased genetic gain in dairy cattle, but in some cases it has resulted in higher inbreeding rates. Therefore, there is need for research on efficient management of inbreeding in genomically-selected dairy cattle populations, especially for local breeds with a small population size. Optimum contribution selection (OCS) minimizes the increase in average kinship while it maximizes genetic gain. However, there is no consensus on how to construct the kinship matrix used for OCS and whether it should be based on pedigree or genomic information. VanRaden’s method 1 (VR1) is a genomic relationship matrix in which centered genotype scores are scaled with the sum of 2p(1-p) where p is the reference allele frequency at each locus, and VanRaden’s method 2 (VR2) scales each locus with 2p(1-p), thereby giving greater weight to loci with a low minor allele frequency. We compared the effects of nine kinship matrices on genetic gain, kinship, inbreeding, genetic diversity, and minor allele frequency when applying OCS in a simulated small dairy cattle population. We used VR1 and VR2, each using base animals, all genotyped animals, and the current generation of animals to compute reference allele frequencies. We also set the reference allele frequencies to 0.5 for VR1 and the pedigree-based relationship matrix. We constrained OCS to select a fixed number of sires per generation for all scenarios. Efficiency of the different matrices were compared by calculating the rate of genetic gain for a given rate of increase in average kinship.ResultsWe found that: (i) genomic relationships were more efficient than pedigree-based relationships at managing inbreeding, (ii) reference allele frequencies computed from base animals were more efficient compared to reference allele frequencies computed from recent animals, and (iii) VR1 was slightly more efficient than VR2, but the difference was not statistically significant.ConclusionsUsing genomic relationships for OCS realizes more genetic gain for a given amount of kinship and inbreeding than using pedigree relationships when the number of sires is fixed. For a small genomic dairy cattle breeding program, we recommend that the implementation of OCS uses VR1 with reference allele frequencies estimated either from base animals or old genotyped animals.

Genomic selection in algae with biphasic lifecycles: A Saccharina latissima (sugar kelp) case study

IntroductionSugar kelp (Saccharina latissima) has a biphasic life cycle, allowing selection on both thediploid sporophytes (SPs) and haploid gametophytes (GPs).MethodsWe trained a genomic selection (GS) model from farm-tested SP phenotypic data and used a mixed-ploidy additive relationship matrix to predict GP breeding values. Topranked GPs were used to make crosses for further farm evaluation. The relationship matrix included 866 individuals: a) founder SPs sampled from the wild; b) progeny GPs from founders; c) Farm-tested SPs crossed from b); and d) progeny GPs from farm-tested SPs. The complete pedigree-based relationship matrix was estimated for all individuals. A subset of founder SPs (n = 58) and GPs (n = 276) were genotyped with Diversity Array Technology and whole genome sequencing, respectively. We evaluated GS prediction accuracy via cross validation for SPs tested on farm in 2019 and 2020 using a basic GBLUP model. We also estimated the general combining ability (GCA) and specific combining ability (SCA) variances of parental GPs. A total of 11 yield-related and morphology traits were evaluated.ResultsThe cross validation accuracies for dry weight per meter (r ranged from 0.16 to 0.35) and wet weight per meter (r ranged 0.19 to 0.35) were comparable to GS accuracy for yield traits in terrestrial crops. For morphology traits, cross validation accuracy exceeded 0.18 in all scenarios except for blade thickness in the second year. Accuracy in a third validation year (2021) was 0.31 for dry weight per meter over a confirmation set of 87 individuals.DiscussionOur findings indicate that progress can be made in sugar kelp breeding by using genomic selection.

Comparing Methods to Select Candidates for Re-Genotyping to Impute Higher-Density Genotype Data in a Japanese Black Cattle Population: A Case Study

As optimization methods to identify the best animals for dense genotyping to construct a reference population for genotype imputation, the MCA and MCG methods, which use the pedigree-based additive genetic relationship matrix (A matrix) and the genomic relationship matrix (G matrix), respectively, have been proposed. We assessed the performance of MCA and MCG methods using 575 Japanese Black cows. Pedigree data were provided to trace back up to five generations to construct the A matrix with changing the pedigree depth from 1 to 5 (five MCA methods). Genotype information on 36,426 single-nucleotide polymorphisms was used to calculate the G matrix based on VanRaden's methods 1 and 2 (two MCG methods). The MCG always selected one cow per iteration, while MCA sometimes selected multiple cows. The number of commonly selected cows between the MCA and MCG methods was generally lower than that between different MCA methods or between different MCG methods. For the studied population, MCG appeared to be more reasonable than MCA in selecting cows as a reference population for higher-density genotype imputation to perform genomic prediction and a genome-wide association study.

