But, the expansion of NGS technologies and genotyping platforms widen the marker applications for crop improvement and were the basis for the success of GS, which has almost shifted the complete reliance on phenotyping to an increased reliance on genotyping-based selection. Among the number of factors that ascertain the efficiency and accuracy with which the superior lines can be predicted through GS, the type and density of marker used as well as size of reference population limited by high cost genotyping are the most critical factors Jannink et al.
In addition, the population structure i. Population structure produces spurious marker-trait associations in genome-wide association studies due to different allele frequencies among subpopulations, which may inflate estimate of genomic heritability and bias accuracies of genomic predictions Price et al. When population structure exists in both training and validation sets, correcting for population structure led to a significant decrease in accuracy with genomic prediction. Thus, the rapid advances in sequencing technology have led to higher throughput and low cost per sample, and positioning NGS-based genotyping as a cost-effective and efficient agrigenomics tool for performing GS in both model and non-model crop species as well as for crops with large and complex genomes Metzker, ; Kirst et al.
GS have been attempted in Miscanthus sinensis for 17 traits related to phenology, biomass, and cell wall composition using RADSeq, and genome-wide prediction accuracies were investigated to be moderate to high average of 0.
Spindel et al. These entire make the sequence based genotyping an ideal approach for GS and its successful application in crop breeding Figure 3. Genotype-by-sequencing follows a modified RAD-seq based library preparation protocol for NGS and is a simple and highly multiplexed system. The important feature of this system include reduced sample handling, fewer PCR and purification steps, low cost, no reference sequence limits, no size fractionation and efficient barcoding technique Davey et al.
It enables the detection of thousands of millions of SNPs in the large collections of lines that can be used for genetic diversity analysis, linkage mapping, GWAS, GS, and evolutionary studies. Beissinger et al. GBS is becoming increasingly important as a cost-effective and unique tool for genomics-assisted breeding in a range of plant species. Genotyping-by sequencing combines marker discovery and genotyping of large populations, making it an excellent marker platform for breeding applications even in the absence of a reference genome sequence or previous polymorphism discovery.
In crop species with large and complex genomes as well as lack of reference sequence the marker technologies lagged behind, which is an important factor to consider for large scale application of GS in crop plants. The high polyploidy level, large genome size and lack of reference genome wheat were the major hindrance of molecular marker development in the crop species. Genotyping-by-sequencing has recently been applied to large complex genomes of barley Hordeum vulgare L.
The GBS have also been used for de novo genotyping of breeding panels and to develop accurate GS models, for the large, complex, and polyploid wheat genome.
GAB value prediction accuracies were 0. The first evidence of the prediction accuracy of GBS in plants came from Poland et al. Since then GS involving GBS have been reported in multiples of crop species including both model and non-model Table 1. In soybean, prediction accuracy for grain yield, assessed using cross validation, was estimated to be 0. Genotyping cost of GBS per individual is lowest in comparison to array-based and other NGS-based markers in wheat and other non-model crop species Bassi et al.
The fraction of the genome covered by GBS can potentially be much greater than the fraction captured by even the densest SNP arrays currently available in crop plants Gorjanc et al. Furthermore, unlike SNP arrays that are typically developed from a limited sample of individuals, GBS can capture genetic variation that is specific to a population or family of interest.
GBS has the advantage that markers are discovered using the population to be genotyped, thus minimizing ascertainment bias. The applied plant breeding is the ultimate source of improved crop varieties, and has led to green revolution in s. At every time this field was supported and facilitated by the new technologies and approaches. The impact of climate change on crop production and global food security is being discussed currently throughout the world Reynolds, Therefore, to fight against these challenges and maintaining sustainable agriculture, new crop varieties are required at an accelerated rate to increase production as well as withstand better biotic and abiotic stresses.
As discussed that most of the agriculturally important traits are governed by minor effect genes, and with a high occurrence of epistatic interactions such as grain yield, plant growth and stress adaptation etc Mackay, Improvement of these traits through conventional breeding and MAS do not met the expected results to pace with growing human population.
