Showing papers by "Roel F. Veerkamp published in 2014"

Whole-genome sequencing of 234 bulls facilitates mapping of monogenic and complex traits in cattle

Abstract: The 1000 bull genomes project supports the goal of accelerating the rates of genetic gain in domestic cattle while at the same time considering animal health and welfare by providing the annotated sequence variants and genotypes of key ancestor bulls. In the first phase of the 1000 bull genomes project, we sequenced the whole genomes of 234 cattle to an average of 8.3-fold coverage. This sequencing includes data for 129 individuals from the global Holstein-Friesian population, 43 individuals from the Fleckvieh breed and 15 individuals from the Jersey breed. We identified a total of 28.3 million variants, with an average of 1.44 heterozygous sites per kilobase for each individual. We demonstrate the use of this database in identifying a recessive mutation underlying embryonic death and a dominant mutation underlying lethal chrondrodysplasia. We also performed genome-wide association studies for milk production and curly coat, using imputed sequence variants, and identified variants associated with these traits in cattle.

Accuracy of imputation to whole-genome sequence data in Holstein Friesian cattle

Abstract: Background: The use of whole-genome sequence data can lead to higher accuracy in genome-wide association studies and genomic predictions. However, to benefit from whole-genome sequence data, a large dataset of sequenced individuals is needed. Imputation from SNP panels, such as the Illumina BovineSNP50 BeadChip and Illumina BovineHD BeadChip, to whole-genome sequence data is an attractive and less expensive approach to obtain whole-genome sequence genotypes for a large number of individuals than sequencing all individuals. Our objective was to investigate accuracy of imputation from lower density SNP panels to whole-genome sequence data in a typical dataset for cattle. Methods: Whole-genome sequence data of chromosome 1 (1737 471 SNPs) for 114 Holstein Friesian bulls were used. Beagle software was used for imputation from the BovineSNP50 (3132 SNPs) and BovineHD (40 492 SNPs) beadchips. Accuracy was calculated as the correlation between observed and imputed genotypes and assessed by five-fold cross-validation. Three scenarios S40, S60 and S80 with respectively 40%, 60%, and 80% of the individuals as reference individuals were investigated. Results: Mean accuracies of imputation per SNP from the BovineHD panel to sequence data and from the BovineSNP50 panel to sequence data for scenarios S40 and S80 ranged from 0.77 to 0.83 and from 0.37 to 0.46, respectively. Stepwise imputation from the BovineSNP50 to BovineHD panel and then to sequence data for scenario S40 improved accuracy per SNP to 0.65 but it varied considerably between SNPs. Conclusions: Accuracy of imputation to whole-genome sequence data was generally high for imputation from the BovineHD beadchip, but was low from the BovineSNP50 beadchip. Stepwise imputation from the BovineSNP50 to the BovineHD beadchip and then to sequence data substantially improved accuracy of imputation. SNPs with a low minor allele frequency were more difficult to impute correctly and the reliability of imputation varied more. Linkage disequilibrium between an imputed SNP and the SNP on the lower density panel, minor allele frequency of the imputed SNP and size of the reference group affected imputation reliability.

Genomic selection for feed efficiency in dairy cattle

Abstract: Feed is a major component of variable costs associated with dairy systems and is therefore an important consideration for breeding objectives. As a result, measures of feed efficiency are becoming popular traits for genetic analyses. Already, several countries account for feed efficiency in their breeding objectives by approximating the amount of energy required for milk production, maintenance, etc. However, variation in actual feed intake is currently not captured in dairy selection objectives, although this could be possible by evaluating traits such as residual feed intake (RFI), defined as the difference between actual and predicted feed (or energy) intake. As feed intake is expensive to accurately measure on large numbers of cows, phenotypes derived from it are obvious candidates for genomic selection provided that: (1) the trait is heritable; (2) the reliability of genomic predictions are acceptable to those using the breeding values; and (3) if breeding values are estimated for heifers, rather than cows then the heifer and cow traits need to be correlated. The accuracy of genomic prediction of dry matter intake (DMI) and RFI has been estimated to be around 0.4 in beef and dairy cattle studies. There are opportunities to increase the accuracy of prediction, for example, pooling data from three research herds (in Australia and Europe) has been shown to increase the accuracy of genomic prediction of DMI from 0.33 within country to 0.35 using a three-country reference population. Before including RFI as a selection objective, genetic correlations with other traits need to be estimated. Weak unfavourable genetic correlations between RFI and fertility have been published. This could be because RFI is mathematically similar to the calculation of energy balance and failure to account for mobilisation of body reserves correctly may result in selection for a trait that is similar to selecting for reduced (or negative) energy balance. So, if RFI is to become a selection objective, then including it in an overall multi-trait selection index where the breeding objective is net profit is sensible, as this would allow genetic correlations with other traits to be properly accounted for. If genetic parameters are accurately estimated then RFI is a logical breeding objective. If there is uncertainty in these, then DMI may be preferable.

