Predictive ability of genomic selection models in a multi-population perennial ryegrass training set using genotyping-by-sequencing

Key message

Genomic prediction models for multi-year dry matter yield, via genotyping-by-sequencing in a composite training set, demonstrate potential for genetic gain improvement through within-half sibling family selection.

Abstract

Perennial ryegrass (Lolium perenne L.) is a key source of nutrition for ruminant livestock in temperate environments worldwide. Higher seasonal and annual yield of herbage dry matter (DMY) is a principal breeding objective but the historical realised rate of genetic gain for DMY is modest. Genomic selection was investigated as a tool to enhance the rate of genetic gain. Genotyping-by-sequencing (GBS) was undertaken in a multi-population (MP) training set of five populations, phenotyped as half-sibling (HS) families in five environments over 2 years for mean herbage accumulation (HA), a measure of DMY potential. GBS using the ApeKI enzyme yielded 1.02 million single-nucleotide polymorphism (SNP) markers from a training set of n = 517. MP-based genomic prediction models for HA were effective in all five populations, cross-validation-predictive ability (PA) ranging from 0.07 to 0.43, by trait and target population, and 0.40–0.52 for days-to-heading. Best linear unbiased predictor (BLUP)-based prediction methods, including GBLUP with either a standard or a recently developed (KGD) relatedness estimation, were marginally superior or equal to ridge regression and random forest computational approaches. PA was principally an outcome of SNP modelling genetic relationships between training and validation sets, which may limit application for long-term genomic selection, due to PA decay. However, simulation using data from the training experiment indicated a twofold increase in genetic gain for HA, when applying a prediction model with moderate PA in a single selection cycle, by combining among-HS family selection, based on phenotype, with within-HS family selection using genomic prediction.

     

Effect of ryegrass mosaic virus on the quality of perennial ryegrass

In 1976 fields of perennial ryegrass located on two farms in the west of Scotland were found to be far more severely and extensively infected by ryegrass mosaic virus (RMV) than had been seen in earlier surveys of ryegrass fields in 1972 and 1973. The chemical analysis of samples of ryegrass from 10 of the fields examined in 1976 showed that RMV significantly reduced the organic matter content, organic matter digestibility, water soluble carbohydrate content and D value of the grass. Furthermore, the losses suffered by grass with the severe necrotic symptom of RMV were significantly greater than those incurred by the grass with the mild mottle only. A correlation analysis between water soluble carbohydrate level and D value showed that whilst the loss in water soluble carbohydrates in mottled grass accounted for much of the loss in digestibility this was not the case in the necrotic grass where loss of green tissue appeared to be a more important contributory factor.     

Potential of genotyping-by-sequencing for genomic selection in livestock populations

Background

Next-generation sequencing techniques, such as genotyping-by-sequencing (GBS), provide alternatives to single nucleotide polymorphism (SNP) arrays. The aim of this work was to evaluate the potential of GBS compared to SNP array genotyping for genomic selection in livestock populations.

Methods

The value of GBS was quantified by simulation analyses in which three parameters were varied: (i) genome-wide sequence read depth (x) per individual from 0.01x to 20x or using SNP array genotyping; (ii) number of genotyped markers from 3000 to 300 000; and (iii) size of training and prediction sets from 500 to 50 000 individuals. The latter was achieved by distributing the total available x of 1000x, 5000x, or 10 000x per genotyped locus among the varying number of individuals. With SNP arrays, genotypes were called from sequence data directly. With GBS, genotypes were called from sequence reads that varied between loci and individuals according to a Poisson distribution with mean equal to x. Simulated data were analyzed with ridge regression and the accuracy and bias of genomic predictions and response to selection were quantified under the different scenarios.

Results

Accuracies of genomic predictions using GBS data or SNP array data were comparable when large numbers of markers were used and x per individual was ~1x or higher. The bias of genomic predictions was very high at a very low x. When the total available x was distributed among the training individuals, the accuracy of prediction was maximized when a large number of individuals was used that had GBS data with low x for a large number of markers. Similarly, response to selection was maximized under the same conditions due to increasing both accuracy and selection intensity.

Conclusions

GBS offers great potential for developing genomic selection in livestock populations because it makes it possible to cover large fractions of the genome and to vary the sequence read depth per individual. Thus, the accuracy of predictions is improved by increasing the size of training populations and the intensity of selection is increased by genotyping a larger number of selection candidates.

Electronic supplementary material

The online version of this article (doi:10.1186/s12711-015-0102-z) contains supplementary material, which is available to authorized users.     