A Note on the Conditioning of the H-1 Matrix Used in Single-Step GBLUP.

The single-step genomic BLUP (ssGBLUP) is used worldwide for the simultaneous genetic evaluation of genotyped and non-genotyped animals. It is easily extendible to all BLUP models by replacing the pedigree-based additive genetic relationship matrix (A) with an augmented pedigree-genomic relationship matrix (H). Theoretically, H does not introduce any artificially inflated variance. However, inflated genetic variances have been observed due to the incomparability between the genomic relationship matrix (G) and A used in H. Usually, G is blended and tuned with A22 (the block of A for genotyped animals) to improve its numerical condition and compatibility. If deflation/inflation is still needed, a common approach is weighting G-1-A22-1 in the form of τG-1-ωA22-1, added to A-1 to form H-1. In some situations, this can violate the conditional properties upon which H is built. Different ways of weighting the H-1 components (A-1, G-1, A22-1, and H-1 itself) were studied to avoid/minimise the violations of the conditional properties of H. Data were simulated on ten populations and twenty generations. Responses to weighting different components of H-1 were measured in terms of the regression of phenotypes on the estimated breeding values (the lower the slope, the higher the inflation) and the correlation between phenotypes and the estimated breeding values (predictive ability). Increasing the weight on H-1 increased the inflation. The responses to weighting G-1 were similar to those for H-1. Increasing the weight on A-1 (together with A22-1) was not influential and slightly increased the inflation. Predictive ability is a direct function of the slope of the regression line and followed similar trends. Responses to weighting G-1-A22-1 depend on the inflation/deflation of evaluations from A-1 to H-1 and the compatibility of the two matrices with the heritability used in the model. One possibility is a combination of weighting G-1-A22-1 and weighting H-1. Given recent advances in ssGBLUP, conditioning H-1 might become an interim solution from the past and then not be needed in the future.

Genetic and nongenetic variation of heifer fertility in Italian Holstein cattle

Excellent fertility performance is important to maximize farmers' profit and to reduce the number of culled animals. Although female fertility of adult cows has been included in Italian Holstein breeding objectives since 2009, little has been done to quantify genetic variation of heifer fertility characteristics so far. The aim of the present study was to estimate genetic parameters of 4 fertility traits in nulliparous Italian Holstein heifers and to develop an aggregate selection index to improve heifer fertility. Data were retrieved from the national fertility database and included information on insemination, calving, and pregnancy diagnosis dates. The investigated phenotypes (mean ± standard deviation) were age at first insemination (AFI, mo; 17.25 ± 2.89), nonreturn rate at 56 d from the first insemination (NRR56, binary; 0.78 ± 0.41), conception rate at first insemination (CR, binary; 0.61 ± 0.49), and interval from first to last insemination (IFL, d; 26.09 ± 51.85). Genetic parameters were estimated using a 4-trait animal model that included the following fixed effects: herd-year of birth and month of birth for AFI, and herd-year-season of birth and month-year of insemination for IFL, NRR56, and CR; the animal additive genetic effect (fitted to the pedigree-based relationship matrix) was considered as a random term. An aggregate index was developed from the estimated additive genetic (co)variance matrix by considering CR as the breeding goal and AFI, NRR56, and IFL as selection criteria. Heritability estimates from average covariance matrices ranged from 0.012 (CR) to 0.015 (IFL), with the exception of AFI (0.071). Conception rate at first insemination was strongly correlated with both IFL (−0.730) and NRR56 (0.668), and weakly to AFI (−0.065), and the relative emphasis placed on each selection criteria in the aggregate index was 10%, 47%, and 43% for AFI, IFL, and NRR56, respectively. The results of the present study suggest that heifer fertility should be considered as an additional trait in the breeding objectives of Italian Holstein.

A multi-omics framework reveals strawberry flavor genes and their regulatory elements.