Evaluation of measures of correctness of genotype imputation in the context of genomic prediction: a review of livestock applications.

Abstract: In livestock, many studies have reported the results of imputation to 50k single nucleotide polymorphism (SNP) genotypes for animals that are genotyped with low-density SNP panels. The objective of this paper is to review different measures of correctness of imputation, and to evaluate their utility depending on the purpose of the imputed genotypes. Across studies, imputation accuracy, computed as the correlation between true and imputed genotypes, and imputation error rates, that counts the number of incorrectly imputed alleles, are commonly used measures of imputation correctness. Based on the nature of both measures and results reported in the literature, imputation accuracy appears to be a more useful measure of the correctness of imputation than imputation error rates, because imputation accuracy does not depend on minor allele frequency (MAF), whereas imputation error rate depends on MAF. Therefore imputation accuracy can be better compared across loci with different MAF. Imputation accuracy depends on the ability of identifying the correct haplotype of a SNP, but many other factors have been identified as well, including the number of genotyped immediate ancestors, the number of animals with genotypes at the high-density panel, the SNP density on the low- and high-density panel, the MAF of the imputed SNP and whether imputed SNP are located at the end of a chromosome or not. Some of these factors directly contribute to the linkage disequilibrium between imputed SNP and SNP on the low-density panel. When imputation accuracy is assessed as a predictor for the accuracy of subsequent genomic prediction, we recommend that: (1) individual-specific imputation accuracies should be used that are computed after centring and scaling both true and imputed genotypes; and (2) imputation of gene dosage is preferred over imputation of the most likely genotype, as this increases accuracy and reduces bias of the imputed genotypes and the subsequent genomic predictions.

Genomic prediction from whole genome sequence in livestock: the 1000 Bull Genomes Project

Abstract: Advantages of using whole genome sequence data to predict genomic estimated breeding values (GEBV) include better persistence of accuracy of GEBV across generations and more accurate GEBV across breeds. The 1000 Bull Genomes Project provides a database of whole genome sequenced key ancestor bulls, that can be used for imputing sequence variant genotypes into reference sets for genomic prediction. Run 3.0 included 429 sequences, with 31.8 million variants detected. BayesRC, a new method for genomic prediction, addresses some of the challenges associated with using the sequence data, and takes advantage of biological information. In a dairy data set, predictions using BayesRC and imputed sequence data from the 1000 Bull Genomes data were 2% more accurate than with 800k data, and we could demonstrate the method was able to identify causal mutations in some cases. Further improvements will come from more accurate imputation of sequence variant genotypes and improved biological information.

Consequences of splitting whole-genome sequencing effort over multiple breeds on imputation accuracy

