Cow genotyping strategies for genomic selection in a small dairy cattle population

This study compares how different cow genotyping strategies increase the accuracy of genomic estimated breeding values (EBV) in dairy cattle breeds with low numbers. In these breeds, few sires have progeny records, and genotyping cows can improve the accuracy of genomic EBV. The Guernsey breed is a small dairy cattle breed with approximately 14,000 recorded individuals worldwide. Predictions of phenotypes of milk yield, fat yield, protein yield, and calving interval were made for Guernsey cows from England and Guernsey Island using genomic EBV, with training sets including 197 de-regressed proofs of genotyped bulls, with cows selected from among 1,440 genotyped cows using different genotyping strategies. Accuracies of predictions were tested using 10-fold cross-validation among the cows. Genomic EBV were predicted using 4 different methods: (1) pedigree BLUP, (2) genomic BLUP using only bulls, (3) univariate genomic BLUP using bulls and cows, and (4) bivariate genomic BLUP. Genotyping cows with phenotypes and using their data for the prediction of single nucleotide polymorphism effects increased the correlation between genomic EBV and phenotypes compared with using only bulls by 0.163 ± 0.022 for milk yield, 0.111 ± 0.021 for fat yield, and 0.113 ± 0.018 for protein yield; a decrease of 0.014 ± 0.010 for calving interval from a low base was the only exception. Genetic correlation between phenotypes from bulls and cows were approximately 0.6 for all yield traits and significantly different from 1. Only a very small change occurred in correlation between genomic EBV and phenotypes when using the bivariate model. It was always better to genotype all the cows, but when only half of the cows were genotyped, a divergent selection strategy was better compared with the random or directional selection approach. Divergent selection of 30% of the cows remained superior for the yield traits in 8 of 10 folds.     

Across-family marker-assisted selection using selective genotyping strategies in dairy cattle breeding schemes

This study investigated the potential loss expected from marker-assisted selection (MAS) when only a proportion of animals are genotyped using several selective genotyping strategies. A population resembling a commercial dairy cattle population over 25 yr was simulated, and the most informative individuals for genotyping were identified among the potential breeding candidates (young bulls and bull-dams). Two strategies were used to identify the most informative animals. The first genotyping strategy was based on selecting individuals for genotyping with predicted total genetic effect [sum of the predicted quantitative trait locus (QTL) and polygenic effects] close to the truncation point for selection. The second strategy used an index that extended the previous strategy to include the variance due to segregation of the QTL in the parents. The 2 strategies for selective genotyping were applied at the 2 different genotyping levels and compared with random selection of candidates for genotyping and complete genotyping of the potential candidates. All selective genotyping strategies at the same proportion of genotyping showed similar cumulative genetic level. The frequency of the favorable QTL allele increased faster with more animals genotyped. Extra response in total genetic effect (polygenic and QTL) was not significantly different between genotyping all candidates (100%), 20%, and 50% genotyping (except for yr 13), but all MAS strategies resulted in significantly higher response than BLUP until yr 18. With 50% (20%) genotyping of candidates for selection within a population, 95% (89%) of maximum cumulative QTL response was achieved in yr 13. All MAS schemes resulted in a 19% decrease in the rate of inbreeding compared with the BLUP scheme. Therefore, it is possible to use selective genotyping in practical dairy cattle breeding and decrease the genotyping costs with a minimal loss of response compared with complete genotyping of the potential candidates.     