Flavor is essential to consumer preference of foods and is an increasing focus of plant breeding programs. In fruit crops, identifying genes underlying volatile organic compounds has great promise to accelerate flavor improvement, but polyploidy and heterozygosity in many species have slowed progress. Here we use octoploid cultivated strawberry to demonstrate how genomic heterozygosity, transcriptomic intricacy and fruit metabolomic diversity can be treated as strengths and leveraged to uncover fruit flavor genes and their regulatory elements. Multi-omics datasets were generated including an expression quantitative trait loci map with 196 diverse breeding lines, haplotype-phased genomes of a highly-flavored breeding selection, a genome-wide structural variant map using five haplotypes, and volatile genome-wide association study (GWAS) with > 300 individuals. Overlaying regulatory elements, structural variants and GWAS-linked allele-specific expression of numerous genes to variation in volatile compounds important to flavor. In one example, the functional role of anthranilate synthase alpha subunit 1 in methyl anthranilate biosynthesis was supported via fruit transient gene expression assays. These results demonstrate a framework for flavor gene discovery in fruit crops and a pathway to molecular breeding of cultivars with complex and desirable flavor.

Integrating genomic information and productivity and climate-adaptability traits into a regional white spruce breeding program.

Tree improvement programs often focus on improving productivity-related traits; however, under present climate change scenarios, climate change-related (adaptive) traits should also be incorporated into such programs. Therefore, quantifying the genetic variation and correlations among productivity and adaptability traits, and the importance of genotype by environment interactions, including defense compounds involved in biotic and abiotic resistance, is essential for selecting parents for the production of resilient and sustainable forests. Here, we estimated quantitative genetic parameters for 15 growth, wood quality, drought resilience, and monoterpene traits for Picea glauca (Moench) Voss (white spruce). We sampled 1,540 trees from three open-pollinated progeny trials, genotyped with 467,224 SNP markers using genotyping-by-sequencing (GBS). We used the pedigree and SNP information to calculate, respectively, the average numerator and genomic relationship matrices, and univariate and multivariate individual-tree models to obtain estimates of (co)variance components. With few site-specific exceptions, all traits examined were under genetic control. Overall, higher heritability estimates were derived from the genomic- than their counterpart pedigree-based relationship matrix. Selection for height, generally, improved diameter and water use efficiency, but decreased wood density, microfibril angle, and drought resistance. Genome-based correlations between traits reaffirmed the pedigree-based correlations for most trait pairs. High and positive genetic correlations between sites were observed (average 0.68), except for those pairs involving the highest elevation, warmer, and moister site, specifically for growth and microfibril angle. These results illustrate the advantage of using genomic information jointly with productivity and adaptability traits, and defense compounds to enhance tree breeding selection for changing climate.

A Random Regression Model Based on a Single-Step Method for Improving the Genomic Prediction Accuracy of Residual Feed Intake in Pigs.

Residual feed intake (RFI) is considered as a measurement of feed efficiency, which is greatly related to the growth performance in pigs. Daily feeding records can be obtained from automatic feeders. In general, RFI is usually calculated from the total measurement records during the whole test period. This measurement cannot reflect genetic changes in different growth periods during the test. A random regression model (RRM) provides a method to model such type of longitudinal data. To improve the accuracy of genetic prediction for RFI, the RRM and regular animal models were applied in this study, and their prediction performances were compared. Both traditional pedigree-based relationship matrix (A matrix) and pedigree and genomic information-based relationship matrix (H matrix) were applied for these two models. The results showed that, the prediction accuracy of the RRM was higher than that of the animal model, increasing 24.2% with A matrix and 40.9% with H matrix. Furthermore, genomic information constantly improved the accuracy of evaluation under each evaluation model. In conclusion, longitudinal traits such as RFI can describe feed efficiency better, and the RRM with both pedigree and genetic information was superior to the animal model. These results provide a feasible method of genomic prediction using longitudinal data in animal breeding.

PSI-6 Genetic Parameters for Food Allergy Responses in Divergently-selected Pig Lines