Abstract: The aim of this study was to determine the consequences of splitting sequencing effort over multiple breeds for imputation accuracy from a high-density SNP chip towards whole-genome sequence. Such information would assist for instance numerical smaller cattle breeds, but also pig and chicken breeders, who have to choose wisely how to spend their sequencing efforts over all the breeds or lines they evaluate. Sequence data from cattle breeds was used, because there are currently relatively many individuals from several breeds sequenced within the 1,000 Bull Genomes project. The advantage of whole-genome sequence data is that it carries the causal mutations, but the question is whether it is possible to impute the causal variants accurately. This study therefore focussed on imputation accuracy of variants with low minor allele frequency and breed specific variants. Imputation accuracy was assessed for chromosome 1 and 29 as the correlation between observed and imputed genotypes. For chromosome 1, the average imputation accuracy was 0.70 with a reference population of 20 Holstein, and increased to 0.83 when the reference population was increased by including 3 other dairy breeds with 20 animals each. When the same amount of animals from the Holstein breed were added the accuracy improved to 0.88, while adding the 3 other breeds to the reference population of 80 Holstein improved the average imputation accuracy marginally to 0.89. For chromosome 29, the average imputation accuracy was lower. Some variants benefitted from the inclusion of other breeds in the reference population, initially determined by the MAF of the variant in each breed, but even Holstein specific variants did gain imputation accuracy from the multi-breed reference population. This study shows that splitting sequencing effort over multiple breeds and combining the reference populations is a good strategy for imputation from high-density SNP panels towards whole-genome sequence when reference populations are small and sequencing effort is limiting. When sequencing effort is limiting and interest lays in multiple breeds or lines this provides imputation of each breed.

Imputation of non-genotyped individuals based on genotyped relatives: assessing the imputation accuracy of a real case scenario in dairy cattle

Abstract: Imputation of genotypes for ungenotyped individuals could enable the use of valuable phenotypes created before the genomic era in analyses that require genotypes. The objective of this study was to investigate the accuracy of imputation of non-genotyped individuals using genotype information from relatives. Genotypes were simulated for all individuals in the pedigree of a real (historical) dataset of phenotyped dairy cows and with part of the pedigree genotyped. The software AlphaImpute was used for imputation in its standard settings but also without phasing, i.e. using basic inheritance rules and segregation analysis only. Different scenarios were evaluated i.e.: (1) the real data scenario, (2) addition of genotypes of sires and maternal grandsires of the ungenotyped individuals, and (3) addition of one, two, or four genotyped offspring of the ungenotyped individuals to the reference population. The imputation accuracy using AlphaImpute in its standard settings was lower than without phasing. Including genotypes of sires and maternal grandsires in the reference population improved imputation accuracy, i.e. the correlation of the true genotypes with the imputed genotype dosages, corrected for mean gene content, across all animals increased from 0.47 (real situation) to 0.60. Including one, two and four genotyped offspring increased the accuracy of imputation across all animals from 0.57 (no offspring) to 0.73, 0.82, and 0.92, respectively. At present, the use of basic inheritance rules and segregation analysis appears to be the best imputation method for ungenotyped individuals. Comparison of our empirical animal-specific imputation accuracies to predictions based on selection index theory suggested that not correcting for mean gene content considerably overestimates the true accuracy. Imputation of ungenotyped individuals can help to include valuable phenotypes for genome-wide association studies or for genomic prediction, especially when the ungenotyped individuals have genotyped offspring.

Genomic prediction of breeding values using previously estimated SNP variances

Abstract: Genomic prediction requires estimation of variances of effects of single nucleotide polymorphisms (SNPs), which is computationally demanding, and uses these variances for prediction. We have developed models with separate estimation of SNP variances, which can be applied infrequently, and genomic prediction, which can be applied routinely. SNP variances were estimated with Bayes Stochastic Search Variable Selection (BSSVS) and BayesC. Genome-enhanced breeding values (GEBV) were estimated with RR-BLUP (ridge regression best linear unbiased prediction), using either variances obtained from BSSVS (BLUP-SSVS) or BayesC (BLUP-C), or assuming equal variances for each SNP. Datasets used to estimate SNP variances comprised (1) all animals, (2) 50% random animals (RAN50), (3) 50% best animals (TOP50), or (4) 50% worst animals (BOT50). Traits analysed were protein yield, udder depth, somatic cell score, interval between first and last insemination, direct longevity, and longevity including information from predictors. BLUP-SSVS and BLUP-C yielded similar GEBV as the equivalent Bayesian models that simultaneously estimated SNP variances. Reliabilities of these GEBV were consistently higher than from RR-BLUP, although only significantly for direct longevity. Across scenarios that used data subsets to estimate GEBV, observed reliabilities were generally higher for TOP50 than for RAN50, and much higher than for BOT50. Reliabilities of TOP50 were higher because the training data contained more ancestors of selection candidates. Using estimated SNP variances based on random or non-random subsets of the data, while using all data to estimate GEBV, did not affect reliabilities of the BLUP models. A convergence criterion of 10−8 instead of 10−10 for BLUP models yielded similar GEBV, while the required number of iterations decreased by 71 to 90%. Including a separate polygenic effect consistently improved reliabilities of the GEBV, but also substantially increased the required number of iterations to reach convergence with RR-BLUP. SNP variances converged faster for BayesC than for BSSVS. Combining Bayesian variable selection models to re-estimate SNP variances and BLUP models that use those SNP variances, yields GEBV that are similar to those from full Bayesian models. Moreover, these combined models yield predictions with higher reliability and less bias than the commonly used RR-BLUP model.