Breeding value estimation for fat percentage using dense markers on Bos taurus autosome 14

Prediction of breeding values using whole-genome dense marker maps for genomic selection has become feasible with the advances in DNA chip technology and the discovery of thousands of single nucleotide polymorphisms in genome-sequencing projects. The objective of this study was to compare the accuracy of predicted breeding values from genomic selection (GS), selection without genetic marker information (BLUP), and gene-assisted selection (GEN) on real dairy cattle data for 1 chromosome. Estimated breeding values of 1,300 bulls for fat percentage, based on daughter performance records, were obtained from the national genetic evaluation and used as phenotypic data. All bulls were genotyped for 32 genetic markers on chromosome 14, of which 1 marker was the causative mutation in a gene with a large effect on fat percentage. In GS, the data were analyzed with a multiple quantitative trait loci (QTL) model with haplotype effects for each marker bracket and a polygenic effect. Identical-by-descent probabilities based on linkage and linkage disequilibrium information were used to model the covariances between haplotypes. A Bayesian method using Gibbs sampling was used to predict the presence of a putative QTL and the effects of the haplotypes in each marker bracket. In BLUP, the haplotype effects were removed from the model, whereas in GEN, the haplotype effects were replaced by the effect of the genotype at the known causative mutation. The breeding values from the national genetic evaluation were treated as true breeding values because of their high accuracy and were used to compute the accuracy of prediction for GS, BLUP, and GEN. The allele substitution effect for the causative mutation, obtained from GEN, was 0.35% fat. The accuracy of the predicted breeding values for GS (0.75) was as high as for GEN (0.75) and higher than for BLUP (0.51). When some markers close to the QTL were omitted from the model, the accuracy of prediction was only slightly lower, around 0.72. The removal of all markers within 8 cM from the QTL reduced the accuracy to 0.64, which was still much higher than BLUP. It is concluded that, when applied to 1 chromosome and if genetic markers close to the QTL are available, the presented model for GS is as accurate as GEN.     

Strategies for continual application of marker-assisted selection in an open nucleus population

The objectives of this study were to develop and simulate the implementation of several strategies for repeated application of quantitative trait loci (QTL) detection and marker-assisted selection (MAS) and to compare the short-term and continual genetic responses. A finite locus model was simulated with 20 QTL randomly distributed across 30 chromosome. Three hundred markers were evenly spaced across the genome. Allelic effects were sampled from a double exponential distribution. A daughter design was used every generation to determine the marker alleles favorably associated to QTL alleles. The MAS was applied within family to young bulls, before progeny testing, as part of an open nucleus. Young bulls were selected using strategies based on 1) the single marker with greatest contrast (BEST1), 2) the sum of n greatest contrasts (BESTn), 3) the best n contrasts, limited to one per chromosome (LIMn), 4) the sum of all contrasts exceeding a given threshold n (THRESn), and 5) the sum of contrasts exceeding a threshold, but limited to one per chromosome (LIMT). The maximum progress was achieved by strategies that selected upon several markers flanking multiple QTL in each generation. When THRES was applied, the mean true breeding value (TBV) of selected bulls was increased by 11.98% (over conventional selection) versus 6.73% for BEST1 in the first generation. Applying a full genome scan in each generation allowed selection for different QTL across time. By selecting for multiple QTL over time, MAS maintained superiority over conventional selection for many generations.     

Genomic analysis of claw lesions in Holstein cows: Opportunities for genomic selection,quantitative trait locus detection,and gene identification

Claw lesions are the third most important health issue in dairy cattle, after mastitis and reproductive disorders, and genomic selection is a key component for long-term improvement of claw health. The objectives of this study were to assess the feasibility of a genomic evaluation for claw health in French Holstein cows, explore possibilities to increase evaluation accuracy, and gain a better understanding of the genetic determinism of claw health traits. The data set consisted of 48,685 trimmed Holstein cows, including 9,646 that were genotyped; 478 genotyped sires were also used. Seven claw lesion traits were evaluated using BLUP, genomic BLUP, BayesC, and single-step genomic BLUP, and the accuracies obtained using these approaches were measured through a validation study. The BayesC approach was used to detect quantitative trait locus (QTL) regions associated with the 7 individual traits (digital dermatitis, heel horn erosion, interdigital hyperplasia, sole hemorrhage circumscribed, sole hemorrhage diffused, sole ulcer, and white line fissure) based on their Bayes factor. Annotated genes on these regions were reported. Genomic evaluation approaches generally did not allow for greater accuracies than BLUP, except for single-step genomic BLUP. Accuracies were moderate, but best and worst validation animals were correctly discriminated and showed significant differences in lesion frequencies. A total of 192 QTL regions were identified, including 13 with major evidence or involved for 2 of the traits. A high number of genes were present on these regions, and several had functions associated with the immune system. In particular, the EPYC gene is located close to a major evidence QTL for resistance to digital dermatitis that is also a QTL for interdigital hyperplasia (on chromosome 5, around 20.9 MB) and has been associated with Ehlers-Danlos syndrome in cattle. Genomic selection can be used to improve resistance to individual claw lesions, and several possibilities exist to improve accuracies of genomic evaluations.     