In this regard, GS provides new opportunities for increasing the efficiency of plant breeding programs Bernardo and Yu, ; Heffner et al. The GS has the potential to fix all the genetic variation and has ability to accurately select individuals of higher breeding value without the requirement of collecting phenotypes pertaining to these individuals.
This has facilitated a shortening of the breeding cycle and enable rapid selection and intercrossing of early generation breeding material Figure 2. Recent research has shown that GS has the potential to reshape crop breeding, and many authors have concluded that the estimated genetic gain per year applying GS is several times that of conventional breeding Bassi et al. The cost of genotyping has declined dramatically in the era of NGS Davey et al.
This will expand the genetic evaluation of germplasm in crop improvement programs and accelerate the delivery of crop varieties with improved yield, quality, biotic and abiotic stress tolerance, and thus directly benefit attempts to address the challenge of increasing global hunger.
Thus, GS will be the cornerstone for the release of global hunger, and has tremendous impact on crop breeding and variety development Figure 2. It is clear from the above discussion that genotyping no more limit the prediction accuracy of GS. But the technical challenge in implementing the GS in crop plants is the reliability of phenotypic data that creates genotype-phenotype gap GP gap.
The GS predication model used to derive GEBV for all genotyped individuals of the reference set depends upon the precision and accuracy the phenotypic data is taken on TP, and thereby the genetic gain achieved after every generation of selection Meuwissen et al. In this regard, several phenotyping facilities have been developed around the world that can scan and record precise and accurate data for thousands of plants quickly by making use of non-invasive imaging, spectroscopy, image analysis, robotics and high-performance computing facilities Cobb et al.
The HTP helps us to collect high quality accurate phenotyping data. The manual, invasive and destructive methods of plant phenotyping were laborious, costly and less precise, often led inaccuracy in GAB as well as limit the population size.
This importance can be realized by the fact that an International Plant Phenomics Initiative was launched recently to address crop productivity 1. The earlier manual methods of plant phenotyping are now giving way to high-throughput precise non-destructive imaging techniques.
These phenomics facilities make sure to scan thousands of plants in a day so that this phenotyping technology will become similar to high-throughput DNA sequencing in the field of genomics Finkel, Hence, to achieve fruitful results from GS and GAB much more efforts and funds are required to be allocated in this field. In India well established phenomic facility has been not created yet, therefore efforts are required to create such facility in the country to boost agriculture production.
Hence, HTP will change the paradigm of GS and led its effective application in crop plants as well as harness its true benefits for crop improvement Figure 3. The classical breeding had made a significant contribution to crop improvement but was slow in targeting the complex and low heritable quantitative traits.
In this regard, GS has been suggested to have a potential to fix all the genetic variation of complex traits. Many studies have shown tremendous opportunities of GS to increase genetic gain in plant breeding. The important consideration for GS to work in crop plant is the availability of low cost, flexible and high density marker system. Revolution of inexpensive NGS technologies has resulted in increasing number of crop genomes as well as provides the low cost and high density SNP genotyping.
These marker technologies have deeply estimated the population structure of both training and validation set, and have increased the selection accuracy of GS. The NGS markers, as well as methodological refinements such as the implementation of genotype-by-environment interaction in prediction models , are notably contributing to paving the way for a successful implementation of GS in plant breeding.
Furthermore, the GS and HTP together will change the entire paradigm of plant breeding as well as led to the effective increase in genetic gain for complex traits. In the future when the genomic sequencing cost further decreases and WGS become feasible and cost effective for GS, there will be further increase in the prediction accuracy of GS. All authors viz. The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Annicchiarico, P. Accuracy of genomic selection for alfalfa biomass yield in different reference populations. BMC Genomics Arruda, M. Comparing genomic selection and marker-assisted selection for Fusarium head blight resistance in wheat Triticum aestivum. Bassi, F. Breeding schemes for the implementation of genomic selection in wheat Triticum spp.
Plant Sci. Beissinger, T. Marker density and read depth for genotyping populations using genotyping-by-sequencing. Genetics , — Bernardo, R. Prospects for genomewide selection for quantitative traits in maize. Crop Sci. Bhat, J. Phenomics: a challenge for crop improvement in genomic era. Plant Breed. Breseghello, F. Traditional and modern plant breeding methods with examples in rice Oryza sativa L.