Genetic and genomic solutions to improve feed efficiency and reduce environmental impact of dairy cattle

Abstract: Traits related to resource use efficiency are dry matter intake (DMI), residual feed intake (RFI) and methane (CH4) emission. In an experimental dataset of 588 heifers, we showed that it is possible to decrease CH4 emission (predicted from DMI and ration composition) by selecting more efficient cows. Resource use efficiency phenotypes are difficult and expensive to measure, but genomic selection is a promising tool to enable selection for resource efficient cows. Using genomic selection, a reduction in predicted CH4 (g/d) of 15% in 10 years is theoretically possible. For DMI, an international collaboration between 9 countries in Europe, US and Australiasia has been established to assemble DMI data on >6,000 cows with phenotypes and genotypes. With all these developments, genetic selection is likely to make a major contribution to improving resource use efficiency, as long as feeding and management are adapted accordingly.

Consequences for diversity when animals are prioritized for conservation of the whole genome or of one specific allele

Abstract: When animals are selected for one specific allele, for example for inclusion in a gene bank, this may result in the loss of diversity in other parts of the genome. The aim of this study was to quantify the risk of losing diversity across the genome when targeting a single allele for conservation when storing animals in a gene bank. From a small Holstein population, genotyped for 54 001 SNP loci, animals were prioritized for a single allele while maximizing the genomewide diversity using optimal contribution selection. Selection for a single allele was done for five different target frequencies: (i) no restriction on a target frequency; (ii) target frequency = original frequency in population; (iii) target frequency = 0.50; (iv) target frequency of the major allele = 1 (fixation); and (v) target frequency of the major allele = 0 (elimination). To do this, optimal contribution selection was extended with an extra constraint on the allele frequency of the target SNP marker. Results showed that elimination or fixation of alleles can result in substantial losses in genetic diversity around the targeted locus and also across the rest of the genome, depending on the allele frequency and the target frequency. It was concluded that losses of genetic diversity around the target allele are the largest when the target frequency is very different from the current allele frequency.

Genetic Parameters and Accuracy of recording Dry Matter Intake in first parity Holstein-Friesian cows

Abstract: A total of 30,483 records for dry matter intake (DMI) recorded between 1990 and 2011 were available from 1,273 Dutch Holstein-Friesian first-parity cows. Genetic parameters were estimated for the first 324 days in milk (DIM) using random regression models. Estimated heritabilities for DMI ranged from 0.21 to 0.40 across lactation, being highest between 80 and 205 DIM. Entire lactation heritability was 0.46. Genetic correlations between DMI in early lactation and during the rest of the lactation were negative (-0.5), and were most positive between DMI in mid lactation and during the rest of the lactation (0.80). Dry matter intake was genetically different across lactation, especially in early and mid lactation. Accuracies of selection for the entire lactation for DMI were 0.58, 0.47 and 0.33, when it was measured for 15, 10 or 5 weeks, respectively and best recording periods were during mid and late lactation.