Length of Productive Life of Dairy Cows. 2. Variance Component Estimation and Sire Evaluation

Estimates of the genetic component of the cow's true stayability and of an approximation of her functional stayability were computed from length of productive life records of 83,338 grade Holstein daughters of 2182 sires. True stayability is defined as the aptitude of a cow to delay culling, whereas functional stayability refers to the ability to delay involuntary culling. The probability of a cow being culled is described using two Weibull models with time-dependent covariates. The first one includes fixed herd × year and stage of lactation × lactation number effects and a random sire effect. In the second model, a within herd × level of milk production is included in order to correct the sire effects for the major source of voluntary disposal. Such a model, actually characterizing “milk-corrected” stayability, is a first step toward a functional stayability evaluation.The empirical Bayes estimates of the sire variance component are very similar for both models. Rigorously speaking, phenotypic variance of length of productive life cannot be computed because of the presence of time-dependent covariates in the models. If this problem is ignored, a “pseudoheritability” of 8.5% is obtained for both traits. The correlation between the two sire evaluations is only .74. Milk yield is favorably related to true stayability but slightly opposed to milk-corrected stayability. Functional stayability is presented as an appealing and economically important secondary trait to consider in breeding programs as a complement to the current production traits.     

Alfalfa Silage or Hay Versus Corn Silage as the Sole Forage for Lactating Dairy Cows

In Trial 1, three rations were fed to 21 cows in a 3 × 3 Latin square: 60% alfalfa silage, 60% corn silage, and 79% corn silage (dry matter basis) with the balance from corn and soybean meal. Acid detergent fiber measures indicated alfalfa and corn silage were of excellent quality. Milk production was similar on 60% forage rations but lower on 79% corn silage. Milk fat was reduced on 60% corn silage. In Trial 2, four rations were fed to 16 cows in a 4 × 4 Latin square: 63% alfalfa silage, 60% alfalfa hay, 60% corn silage, and 76% corn silage. Alfalfa forages were higher in acid detergent fiber but corn silage was similar to Trial 1. Dry matter digestibility was highest on 60% corn silage, intermediate on 63% alfalfa silage and 76% corn silage, and lowest on 60% alfalfa hay. Milk production was similar on the diets containing 60 and 63% forage and lower on 76% corn silage. Milk protein concentration was reduced on the alfalfa diets. Highest protein secretion and feed conversion was supported by 60% corn silage. In both trials, potentially digestible neutral detergent fiber from alfalfa was more digestible than that from corn silage, and concentrations of urea in milk and blood were highly correlated. Results indicate high quality alfalfa silage is comparable to corn silage for milk production.     

Pectin quantity,composition and physicochemical behaviour as influenced by the purification process

To examine the purification procedure effect on the pectin amount, purity, macromolecular characteristics, and gelling ability, three pectin isolates, namely, alcohol (APP)-, dialysis (DPP)-, and metal ion (MPP)-purified pectins are obtained from acid extracts of yellow passion fruit rind using alcohol-precipitation, dialysis, or ion-binding precipitation. The results show that the amount of MPP (4.1 g/100 g) is significantly (p < 0.05) lower than the amount of the two other pectins (6.8–7.5 g/100 g). In contrast, MPP has a higher galacturonic acid (78.9 g/100 g) and lower neutral sugar (9.7 g/100 g), ash (0.9 g/100 g), and protein (1.4 g/100 g) contents than the remainder (62.4–70.1, 16.0–17.8, 2.7–5.8, 3.1–3.2 g/100 g, respectively). Molecular-weight distribution patterns suggest that MPP is free of neutral sugar oligosaccharide contaminants unlike the others, especially APP. Therefore, the term of ‘the galacturonic acid yield’ is introduced to complement the conventional term of ‘the pectin yield’. Furthermore, MPP gel preparation is likely to set more rapidly, with the gel formed achieving a much higher strength. It is concluded that the pectin amount, composition, and physicochemical properties can be considerably affected by the purification mode.     