Abstract Genetic selection for health and welfare-related traits is paramount in livestock breeding. Piglet allergic responses to soybean protein negatively impact animal growth and welfare. The objectives of this study were to estimate the heritability of soybean allergic responses and genetic correlations between soybean and peanut allergies (r1,2) in divergently-selected pig lines. The base population consisted of nine sire lines (primarily Yorkshire and Landrace) and two dam sources (Yorkshire × Chester White-F1 and Landrace × Yorkshire-F1). Soybean and peanut-allergic responses were measured through a skin test (0–6 scale; 0-no allergic responses; 6-severe allergic responses; characterized by wheal and flair) on pigs fed a diet containing soybean meal for 21 d post-weaning. A total of 5,505 animals from nine generations of two contrasting lines (i.e., high and low skin test reaction) were included in the analyses. The statistical model included contemporary group, breeding lines, replication, sex, and weaning weight (linear covariate) as fixed effects (P-value < 0.05), and additive genetic, maternal genetic, and maternal permanent environment as random effects. Threshold and linear Bayesian models were used to estimate genetic parameters, using a pedigree-based relationship matrix containing 9,201 animals. The heritability estimates for the general soybean allergic response were 0.199 ± 0.045 and 0.119 ± 0.025 for the threshold (liability scale) and linear models, respectively, suggesting that soybean allergic responses are heritable and can be improved through selective breeding. A weak negative genetic correlation between allergic responses and birth weight was observed (r1,2 = -0.253 ± 0.192), which shows that the genetic variance of soybean allergy is less dependent on birth weight. However, a high positive genetic correlation was estimated between soybean and peanut-allergic responses (r1,2 = 0.89 ± 0.048), which indicates a potential cross-reactivity of soybean and peanut allergies. Our findings suggest that it is possible to reduce food allergy responses in pigs through selective breeding.

PSI-10 Genetic parameters for soybean allergy response in divergently-selected pigs

Abstract Genetically selecting for improved animal resilience has become of great interest in modern livestock breeding. In this context, piglets experience a transitory allergic response to soybean proteins that has a negative impact on their growth and welfare. The main objective of this study was to determine whether soybean allergy response is a heritable trait, using divergently selected pig lines. The base generation pigs were generated from nine sire lines (primarily Yorkshire and Landrace) and two dam sources (Yorkshire × Chester White F1 and Landrace × Yorkshire F1). Soybean allergic response was measured, following a skin test protocol, on pigs that received a diet containing 28% soybean meal for 21 consecutive days post-weaning. The skin test was performed with an intradermal injection of soybean protein. Wheel and flare scores (ranging from 0 to 3) were assigned approximately 60 minutes post-injection. Based on the skin test total scores (i.e., the sum of the wheal and flair scores, ranging from 0 to 6), animals were selected to create two (a high and a low) soybean protein reacting lines, and bred within lines for nine generations. Phenotypes used in this study were measured on a total of 5,505 animals born from generations one to nine, in both high and low reacting lines. Significant systematic effects (P < 0.05) included in the statistical model were birth year, birth season, sex, and weaning weight (as a covariate). Genetic parameters for the skin test were estimated using a pedigree-based relationship matrix containing 9,201 animals, and a threshold animal mixed model. The genetic variance estimated for soybean allergy was 0.091, indicating that there is genetic variability in this trait. The heritability estimated for soybean allergy response was 0.139, which indicates that food allergy response is a heritable trait and can be improved through selective breeding.

Theoretical and empirical comparisons of expected and realized relationships for the X-chromosome

BackgroundX-chromosomal loci present different inheritance patterns compared to autosomal loci and must be modeled accordingly. Sexual chromosomes are not systematically considered in whole-genome relationship matrices although rules based on genealogical or marker information have been derived. Loci on the X-chromosome could have a significant contribution to the additive genetic variance, in particular for some traits such as those related to reproduction. Thus, accounting for the X-chromosome relationship matrix might be informative to better understand the architecture of complex traits (e.g., by estimating the variance associated to this chromosome) and to improve their genomic prediction. For such applications, previous studies have shown the benefits of combining information from genotyped and ungenotyped individuals.ResultsIn this paper, we start by presenting rules to compute a genomic relationship matrix (GRM) for the X-chromosome (GX) without making any assumption on dosage compensation, and based on coding of gene content with 0/1 for males and 0/1/2 for females. This coding adjusts naturally to previously derived pedigree-based relationships (S) for the X-chromosome. When needed, we propose to accommodate and estimate dosage compensation and genetic heterogeneity across sexes via multiple trait models. Using a Holstein dairy cattle dataset, including males and females, we then empirically illustrate that realized relationships (GX) matches expectations (S). However, GX presents high deviations from S. GX has also a lower dimensionality compared to the autosomal GRM. In particular, individuals are frequently identical along the entire chromosome. Finally, we confirm that the heritability of gene content for markers on the X-chromosome that are estimated by using S is 1, further demonstrating that S and GX can be combined. For the pseudo-autosomal region, we demonstrate that the expected relationships vary according to position because of the sex-gradient. We end by presenting the rules to construct the 'H matrix’ by combining both relationship matrices.ConclusionsThis work shows theoretically and empirically that a pedigree-based relationship matrix built with rules specifically developed for the X-chromosome (S) matches the realized GRM for the X-chromosome. Therefore, applications that combine expected relationships and genotypes for markers on the X-chromosome should use S and GX.

