Factor analysis of evaluated and linearly scored traits in Swedish Warmblood horses.

Assessment protocols to describe the various aspects of conformation, gait and jumping traits on a linear scale were introduced at young horse tests for Swedish Warmblood horses in 2013. The traits scored on a linear scale are assumed to be less subjective and more easily compared across populations than the traditional evaluated traits that are scored relative to the breeding goal. However, the resulting number of traits is considerable, and several of the traits are correlated. The aim of this study was to investigate the interrelationship between the different evaluated and linearly scored traits in Swedish Warmbloods using factor analysis. In total, 20,935 horses born 1996-2017 had information on evaluated traits, and 5450 of these also had linearly scored trait records assessed since 2014 when the protocol was updated. A factor analysis with varimax rotation was performed separately for evaluated and linearly scored traits using the Psych package in R. Height at withers was included in both analyses. A total of four factors for evaluated traits and 14 factors for linearly scored traits were kept for further analysis. Missing values for individual traits in horses with linearly scored trait records were imputed based on correlated traits before factor scores were calculated using factor loadings. Genetic parameters for, and correlations between, the resulting underlying factors were estimated using multiple-trait animal models in the BLUPF90 package. Heritability estimates were on a similar level as for the traits currently used in the genetic evaluation, ranging from 0.05 for the factor for linearly scored traits named L.behaviour (dominated by traits related to behaviour) to 0.59 for the factor for evaluated traits named E.size (dominated by height at withers and conformation). For both types of traits, separate factors were formed for jumping and gait traits, as well as for body size. High genetic correlations were estimated between such corresponding factors for evaluated traits and factors for linearly scored traits. In conclusion, factor analysis could be used to reduce the number of traits to be included in multiple-trait genetic evaluation or in genomic analysis for warmblood horses. It can also contribute to a better understanding of the interrelationships among the assessed traits and be useful to decide on subgroups of traits to be used in several multiple-trait evaluations on groups of original traits.

Genetic evaluation of claw health traits accounting for potential preselection of cows to be trimmed.

Claw lesions are one of the most important health issues in dairy cattle. Although the frequency of claw lesions depends greatly on herd management, the frequency can be lowered through genetic selection. A genetic evaluation could be developed based on trimming records collected by claw trimmers; however, not all cows present in a herd are usually selected by the breeder to be trimmed. The objectives of this study were to investigate the importance of the preselection of cows for trimming, to account for this preselection, and to estimate genetic parameters of claw health traits. The final data set contained 25,511 trimming records of French Holstein cows. Analyzed claw lesion traits were digital dermatitis, heel horn erosion, interdigital hyperplasia, sole hemorrhage circumscribed, sole hemorrhage diffused, sole ulcer, and white line fissure. All traits were analyzed as binary traits in a multitrait linear animal model. Three scenarios were considered: including only trimmed cows in a 7-trait model (scenario 1); or trimmed cows and contemporary cows not trimmed but present at the time of a visit (considering that nontrimmed cows were healthy) in a 7-trait model (scenario 2); or trimmed cows and contemporary cows not trimmed but present at the time of a visit (considering lesion records for trimmed cows only), in an 8-trait model, including a 0/1 trimming status trait (scenario 3). For scenario 3, heritability estimates ranged from 0.02 to 0.09 on the observed scale. Genetic correlations clearly revealed 2 groups of traits (digital dermatitis, heel horn erosion, and interdigital hyperplasia on the one hand, and sole hemorrhage circumscribed, sole hemorrhage diffused, sole ulcer, and white line fissure on the other hand). Heritabilities on the underlying scale did not vary much depending on the scenario: the effect of the preselection of cows for trimming on the estimation of heritabilities appeared to be negligible. However, including untrimmed cows as healthy caused bias in the estimation of genetic correlations. The use of a trimming status trait to account for preselection appears promising, as it allows consideration of the exhaustive population of cows present at the time a trimmer visited a farm without causing bias in genetic parameters.

A novel generalized ridge regression method for quantitative genetics.

As the molecular marker density grows, there is a strong need in both genome-wide association studies and genomic selection to fit models with a large number of parameters. Here we present a computationally efficient generalized ridge regression (RR) algorithm for situations in which the number of parameters largely exceeds the number of observations. The computationally demanding parts of the method depend mainly on the number of observations and not the number of parameters. The algorithm was implemented in the R package bigRR based on the previously developed package hglm. Using such an approach, a heteroscedastic effects model (HEM) was also developed, implemented, and tested. The efficiency for different data sizes were evaluated via simulation. The method was tested for a bacteria-hypersensitive trait in a publicly available Arabidopsis data set including 84 inbred lines and 216,130 SNPs. The computation of all the SNP effects required <10 sec using a single 2.7-GHz core. The advantage in run time makes permutation test feasible for such a whole-genome model, so that a genome-wide significance threshold can be obtained. HEM was found to be more robust than ordinary RR (a.k.a. SNP-best linear unbiased prediction) in terms of QTL mapping, because SNP-specific shrinkage was applied instead of a common shrinkage. The proposed algorithm was also assessed for genomic evaluation and was shown to give better predictions than ordinary RR.

Relative impact of alpha-tocopherol, copper and fatty acid composition on the occurrence of oxidized milk flavour.