Food Chem. Cabrera-Bosquet, L. High-throughput phenotyping and genomic selection: the frontiers of crop breeding converge. Plant Biol. Castro, A. Mapping and pyramiding of qualitative and quantitative resistance to stripe rust in barley.
Cobb, J. Next-generation phenotyping: requirements and strategies for enhancing our understanding of genotype—phenotype relationships and its relevance to crop improvement. Collard, B. Marker-assisted selection: an approach for precision plant breeding in the twenty-first century. B Biol. Collins, F. A vision for the future of genomics research.
Nature , — Crossa, J. Genomic prediction in maize breeding populations with genotyping-by-sequencing. G3 Bethesda 3, — Prediction of genetic values of quantitative traits in plant breeding using pedigree and molecular markers. Genomic prediction of gene bank wheat landraces. G3 Bethesda 6, — Genomic selection has been well established in the field of animal breeding, but is in its beginning in crops plants and forest tree breeding. Genome-wide selection or genomic selection estimates marker effects across the full ordering of the breeding population BP supported the prediction model developed within the training population TP.
Training population could be a group of related individuals such as half-sibs or lines that are each phenotypes and genotypes. Breeding population typically is just genotyped not phenotypes. Hence, Genomic selection depends on the degree of genetic similarity between training population and breeding population within the Linkage disequilibrium between marker and trait loci. Breeding values have not been a preferred index in plant breeding, however it is in animal breeding.
Once plan of genomic estimated breeding value GEBV was planned, it had been considered an unrealistic approach due to lack of enormous scale genotyping technologies. However, currently, it has been a possible approach with recent advances in high throughput genotyping platforms 3rd generation platforms. Generally processes of genomic selection and marker assisted selection used for Quantitative Traits are shown in Figure 1.
Self-pollinated crop genomic selection vs. The main schemes of the two approaches are similar, wherever each marker assisted selection and genomic selection consist of breeding and training phases. In the training phase, phenotypes and genome-wide GW genotypes are investigated in an exceedingly set of a population, i. Among populations, important relationships between phenotypes and genotypes are expected utilizing statistical models.
Within the breeding phase, genotype data are obtained in an exceedingly breeding population, on the basis of genotypic information favorable individuals are selected. There are three prominent variations between the two approaches: 1 within the training section, quantitative trait loci QTLs are known in marker assisted selection whereas formulae for genetic estimation of breeding value prediction are generated in genomic selection, called genomic selection models; 2 within the breeding section, genotype data are solely needed for targeted regions in marker assisted selection, whereas genomic selection genotype data are considered to be mandatory in genomic selection 3 within the breeding phase, favorable individuals are selected on the bases of the linked markers in marker assisted selection, whereas GEBVs are used for selection in GS.
Thus, GS collectively analyses all the genetic variance of every individual by summing the marker impacts of GEBV and it is expected to deal with little effect genes that cannot be captured by traditional MAS.
The statistical ways employed by GS are comparatively new the plant-breeding community. The ways of marker-assisted selection MAS or marker-assisted recurrent selection MARS assume that the user is aware of that alleles are favorable, and what their average effects on the phenotype are.
This assumption is viable for major-gene traits however not for quantitative traits that are influenced by several loci of little impact and the environment. To deal with quantitative traits, new statistical approaches that might account for this uncertainty were required to get the most effective predictions potential.
Finding problem with locus identification, entailed that the consequences for all marker loci be at the same time estimated. Once a prediction based on allele effects, the allele becomes the unit of analysis.
Alleles are so the units that need to be replicated inside and across environments. However that replication will occur in spite of the particular lines carrying the alleles such lines themselves no longer need to be replicated. Within the breeding context, removing the requirement for line replication opens the likelihood of dramatically increasing the amount of lines pushed through the pipeline of a breeding program, and successively of accelerating selection intensity.
Genomic selection is to assemble a training population for individuals for which both genotypes and phenotypes are available and use those data to create a statistical model that relates variation in observed genotypes marker loci to variation in the observed phenotypes of the individuals.