Genomic prediction of growth traits for pigs in the presence of genotype by environment interactions using single-step genomic reaction norm model.

Economically important traits are usually complex traits influenced by genes, environment and genotype-by-environment (G×E) interactions. Ignoring G×E interaction could lead to bias in the estimation of breeding values and selection decisions. A total of 1,778 pigs were genotyped using the PorcineSNP80 BeadChip. The existence of G×E interactions was investigated using a single-step reaction norm model for growth traits of days to 100kg (AGE) and backfat thickness adjusted to 100kg (BFT), based on a pedigree-based relationship matrix (A) or a genomic-pedigree joint relationship matrix (H). In the reaction norm model, the herd-year-season effect was measured as the environmental variable (EV). Our results showed no G×E interactions for AGE, but for BFT. For both AGE and BFT, the genomic reaction norm model (H) produced more accurate predictions than the conventional reaction norm model (A). For BFT, the accuracies were greater based on the reaction norm model than those based on the reduced model without exploiting G×E interaction, with EV ranging from 0.5 to 1, and accuracy increasing by 3.9% and 4.6% in the reaction norm model based on A and H matrices, respectively, while reaction norm model yielded approximately 8.4% and 7.9% lower accuracy for EVs ranging from 0 to 0.4, based on A and H matrices, respectively. In addition, for BFT, the highest accuracy was obtained in the BJLM6 farm for realizing directional selection. This study will help to apply G×E interactions to practical genomic selection.

Effect of genomic selection and genotyping strategy on estimation of variance components in animal models using different relationship matrices

BackgroundThe traditional way to estimate variance components (VC) is based on the animal model using a pedigree-based relationship matrix (A) (A-AM). After genomic selection was introduced into breeding programs, it was anticipated that VC estimates from A-AM would be biased because the effect of selection based on genomic information is not captured. The single-step method (H-AM), which uses an H matrix as (co)variance matrix, can be used as an alternative to estimate VC. Here, we compared VC estimates from A-AM and H-AM and investigated the effect of genomic selection, genotyping strategy and genotyping proportion on the estimation of VC from the two methods, by analyzing a dataset from a commercial broiler line and a simulated dataset that mimicked the broiler population.ResultsVC estimates from H-AM were severely overestimated with a high proportion of selective genotyping, and overestimation increased as proportion of genotyping increased in the analysis of both commercial and simulated data. This bias in H-AM estimates arises when selective genotyping is used to construct the H-matrix, regardless of whether selective genotyping is applied or not in the selection process. For simulated populations under genomic selection, estimates of genetic variance from A-AM were also significantly overestimated when the effect of genomic selection was strong. Our results suggest that VC estimates from H-AM under random genotyping have the expected values. Predicted breeding values from H-AM were inflated when VC estimates were biased, and inflation differed between genotyped and ungenotyped animals, which can lead to suboptimal selection decisions.ConclusionsWe conclude that VC estimates from H-AM are biased with selective genotyping, but are close to expected values with random genotyping.VC estimates from A-AM in populations under genomic selection are also biased but to a much lesser degree. Therefore, we recommend the use of H-AM with random genotyping to estimate VC for populations under genomic selection. Our results indicate that it is still possible to use selective genotyping in selection, but then VC estimation should avoid the use of genotypes from one side only of the distribution of phenotypes. Hence, a dual genotyping strategy may be needed to address both selection and VC estimation.

Evaluation of the utility of genomic information to improve genetic evaluation of feed efficiency traits of the Pacific white shrimp Litopenaeus vannamei