Breeding Value for Dry Matter Intake for Dutch Bulls based on DGV for DMI and BV for Predictors

Abstract: Dry matter intake (DMI) was available for (part of) 3,179 lactations in the Netherlands. Using a fixed regression test-day model, genetic variances were estimated for yield and DMI in parity 1, 2 and 3+, and stature, body depth and chest width in parity 1. The combined pedigree and genomic relationship matrix was used to predict genomic breeding values (DGV) for DMI and used to back solve the solutions to obtain SNP effects. Using genetic correlations with predictor traits, combined breeding values (GEBV) for DMI were produced. The s.d. of the combined GEBV was 1.2 kg/d and the median for the reliability was 0.56. For DGV s.d. was 0.45 and median reliability was 0.18. Genetic trend showed an increase of 1.1 kg DMI/d per decade for the combined GEBV, versus 0.12 for the DGV. Future developments include the expansion of the reference population in collaboration with other partners worldwide.

A comparison of principal component regression and genomic REML for genomic prediction across populations

Abstract: Genomic prediction faces two main statistical problems: multicollinearity and n ≪ p (many fewer observations than predictor variables). Principal component (PC) analysis is a multivariate statistical method that is often used to address these problems. The objective of this study was to compare the performance of PC regression (PCR) for genomic prediction with that of a commonly used REML model with a genomic relationship matrix (GREML) and to investigate the full potential of PCR for genomic prediction. The PCR model used either a common or a semi-supervised approach, where PC were selected based either on their eigenvalues (i.e. proportion of variance explained by SNP (single nucleotide polymorphism) genotypes) or on their association with phenotypic variance in the reference population (i.e. the regression sum of squares contribution). Cross-validation within the reference population was used to select the optimum PCR model that minimizes mean squared error. Pre-corrected average daily milk, fat and protein yields of 1609 first lactation Holstein heifers, from Ireland, UK, the Netherlands and Sweden, which were genotyped with 50 k SNPs, were analysed. Each testing subset included animals from only one country, or from only one selection line for the UK. In general, accuracies of GREML and PCR were similar but GREML slightly outperformed PCR. Inclusion of genotyping information of validation animals into model training (semi-supervised PCR), did not result in more accurate genomic predictions. The highest achievable PCR accuracies were obtained across a wide range of numbers of PC fitted in the regression (from one to more than 1000), across test populations and traits. Using cross-validation within the reference population to derive the number of PC, yielded substantially lower accuracies than the highest achievable accuracies obtained across all possible numbers of PC. On average, PCR performed only slightly less well than GREML. When the optimal number of PC was determined based on realized accuracy in the testing population, PCR showed a higher potential in terms of achievable accuracy that was not capitalized when PC selection was based on cross-validation. A standard approach for selecting the optimal set of PC in PCR remains a challenge.

(A)cross-breed Genomic Prediction

Abstract: Genomic prediction holds the promise to use information of other populations to improve prediction accuracy. Thus far, empirical evaluations showed limited benefit of multi-breed compared to single reed genomic prediction. We compared prediction accuracy of different models based on two losely related and one unrelated line of layer chickens. Multi-breed genomic prediction may be successful when lines are closely related, and when the number of training animals of the additional line is large compared to the line itself. Multi-breed genomic prediction requires models that are lexible enough to use beneficial and ignore detrimental sources of information in the training data. Combining linear and non-linear models may lead to small increases in accuracy of multibreed genomic prediction. Multitrait models, modelling a separate trait for each breed, appear especially beneficial when elationships between breeds are very low, or when the genetic correlation between breeds is negative.

Genetic Architecture and Biological Basis of Feed Efficiency in Dairy Cattle

Abstract: The genetic architecture of residual feed intake (RFI) and related traits was evaluated using a dataset of 2,894 cows. A Bayesian analysis estimated that markers accounted for 14% of the variance in RFI, and that RFI had considerable genetic variation. Effects of marker windows were small, but QTL peaks were identified. Six of the 8 chromosomes harboring QTL influencing RFI did not contain QTL influencing dry matter intake (DMI), net energy for lactation, or metabolic body weight. In contrast, 7 of 9 chromosomes with QTL influencing DMI also harbored QTL for one or more of the other traits evaluated. These results represent the first genomic analysis of RFI using a large (~3,000 animals) international dataset. In general they suggest RFI is a trait that should respond to selection, and that its genetic regulation is different from that of DMI.