The resistance of the perennial ryegrass cultivar Endura to ryegrass mosaic virus

The perennial ryegrass cv. Endura is particularly resistant to infection with two ryegrass mosaic virus isolates, RMV-Roth and RMV-Sax. A few plants appeared immune to RMV-Roth but were infected by RMV-Sax. Two plants that developed only mild symptoms after infection with RMV-Roth contained few virus particles, but passage through these resistant plants caused no detectable change in the virulence of RMV-Roth towards S 22 Italian ryegrass. In offsets derived from one of these resistant plants, RMV-Sax caused severe symptoms and attained a high virus concentration but it was unable to infect if the plant was already infected with RMV-Roth.     

The effect of ryegrass mosaic virus on the yield of perennial ryegrass swards

The influence of ryegrass mosaic virus (RMV) on yield of perennial ryegrass swards was examined under two harvesting frequencies and at two levels of nitrogen fertilizer, in a field trial planted with a clone of ryegrass, cv. S 321, using small plots established with proportions of healthy and infected plants. RMV reduced the height and the number of tillers of plants grown in pots and reduced the height also of plants in the field. The swards were planted in August and at a harvest 8 wk later RMV reduced the yield from 2–12 to 1–52 t dry matter (d.m.)/ha. In the first full harvest year, RMV caused only a small reduction in yield where no N fertilizer was applied and the application of a very small amount of fertilizer would have recouped the loss. On the other hand RMV severely restricted the increase in yield resulting from heavy dressings of the fertilizer. Thus where 400 kg N/ha was applied, RMV reduced the total annual yield from 18-6 to 13-8 t d.m./ha. The effect of the virus was especially great both in the Spring, the period of maximum herbage production, and when the level of fertilizer was high.     

Mechanism of paraquat tolerance in perennial ryegrass

Abstract The mechanism of paraquat tolerance was investigated in lines of perennial ryegrass (Lolium perenne L.) which had been selected for resistance to the herbicide. Uptake, metabolism and translocation of paraquat were studied. Susceptible cultivars and a tolerant line were not found to differ in uptake of radioactive paraquat applied to the leaf surface or supplied to the cut ends of excised leaves. Distribution of herbicide within leaf tissue was similar in tolerant and susceptible plants and no metabolites of ¹⁴C-paraquat were detected in tolerant or susceptible plants treated with sub-lethal concentrations of the herbicide. Autoradiography and quantitative determinations showed much variation in translocation of ¹⁴C-paraquat out of treated leaves of intact plants, but the variation was not related to the degree of susceptibility to the herbicide. It is concluded that paraquat tolerance in perennial ryegrass is unlikely to depend upon reduced uptake, enhanced metabolism or altered translocation of the herbicide.     

Selection of perennial and Italian ryegrass plants resistant to ryegrass mosaic virus

Perennial ryegrass plants collected from fields and Italian ryegrass plants grown from seed were selected for resistance to infection by ryegrass mosaic virus (RMV) by repeated manual inoculation. Two of 108 perennial ryegrass plants and one of 150 Italian ryegrass plants were symptomless after seven and nine inoculations respectively. These three plants were propagated vegetatively. Plants of the two perennial ryegrass clones showed no symptoms after further manual inoculations with the initial isolate of RMV, or with an inoculum from infected plants collected from several fields, or after inoculation by viruliferous mites. Electron microscopy and back tests indicated that the plants were virus free. Some plants of the selected Italian ryegrass clone became infected after a further inoculation with mites or sap, but fewer than similarly inoculated unselected plants.     

Decreased severity of ryegrass mosaic virus after passage through resistant perennial ryegrass

After exposure to infection in the field, the proportion of plants showing distinct symptoms of ryegrass mosaic virus (RMV) was less in perennial than in Italian ryegrass. The perennial ryegrass cv. Mascot had a smaller proportion of plants with symptoms than cv. S.23. Far milder symptoms were induced in test plants by RMV from naturally infected perennial ryegrass plots than from Italian ryegrass plots. Within perennial ryegrass, RMV from cv. Mascot caused milder symptoms than that from cv. S.23. Severe RMV isolated from Italian ryegrass cv. Trident (RMVT) became milder after one passage through cv. Mascot, although not as mild as RMV obtained from field plots of cv. Mascot (RMVM). Families from two highly resistant perennial ryegrass clones and two randomly selected clones of cv. S.23 crossed in all possible combinations varied in symptom severity when inoculated with RMVT but not when inoculated with RMVM. Families inoculated with RMVT also yielded virus which varied in the severity of symptoms induced in test plants, families with severe symptoms yielding severer virus. Thus, much of the variation in the resistance of these clones could be attributed to infection with RMV of differing severity. Resistance was controlled by several genes which were mainly additive in their effect.     