This study evaluated the utility of genomic information to improve genetic evaluation of feed efficiency traits of the Pacific white shrimp Litopenaeus vannamei. Three methods including the pedigree-based best linear unbiased prediction (BLUP), genomic best linear unbiased prediction (GBLUP) and single-step GBLUP (ssGBLUP) were compared for the prediction accuracy of feed efficiency ratio (FER) and residual feed intake (RFI), as well as a growth trait, average daily weight gain (ADG). The dataset used for analysis consisted of 474 offspring from 34 full-sib families, with a pedigree of three generations, and all the offspring were genotyped on 35,869 SNPs. Two univariate animal models were used, one with the common environmental effect (Model 1) and one without (Model 2). The genomic and pedigree-genomic relationship matrices based heritability estimates of RFI, FER and ADG by Model 1 ranged from 0.187 ± 0.156 to 0.388 ± 0.167, with the common environmental effect in the range of 0.057 ± 0.080 to 0.283 ± 0.108, but it was unable to estimate both additive genetic and common environmental effects simultaneously using the pedigree-based relationship matrix. Across the three matrices, heritability estimates of RFI, FER and ADG by Model 2 ranged from 0.476 ± 0.101 to 0.796 ± 0.146, higher than the corresponding ones using Model 1. Cross-validation was used to evaluate the prediction accuracy of breeding values. For GBLUP and ssGBLUP, genomic prediction accuracy using Model 1 was significantly increased by 28.6–29.0% for RFI, 11.6–12.5% for FER and 62.2–69.4% for ADG, compared with using Model 2, indicative of the advantage of Model 1 over Model 2. When using Model 2, genomic predictions were more accurate than the pedigree-based predictions, with an increase of accuracy ranging from 8.9% to 13.3% across the three traits. Furthermore, accuracy of GBLUP was slightly higher than that of ssGBLUP. The results indicate that the genomic-based methods, especially GBLUP, combined with an animal model containing the common environmental effect, have obvious advantage in estimating variance components and predicting breeding values for feed efficiency traits as well as growth of L.vannamei compared with traditional BLUP.

Effects of ignoring inbreeding in model-based accuracy for BLUP and SSGBLUP.

Model-based accuracy, defined as the theoretical correlation between true and estimated breeding value, can be obtained for each individual as a function of its prediction error variance (PEV) and inbreeding coefficient F, in BLUP, GBLUP and SSGBLUP genetic evaluations. However, for computational convenience, inbreeding is often ignored in two places. First, in the computation of reliability=1-PEV/(1+F). Second, in the set-up, using Henderson's rules, of the inverse of the pedigree-based relationship matrix A. Both approximations have an effect in the computation of model-based accuracy and result in wrong values. In this work, first we present a reminder of the theory and extend it to SSGBLUP. Second, we quantify the error of ignoring inbreeding with real data in three scenarios: BLUP evaluation and SSGBLUP in Uruguayan dairy cattle, and BLUP evaluations in a line of rabbit closed for >40 generations with steady increase of inbreeding up to an average of 0.30. We show that ignoring inbreeding in the set-up of the A-inverse is equivalent to assume that non-inbred animals are actually inbred. This results in an increase of apparent PEV that is negligible for dairy cattle but considerable for rabbit. Ignoring inbreeding in reliability=1-PEV/(1+F) leads to underestimation of reliability for BLUP evaluations, and this underestimation is very large for rabbit. For SSGBLUP in dairy cattle, it leads to both underestimation and overestimation of reliability, both for genotyped and non-genotyped animals. We strongly recommend to include inbreeding both in the set-up of A-inverse and in the computation of reliability from PEVs.

Performance of pedigree and various forms of marker‐derived relationship coefficients in genomic prediction and their correlations

In recent years, with development and validation of different genotyping panels, several methods have been proposed to build efficient similarity matrices among individuals to be used for genomic selection. Consequently, the estimated genetic parameters from such information may deviate from their counterpart using traditional family information. In this study, we used a pedigree-based numerator relationship matrix (A) and three types of marker-based relationship matrices ( ) including two identical by descent, that is and and one identical by state, as well as four Gaussian kernel ( ) similarity kernels with different smoothing parameters to predict yet to be observed phenotypes. Also, we used different kinship matrices that are a linear combination of marker-derived IBD or IBS matrices with A, constructed as , where the weight ( ) assigned to each source of information varied over a grid of values. A Bayesian multiple-trait Gaussian model was fitted to estimate the genetic parameters and compare the prediction accuracy in terms of predictive correlation, mean square error and unbiasedness. Results show that the estimated genetic parameters (heritability and correlations) are affected by the source of the information used to create kinship or the weight placed on the sources of genomic and pedigree information. The superiority of GK-based model depends on the smoothing parameters (θ) so that with an optimum θ value, the GK-based model statistically yielded better performance (higher predictive correlation, lowest MSE and unbiased estimates) and more stable correlations and heritability than the model with IBD, IBS or kinship matrices or any of the linear combinations.