Genetic Analysis of Longevity in Dutch Dairy Cattle using Random Regression

Abstract: Longevity, productive life, or lifespan of dairy cattle is an important trait for dairy farmers, and it is defined as the time from first calving to the last test date for milk production. Methods for genetic evaluations need to account for censored data; that is, records from cows that are still alive. The aim of this study was to investigate whether these methods also need to take account of survival being genetically a different trait across the entire lifespan of a cow. The data set comprised 112,000 cows with a total of 3,964,449 observations for survival per month from first calving until 72 mo in productive life. A random regression model with second-order Legendre polynomials was fitted for the additive genetic effect. Alternative parameterizations were (1) different trait definitions for the length of time interval for survival after first calving (1, 3, 6, and 12 mo); (2) linear or threshold model; and (3) differing the order of the Legendre polynomial. The partial derivatives of a profit function were used to transform variance components on the survival scale to those for lifespan. Survival rates were higher in early life than later in life (99 vs. 95%). When survival was defined over 12-mo intervals survival curves were smooth compared with curves when 1-, 3-, or 6-mo intervals were used. Heritabilities in each interval were very low and ranged from 0.002 to 0.031, but the heritability for lifespan over the entire period of 72 mo after first calving ranged from 0.115 to 0.149. Genetic correlations between time intervals ranged from 0.25 to 1.00. Genetic parameters and breeding values for the genetic effect were more sensitive to the trait definition than to whether a linear or threshold model was used or to the order of Legendre polynomial used. Cumulative survival up to the first 6 mo predicted lifespan with an accuracy of only 0.79 to 0.85; that is, reliability of breeding value with many daughters in the first 6 mo can be, at most, 0.62 to 0.72, and changes of breeding values are still expected when daughters are getting older. Therefore, an improved model for genetic evaluation should treat survival as different traits during the lifespan by splitting lifespan in time intervals of 6 mo or less to avoid overestimated reliabilities and changes in breeding values when daughters are getting older.

Genetics and genomics of fertility in dairy cows

Abstract: GENETICS OF FERTILITY The existence of significant genetic variation in fertility is generally accepted (Pryce and Veerkamp 2001, Flint 2006, Rydhmer and Berglund 2006). Heritability for fertility traits commonly used in animal breeding is relatively low, due to the large unexplainable residual variation in statistical models trying to predict traits like calving interval and pregnancy rate at the individual cow level. The existence of strong genetic effects are evidenced by differences in mean calving interval of up to 30 days between daughters of different sires. Similarly, the difference in pregnancy rate between daughters of extreme Holstein sires is as high as 7%, which equates to roughly 29 days open per lactation (Weigel 2006). There is also overwhelming evidence that increasing genetic merit for yield without considering genetic merit for fertility, reduces fertility (Pryce and Veerkamp 2001, Veerkamp and others 2003). Over the previous two decades the interval from calving to conception increased by 24 days in the US (Shook 2006). Illustrative of this trend, Holstein herds in south-eastern states reported increases in average days open of 40d and over between 1982 and 1999, whilst conception rates decreased from about 50% to 34%. Unfavourable genetic changes in conception interval since 1980 accumulated to 1.0 genetic standard deviations and genetics has been estimated to account for onethird of the decline in pregnancy rate (Shook 2006). To date, none of the leading dairy cattle breeding programmes select on yield only, and by using multi-trait selection they have included fertility in their selection indices (Miglior and others 2005). This multi-trait selection has reversed the negative genetic trend in most countries. For example, in the Netherlands (Figure 1) the genetic potential of cows showed a steady decline, e.g. genetic potential for interval calving first insemination increased by 15 days before the millennium, but since the introduction of fertility indices this genetic trend levelled off and is improving again. Similar patterns are observed for the other fertility traits. The most limiting factors for even faster genetic progress for fertility are the genetic association between fertility and yield, the low accuracy of breeding values due to poor recording, and the long generation interval.