Economic Consequences of Holstein Artificial Insemination Sire Selection by Independent Culling

Objectives were to estimate the profitability changes in AI sire selection by using alternative independent culling restrictions. Net present values of semen were calculated for 439 Holstein sires available for purchase after the 1987 USDA Sire Summary. This measure of profit was used to quantify the net incomes from milk sacrificed by choosing bulls by independent culling restrictions in PD and repeatability for milk, fat test, and type score. Selections were the highest ranking bulls for PD milk or PD dollars that exceeded selection restrictions and cost < $40 per unit of semen.Profits were reduced from $13 (.4 SD) when PD for type score was required to be > 0 to $59 (1.6 SD) when it exceeded 1.5 points. Additional restrictions on repeatability of PD type score and PD of fat percent and milk did not decrease further profit. Culling for fat percent resulted in net income losses ranging from nil when PD > −.05% to $39 (1.0 SD) when PD > .15%. Profits were not reduced by culling for PD milk except when repeatability was required to be greater than 50%. If independent culling is used in sire selection, profit losses would be minimized by selecting on net income rather than gross income or milk.     

Strategies for estimating genetic parameters in marker-assisted best linear unbiased predictor models in dairy cattle

An appropriate strategy to estimate variance components and breeding values in genetic models with quantitative trait loci (QTL) was developed for a dairy cattle breeding scheme by utilizing simulated data. Reliable estimates for variance components in QTL models are a prerequisite in fine-mapping experiments and for marker-assisted genetic evaluations. In cattle populations, only a small fraction of the population is genotyped at genetic markers, and only these animals are included in marker-assisted genetic evaluation models. Phenotypic information in these models are precorrected phenotypes [daughter yield deviations (DYD) for bulls, yield deviations (YD) for cows] estimated by standard animal models from the entire population. Because DYD and YD may represent different amounts of information, the problem of weighting these 2 types of information appropriately arises. To detect the best combination of phenotypes and weighting factors, a stochastic simulation for a trait representing milk yield was used. The results show that DYD models are generally optimal for estimating QTL variance components, but properties of estimates depend strongly on weighting factors. An example for the benefit in selection of using YD is shown for the selection among paternal half-sibs inheriting alternative QTL alleles. Even if QTL effects are small, marker-assisted best unbiased linear prediction can improve the selection among half-sibs, because the Mendelian sampling variance within family can be exploited, especially in DYD-YD models. Marker-assisted genetic evaluation models should also include YD for cows to ensure that marker-assisted selection improves selection even for moderate QTL effects (≥10%). A useful strategy for practical implementation is to estimate variance components in DYD models and breeding values in DYD-YD models.     

Availability of lactoferrin as a natural solubilizer of iron for food products

Lactoferrin (LF) was investigated as a natural solubilizer of iron. Under neutral conditions, 1 mole of LF solubilized 200 moles of iron. In a single dose oral toxicity study, 2.5 g iron solubilized by lactoferrin (FeLF) per kg body weight had no side effects on the rat gastric mucosa, while ferrous sulphate and ferrous citrate (at the same level with respect to iron) caused some serious injuries. FeLF was orally administered to rats for 28 days to investigate its potential toxicity. Haemoglobin content and its regeneration efficiency in rats fed with FeLF were superior to those observed with ferrous sulphate, which is the most frequently used iron supplement. The availability of FeLF in a rat model of iron-deficiency anaemia was demonstrated. LF is useful as a natural iron solubilizer for food products and neutraceuticals from the viewpoints of safety, bioavailability, and productivity.     