Improved dairy cattle mating plans at herd level using genomic information

From 2012 to 2018, 223 180 Montbéliarde females were genotyped in France and the number of newly genotyped females increased at a rate of about 33% each year. With female genotyping information, farmers have access to the genomic estimated breeding values of the females in their herd and to their carrier status for genetic defects or major genes segregating in the breed. This information, combined with genomic coancestry, can be used when planning matings in order to maximize the expected on-farm profit of future female offspring. We compared different mating allocation approaches for their capacity to maximize the expected genetic gain while limiting expected progeny inbreeding and the probability to conceive an offspring homozygous for a lethal recessive allele. Three mate allocation strategies (random mating (RAND), sequential mating (gSEQ€) and linear programing mating (gLP€)) were compared on 160 actual Montbéliarde herds using male and female genomic information. Then, we assessed the benefit of using female genomic information by comparing matings planned using only female pedigree information with the equivalent strategy using genomic information. We measured the benefit of adding genomic expected inbreeding and risk of conception of an offspring homozygous for a lethal recessive allele to Net merit in mating plans. The influence of three constraints was tested: by relaxing the constraint on availability of a particular semen type (sexed or conventional) for bulls, by adding an upper limit of 8.5% coancestry between mate pairs or by using a more stringent maximum use of a bull in a herd (5% vs 10%). The use of genomic information instead of pedigree information improved the mate allocation method in terms of progeny expected genetic merit, genetic diversity and risk to conceive an offspring homozygous for a lethal recessive allele. Optimizing mate allocation using linear programming and constraining coancestry to a maximum of 8.5% per mate pair reduced the average coancestry with a small impact on expected Net Merit. In summary, for male and female selection pathways, using genomic information is more efficient than using pedigree information to maximize genetic gain while constraining the expected inbreeding of the progeny and the risk to conceive an offspring homozygous for a lethal recessive allele. This study also underlines the key role of semen type (sexed vs conventional) and the associated constraints on the mate allocation algorithm to maximize genetic gain while maintaining genetic diversity and limiting the risk to conceive an offspring homozygous for a lethal recessive allele.     

Linkage disequilibrium in synthetic varieties of perennial ryegrass

Synthetic varieties obtained after three to four panmictic generations are variable, not structured and so can be used for association studies. The pattern of linkage disequilibrium (LD) decay determines whether a genome scan or a candidate gene approach can be used for an association study between genotype and phenotype. Our goal was to evaluate the effect of the number of parents used to build the synthetic varieties on the pattern of LD decay. LD was investigated in the gibberelic acid insensitive gene (GAI) region in three synthetic varieties of perennial ryegrass (Lolium perenne L.) chosen for their contrasted number of parents in the initial polycrosses. Results were compared with those obtained from a core collection. STS and SSR markers were used to evaluate variation, structuration and LD in each variety. As expected, the varieties variability increased with the number of parents almost up to the core collection variability. No structuration was observed in the varieties. Significant LDs were observed up to 1.6 Mb in a variety originated from six related parents and not above 174 kb in a variety originated from 336 parents. These results suggest that a candidate gene approach can be used when varieties have a large number of parents and a genome scan approach can be envisaged in specific regions when varieties have a low number of parents. Nevertheless, we strongly recommend to estimate the pattern of LD decay in the population and in the genomic region studied before performing an association study.     

Fungicide seed treatments increase growth of perennial ryegrass

Field, laboratory and glasshouse experiments were carried out to measure effects of seed treatments with captan or thiram on growth of perennial ryegrass (Lolium perenne L.). Field-sown captan- or thiram-treated seed gave twice as many seedlings as untreated seed. Spaced plants growing from fungicide-treated seed produced almost 6 times more dry matter 16 weeks after sowing than those from untreated seed. This effect, though diminishing with time, was still apparent more than a year after sowing. Fungicides in sterile agar growth medium were phytotoxic to seedlings at concentrations of 10μg/ml and greater. Seedlings grown from treated seeds sown from 5 to 15 mm away from developing colonies of the virulent seedling pathogenFusarium oxysporum Schlecht. were more than 4 times larger than those grown from untreated seeds. Captan-treated seed sown into pots containing field soil produced more and larger seedlings than untreated seed. Methyl bromide fumigation of the same soil also increased both number and size of seedlings. Fungicidal, rather than direct chemical effects, at early stages of seedling growth, account for increased growth of plants from fungicide-treated seed.     

Estimating genomic coexpression networks using first-order conditional independence

We describe a computationally efficient statistical framework for estimating networks of coexpressed genes. This framework exploits first-order conditional independence relationships among gene-expression measurements to estimate patterns of association. We use this approach to estimate a coexpression network from microarray gene-expression measurements from Saccharomyces cerevisiae. We demonstrate the biological utility of this approach by showing that a large number of metabolic pathways are coherently represented in the estimated network. We describe a complementary unsupervised graph search algorithm for discovering locally distinct subgraphs of a large weighted graph. We apply this algorithm to our coexpression network model and show that subgraphs found using this approach correspond to particular biological processes or contain representatives of distinct gene families.     