The oxidative process in milk fat, resulting in spontaneous oxidized off-flavour (SOF), is commonly assumed to depend on contents of pro- and antioxidants in milk and availability of fatty acids acting as their substrate. An important antioxidant in milk is alpha-tocopherol whereas the most potent prooxidant is the metal ion copper. The separate effects of alpha-tocopherol, copper, and milk fatty acid profile, and their combined effect on SOF development were examined in milk from 44 multiparous cows fed different roughage types and different amounts of dietary, unsaturated fat. A clear association between concentrations of copper and poly-unsaturated fatty acids in milk and the risk for developing SOF was found. Heritability estimates suggest that occurrence of SOF is partly under genetic control which indicates that milk quality may be compromised if breeding bulls are selected that carry genotypes predisposing for milk prone to develop SOF.

Genetic heterogeneity of residual variance - estimation of variance components using double hierarchical generalized linear models.

BACKGROUND: The sensitivity to microenvironmental changes varies among animals and may be under genetic control. It is essential to take this element into account when aiming at breeding robust farm animals. Here, linear mixed models with genetic effects in the residual variance part of the model can be used. Such models have previously been fitted using EM and MCMC algorithms. RESULTS: We propose the use of double hierarchical generalized linear models (DHGLM), where the squared residuals are assumed to be gamma distributed and the residual variance is fitted using a generalized linear model. The algorithm iterates between two sets of mixed model equations, one on the level of observations and one on the level of variances. The method was validated using simulations and also by re-analyzing a data set on pig litter size that was previously analyzed using a Bayesian approach. The pig litter size data contained 10,060 records from 4,149 sows. The DHGLM was implemented using the ASReml software and the algorithm converged within three minutes on a Linux server. The estimates were similar to those previously obtained using Bayesian methodology, especially the variance components in the residual variance part of the model. CONCLUSIONS: We have shown that variance components in the residual variance part of a linear mixed model can be estimated using a DHGLM approach. The method enables analyses of animal models with large numbers of observations. An important future development of the DHGLM methodology is to include the genetic correlation between the random effects in the mean and residual variance parts of the model as a parameter of the DHGLM.

Association between radiographic assessment of hip status and subsequent incidence of veterinary care and mortality related to hip dysplasia in insured Swedish dogs.

Our objective was to evaluate the association between grading of hip status as assessed by radiographic examination (hip screening) and subsequent incidence of veterinary care and mortality related to hip dysplasia (HD) in five breeds of insured dogs in Sweden. Screening results for hip status from the Swedish Kennel Club and data on veterinary care and mortality from the insurance company Agria were merged based on the registration number of the dog. Dogs of five breeds (Bernese Mountain Dogs, German Shepherds, Golden Retrievers, Labrador Retrievers, and Rottweilers) screened during 1995-2004 and covered by an insurance plan for veterinary care or life at the time of screening were included. The study populations included between 1667 and 10,663 dogs per breed. Breed-specific multivariable Cox proportional-hazards analyses were performed to evaluate the impact of radiographic hip status on time from hip screening to first HD-related veterinary and life claim, respectively. The effects of gender, birth season, and a time-varying covariate of year were also studied. Additional analyses, on the five breeds combined, were performed to investigate the effects of hip status, breed, and the interaction between hip status and breed. The effect of hip status was highly significant (P<0.001) for both life and veterinary claims related to HD in all five breeds with increased hazard ratio (HR) for deteriorating hip status. Dogs with moderate or severe hip status at screening had a markedly increased hazard of HD-related veterinary care and mortality compared with dogs assessed as free or mild. The time-varying covariate of year showed a significantly higher HR in the last time period for German Shepherds and Labrador Retrievers in the analyses of veterinary claims. In the analyses on all five breeds, German Shepherds had the highest HR for both veterinary care and mortality related to HD, followed by Bernese Mountain Dogs. Golden and Labrador Retrievers had the lowest HR. The effect of hip status on the hazard was the same irrespective of breed. However, as a consequence of differences between breeds in overall risk, the predictive ability of screening results for subsequent incidence of HD-related problems for individual dogs was breed-dependent. Based on the strong association between radiographic hip status and incidence of HD-related veterinary care and mortality, and the previously reported moderate heritability of hip status, we conclude that selection based on screening results for hip status can be expected to reduce the risk of HD-related clinical problems.

Fuzzy classification of phantom parent groups in an animal model.

BACKGROUND: Genetic evaluation models often include genetic groups to account for unequal genetic level of animals with unknown parentage. The definition of phantom parent groups usually includes a time component (e.g. years). Combining several time periods to ensure sufficiently large groups may create problems since all phantom parents in a group are considered contemporaries. METHODS: To avoid the downside of such distinct classification, a fuzzy logic approach is suggested. A phantom parent can be assigned to several genetic groups, with proportions between zero and one that sum to one. Rules were presented for assigning coefficients to the inverse of the relationship matrix for fuzzy-classified genetic groups. This approach was illustrated with simulated data from ten generations of mass selection. Observations and pedigree records were randomly deleted. Phantom parent groups were defined on the basis of gender and generation number. In one scenario, uncertainty about generation of birth was simulated for some animals with unknown parents. In the distinct classification, one of the two possible generations of birth was randomly chosen to assign phantom parents to genetic groups for animals with simulated uncertainty, whereas the phantom parents were assigned to both possible genetic groups in the fuzzy classification. RESULTS: The empirical prediction error variance (PEV) was somewhat lower for fuzzy-classified genetic groups. The ranking of animals with unknown parents was more correct and less variable across replicates in comparison with distinct genetic groups. In another scenario, each phantom parent was assigned to three groups, one pertaining to its gender, and two pertaining to the first and last generation, with proportion depending on the (true) generation of birth. Due to the lower number of groups, the empirical PEV of breeding values was smaller when genetic groups were fuzzy-classified. CONCLUSION: Fuzzy-classification provides the potential to describe the genetic level of unknown parents in a more parsimonious and structured manner, and thereby increases the precision of predicted breeding values.