Carotenoid composition of two Brazilian genotypes of acerola (Malpighia punicifolia L.) from two harvests

Acerola fruit is native to Central America and adapted very well to Brazil, which in turn became the major worldwide acerola producer, consumer and exporter. Two acerola genotypes were harvested from a Brazilian plantation during the 2003 and 2004 summer harvests. Both genotypes presented β-carotene (265.5–1669.4 μg/100 g), lutein (37.6–100.7 μg/100 g), β-cryptoxanthin (16.3–56.5 μg/100 g) and α-carotene (7.8–59.3 μg/100 g) as the major carotenoids. In both harvests, the β-carotene, β-cryptoxanthin and α-carotene levels were higher in the Olivier genotype, whereas the lutein content was higher in the Waldy Cati 30 genotype. Due to higher sunlight exposure, the fruits harvested in 2004 showed higher total carotenoid contents than those from the 2003 harvest. Acerola, especially the Olivier genotype, with 148–283 RE/100 g, can be considered as a good source of provitamin A.     

Exploring the effects of genotypical and phenotypical variations in bitter taste sensitivity on perception,liking and intake of brassica vegetables in the UK

Brassicaceous vegetables (BV) have chemoprotective effects and yet consumption of BV in the UK is low. Previous studies suggest perception, liking and intake of BV are influenced by bitter taste sensitivity which this study further explores. Phenotypical taste sensitivity of 136 subjects was classified using propythiouracil (PROP) and sodium chloride and fungiform papillae density (FPD) was measured from tongue images. Polymorphisms of TAS2R38 and gustin (CA6) genes were analysed. Liking and bitterness of four raw vegetables (two BV (broccoli and white cabbage) and two non-BV (spinach and courgette)), as well as habitual consumption, were evaluated.There was a significant association between TAS2R38 genotype and PROP taster status (p < 0.0001) and between FPD and PROP taster status (p = 0.029). Individuals with greater sensitivity for PROP predominantly had TAS2R38 PAV/PAV genotype and greater FPD. BV were perceived as more bitter than non-BV (p < 0.0001) with PAV/PAV subjects perceiving significantly stronger bitter intensity. There was a significant difference in liking for the four vegetables (p = 0.002), and between consumers of different TAS2R38 genotype (p = 0.0024). Individuals with TAS2R38 AVI/AVI genotype liked BV more. Regarding intake, both PAV/PAV and AVI/AVI individuals consumed more total vegetables and BV than PAV/AVI. Although PROP nontasters tended to consume more vegetables and BV than the other two phenotype groups, liking and vegetable intake were not significantly affected by taste phenotype. Although there was not a significant effect of CA6 genotype on bitterness ratings, there was a significant interaction between CA6 and TAS2R38, and in addition CA6 genotype was significantly associated with BV intake. However, these effects require validation as the proportions of the population with the CA6 G/G genotype was extremely small (7%).Our results confirmed that bitter taste perception in vegetables was influenced by both genotype and phenotype of bitter taste sensitivity. Moreover, our findings demonstrated that neither genotype nor phenotype of taste sensitivity alone accurately predict vegetable liking and intake as demographic factors were found to have a substantial influence.     

Chewing activities and particle size of rumen digesta and feces of precision-fed dairy heifers fed different forage levels with increasing levels of distillers grains