Estimation of genetic parameters and breeding values in honey bees

Abstract: Estimation of breeding values and variance components in honey bees is complex due to bees’ reproduction system. It complicates calculation of the numerator relationships matrix (NRM). Mixed model methodology to estimate breeding values in honey bees was developed by Bienefeld (2007) (BER). To invert NRM they used a diagonal matrix of Mendelian sampling terms as an approximation of a more realistic block diagonal matrix. Brascamp and Bijma (2014) developed the necessary algebra for elements in NRM including these blocks and compared breeding value estimation using both methods with the same genetic parameters for both. In this paper we estimate these genetic parameters with ASReml. It appears that BER in particular underestimates the additive genetic variance for worker effect and also estimates the genetic correlation (true value -0.50) closer to zero. For the ranking of breeding animals differences are small.

Genomic Prediction with 12.5 Million SNPs for 5503 Holstein Friesian Bulls

Abstract: This study reports the first preliminary results of genomic prediction with whole-genome sequence data (12,590,056 SNPs) for 5503 bulls with accurate phenotypes. Two methods were compared: genome-enabled best linear unbiased prediction (GBLUP) and a Bayesian approach (BSSVS). Results were compared with results using BovineHD genotypes (631,428 SNPs). Results were reported for somatic cell score, interval between first and last insemination, and protein yield. For all traits, and both methods genomic prediction with sequence data showed similar results compared to BovineHD and GBLUP showed similar results compared to BSSVS. However, it remains to be seen if reliability of BSSVS with sequence data will improve after more sampling cycles have been finished.

Genetic Parameters for Endocrine Fertility Traits from In-line Milk Progesterone Records in Dairy Cows

Abstract: In-line milk progesterone records (n = 163,145) collected from June 2009 through November 2013 for 2,274 lactations of Holstein-Friesian cows in 12 commercial herds in the Netherlands were analyzed for commencement of luteal activity (CLA), luteal activity during first 60 days in milk (LA60), proportion of samples with luteal activity (PLA), interval from commencement of luteal activity to first service, length of first luteal phase, and number of inter-ovulatory intervals before first service. Heritability (0.13, 0.10, and 0.05) and repeatability estimates (0.26, 0.21, and 0.16) were greatest for CLA, PLA and LA60, respectively, compared with other traits. Genetic correlations were 0.96 to 0.99 between these traits. This study shows that in-line P4 records can be used to define and explore several heritable endocrine fertility traits that can be used in genetic improvement of fertility by selection.

Consequences of Splitting Sequencing Effort over Multiple Breeds on Imputation Accuracy

Abstract: Imputation from a high-density SNP panel (777k) to whole-genome sequence with a reference population of 20 Holstein resulted in an average imputation accuracy of 0.70, and increased to 0.83 when the reference population was increased by including 3 other dairy breeds with 20 animals each. When the same amount of animals from the Holstein breed were added the accuracy improved to 0.88. Imputation of variants with very low minor allele frequency in Holstein that were also segregating in the mixed breed reference population benefitted from the inclusion of other breeds in the reference population, whereas Holstein specific variants benefitted from the large Holstein reference population. This study shows that splitting sequencing effort over multiple breeds is a good strategy for imputation from high-density SNP panels towards whole-genome sequence when reference populations are small

Simultaneous deregression of cow and bull breeding values

Abstract: The next step to increase the accuracy of genomic prediction is to extend reference populations with cows next to daughter proven bulls. Cows typically have estimated breeding values (EBV) with considerably lower reliabilities compared to bulls. This suggests that commonly used (approximate) deregression procedures for bulls may not be appropriate for cows. The objective of this study was to test an alternative approach to simultaneously de-regress EBV of cows and bulls, and to derive appropriate weights for those de-regressed EBV. First, the appropriate weights of the de-regressed EBV were derived, and then the de-regressed EBV were computed using those weights. The analyses showed that the methods were well able to accurately de-regress EBV and compute their weights, both for bulls and cows. Despite observed discrepancies between intermediate results and simulated values, final EBV and reliabilities correlated very well with original values.