A re-examination of phosphorus-lime interactions in perennial ryegrass

The results of a previous study had suggested that under conditions of limited P availability, Ca may be able to compensate for P in the shoot tissue of perennial ryegrass. To verify this preliminary finding, a factorial experiment was set up which simultaneously tested the effects of Ca and P fertilization on the yield and chemical composition of perennial ryegrass. Calcium was supplied as either lime or gypsum in order to differentiate between the effects of Ca and pH on the response of perennial ryegrass to P fertilization. In the final stage of the experiment a Zn treatment was included, to see whether altering the P/Zn ratios of plant shoots had any influence on the purported interaction between Ca and P. The results demonstrated that the P-sparing effect of lime occurs, at least partly, because Ca application improves the efficiency of absorbed P for DM production. However, it was reasonably clear that the site of the interaction between Ca and P was the soil-root interface, and not shoot tissue. It was suggested that under conditions of limited P supply, Ca stablizes root membranes and thereby minimizes both the efflux losses of nutrients from root tissue, and the compensatory flow of photosynthates from shoots to roots. No interaction was observed between P and Zn treatments in this study. Instead, a positive interaction was found between lime and Zn treatments, which suggests that the stabilizing action of Ca on root membranes requires Zn as a co-stabilizing factor. It is proposed that chemical analysis of shoot tissue alone may not be sufficient to accurately diagnose the P, Ca or Zn status of whole plants, since the critical levels of these elements in shoots appear to bear little relation to their requirements in the rhizosphere.     

Molecular genetics of fructan metabolism in perennial ryegrass

Fructans are the main storage carbohydrates of temperate grasses, sustaining regrowth immediately after defoliation, as well as contributing to the nutritive value of feed. Fructan metabolism is based on the substrate sucrose and involves fructosyltransferases (FTs) for biosynthesis and fructan exohydrolases (FEHs) for degradation. Sucrose is also utilized by invertases (INVs), which hydrolyse it into its constituent monosaccharides for use in metabolism. The isolation, molecular characterization, functional analysis, and phylogenetic relationships of genes encoding FTs, FEHs, and INVs from temperate grasses are reviewed, with an emphasis on perennial ryegrass (Lolium perenne L.). The roles these enzymes play in fructan accumulation and remobilization, and future biotechnological applications in molecular plant breeding are discussed.     

A rapid method by flow cytometry for estimating persistence of tetraploid perennial ryegrass in pasture mixtures with diploid perennial ryegrass

A simple and rapid method for the estimation of the proportion of diploid and tetraploid perennial ryegrass in a mixed sward is described, based on the measurement of the nuclear DNA content of the leaf tissue cells by flow cytometry. The error of estimation was less than 5 %.     

Stem-boring Diptera in perennial ryegrass in relation to fertiliser. I. Nitrogen level and form

The effects of applying nitrogen to a perennial ryegrass sward were studied in relation to the size and nature of populations of dipterous stem-bores. Two levels of both ammonium sulphate and calcium nitrate were applied to the sward. Larvae were more numerous in the plots treated with high levels of nitrogen regardless of the form applied. Also, there were more larvae in the plots treated with ammonium sulphate than in those treated with calcium nitrate. A limited range of stem-borer species predominated. These were Oscinella vastator, two forms of O. frit and Geomyza tripunctata. Stem-borer species were distributed differently between plots receiving high and low nitrogen levels; G. tripunctata occurred almost exclusively on plots receiving high nitrogen applications while O. vastator was more frequent in those receiving low levels of nitrogen.     

Development of a genomic microsatellite library in perennial ryegrass (Lolium perenne) and its use in trait mapping

Background and Aims: Perennial ryegrass (Lolium perenne) is one of the key forageand amenity grasses throughout the world. In the UK it accountsfor 70 % of all agricultural land use with an estimated farmgate value of £6 billion per annum. However, in termsof the genetic resources available, L. perenne has lagged behindother major crops in Poaceae. The aim of this project was thereforethe construction of a microsatellite-enriched genomic libraryfor L. perenne to increase the number of genetic markers availablefor both marker-assisted selection in breeding programmes andgene isolation. Methods: Primers for 229 non-redundant microsatellite markers were designedand used to screen two L. perenne genotypes, one amenity andone forage. Of the 229 microsatellites, 95 were found to showpolymorphism between amenity and forage genotypes. A selectionof microsatellite primers was selected from these 95 and usedto screen two mapping populations derived from intercrossingand backcrossing the two forage and amenity grass genotypes. Key Results and Conclusions: The utility of the resulting genetic maps for analysis of thegenetic control of target traits was demonstrated by the mappingof genes associated with heading date to linkage groups 4 and7.