The objective of this study was to determine the effects of 2 differing forage to concentrate ratios (F:C) and various levels of corn dry distillers grain with solubles (DDGS) replacing canola meal in precision-fed dairy heifer rations on chewing behavior, rumen pH and fill, and particle size of rumen contents and feces. A split plot design with F:C as whole plot and DDGS inclusion level as subplot was administered in a 4-period 4 × 4 Latin square. Eight rumen-cannulated Holstein heifers (12.5 ± 0.5 mo of age and 344 ± 15 kg of body weight, respectively) housed in individual stalls were allocated to F:C 50:50 (low forage) or 75:25 [high forage (HF); dry matter basis] and to a sequence of DDGS level (0, 7, 14, and 21%; dry matter basis). Forage was a mix of 50% corn silage and 50% grass hay (dry matter basis). Diets were fed once daily and formulated to provide equal amounts of nutrients and body weight gain. No differences were found for rumen pH between dietary treatments. Time spent eating tended to be longer for HF and was not affected by DDGS inclusion rate. Ruminating time did not differ by F:C, but linearly increased as DDGS increased (422 to 450 ± 21 min/d). Total chewing time tended to be longer for HF and to increase linearly as DDGS increased (553 to 579 ± 33 min/d). Wet rumen digesta weight and volume were greater for HF. Geometric mean particle length of rumen contents was greater for HF 2 h prefeeding when analyzed with solubles (particles <0.15 mm). Proportion of rumen solubles decreased as DDGS increased 5 h postfeeding. Fecal geometric mean particle length and proportion of particles >1.18 mm increased with increasing levels of DDGS and did not change with F:C. Total chewing time increased by the addition of DDGS and higher F:C. Heifers can compensate for lower physically effective neutral detergent fiber by modifying their chewing behavior. Rumen pH was never at a level that could induce acidosis, and lower eating time at lower F:C was somewhat compensated by time spent ruminating per unit of physically effective neutral detergent fiber intake. Dry distillers grains with solubles, when used in dairy heifer rations as a replacement for canola meal, yielded similar rumen digestion traits.     

In Situ Particle Size Reduction of Alfalfa and Timothy Hay as Influenced by Form and Particle Size

Our objectives were to characterize particle size reduction of alfalfa and timothy hay by in situ incubation. Alfalfa and timothy hay were chopped at .64-cm theoretical length of cut. Subsamples were dry sieved through a 3.2-mm screen, yielding particles > 3.2 and < 3.2 mm, and another subsample ground (5.0 mm). Each forage treatment was placed in polyester bags and ruminally incubated for up to 100 h. Original and residues were wet sieved. Change in geometric mean diameter (initial to lowest) with time for both forages was less as particle size decreased from chopped to ground; however, percent change was similar between chopped and ground and between sieved > 3.2 and < 3.2 mm. Regression of geometric mean diameter on time for alfalfa exhibited quadratic relationships for all particle categories. For timothy, ground and sieved > 3.2 mm exhibited linear, whereas chopped and sieved < 3.2 mm exhibited quadratic relationships with time. Large particle fractions for both forages were reduced proportionately more than particle fractions that were small initially. Particle size distribution showed particles > 1.18 mm accounted for 63.0 to 98.4% of the initial DM, which was reduced to 14.9 to 24.1% by 100 h. Microbial fermentation and detrition associated with ruminal movements appear to contribute substantially to particle breakdown.     

Optimal Genetic Improvement for the High Producing Cow

Many 40,000-lb (18,144 kg) cows exist in today's dairy population, but herds capable of 40,000-lb averages for all cows remain to be developed. Traditional genetic improvement practices, based on consistent use of current high-ranking AI bulls selected to improve economically important traits, will remain important tools to develop high producing cows. Future breeding strategies will likely include attention to how high production is achieved and may include direct selection for increased appetite or some measure of energy balance to support high production, reproduction, and immune function. Direct selection for improved fertility, perhaps involving traits not presently used in herd management, may prove to be necessary as yields increase. Roles for evolving technologies such as marker-assisted selection, manipulation of the bovine genome, and cloning remain unclear, but will likely be incorporated into traditional progeny testing schemes. Equipment to routinely monitor physiological functions may encourage the establishment of large progeny test herds with expanded data recording capability. The expense could lead to proprietary genetic lines and private genetic evaluation systems such as exist in poultry and swine. Pedigree information will become more important in commercial herds to manage inbreeding. The dairy industry can expect to benefit from current research efforts in human genetics. However, current funding of dairy breeding research in the United States will limit the number of individuals trained in methods to implement those results in dairy cattle.