The statistical model obtained from genotype and phenotype is then applied to a prediction population comprised of individuals for which genotypes are available, but phenotypes are not. This similarity may exist because breeding population is selected from training population or descended from training population or because density of markers is so high that every trait locus is in disequilibrium with at least one marker locus across the entire population of the target species.
The training population is genotyped and phenotyped to train the genomic selection GS prediction model. Genotypic information from the breeding material is then fed into the model to calculate genome estimated breeding values GEBV for these lines Figure 2.
Genomic selection scheme. Information on phenotype and genotype for a training population allows estimating parameters for the model. Modified: Castro et al. Traditional marker assisted selection, whereas helpful for merely transmitted traits controlled by few loci, loses effectiveness because the number of loci will increase.
This is often true for individual quantitative traits or once multiple traits are below selection. Quantitative traits like grain yield, abiotic stress have verified hard to enhance with marker-assisted selection. The main limitations are i tiny population sizes and traditional statistical strategies that have inadequate power to find and accurately estimate effects of small-effect quantitative trait loci QTL and ii gene x gene interactions epistasis and iii genotype x environment interactions G.
E that have restricted the exchangeability of quantitative trait loci result estimates across populations and environments. The Beavis effect is a statistical phenomenon in biology that refers to the overestimation of the effect size of quantitative trait loci QTL as a result of small sample sizes in QTL studies.
The availability of low cost and extensive molecular markers in plants has allowed breeders to raise however molecular markers might best be used to win breeding progress. Additionally advances in high-throughput genotyping have markedly reduced the value per data point of molecular markers and increasing genome coverage.
This reduction was in the main the results of three parallel developments [ 2 ] i the invention of huge numbers of single nucleotide polymorphism SNP markers in several species; ii development of high-throughput technologies, like multiplexing and gel-free deoxyribonucleic acid arrays, for screening SNP polymorphisms; and iii automation of the marker-genotyping method, together with efficient procedures for deoxyribonucleic acid extraction [ 2 ].
Phenotyping prices are increased Genotyping prices are being reduced and marker densities are being increased speedily. Statistical strategies are inadequate for improving polygenic traits controlled by several loci of small impact. There will be more markers explanatory variables than lines observations that introduce statistical issues. Drawback of small p number of traits and enormous m number of markers ends up in a lack of degrees of freedom. The foremost acceptable statistical model is required to at the same time estimate several marker effects from a limited range of phenotypes.
To overcome these issues, a range of ways, e. The most economical use of GS is to exchange expensive and long phenotyping by a prediction of the genetic worth of the character below selection or any multi trait index.
Thus, the foremost expected advantage is to shorten selection cycles. Progeny testing schemes have a high accuracy of selection, however the time interval is also additional, takes long term to perform a cycle of selection that decreases the genetic gain.
Selection accuracy is adequate to the correlation between selection criteria and breeding value i. The power to calculate extremely correct GEBVs and also the potential to drastically cut back makeup analysis frequency and selection cycle time expedited a speedy adoption of genomic selection and is revolutionizing the oxen breeding trade Figure 3.
The GEBV is generally equal to g x i. Further similarities among GS models can be seen by recognizing that they all seek to minimize a certain cost function. In least squares analysis, the well-known cost function is simply the sum of squared residuals. CV entails splitting the data into training and validation set.
The ratio of observations in each set varies, but often a fivefold CV is used, that is, the data set is randomly divided into five sets, with four sets being combined to form the training set and the remaining set designated as the validation set. Each subset of the data is used as the validation set once, before applying of the prediction model to the breeding population, the accuracy of the model should be tested.
For this, most of the training population is used to create a prediction model, which is then used to estimate the genomic estimation breeding values of the remaining individuals in the training population, using genotypic data only.
Once valid, the model is often applied to a breeding population to calculate GEBVs of lines that genotypical, however no phenotypical, information is available. This correlation provides an estimate of selection accuracy and thus directly relates GEBV prediction accuracy to selection response [ 2 ]. Other statistics such as mean-square error MSE are used occasionally [ 3 ]. The assumption could be violated if the training and validation data were collected in the same environment.
In that case, genotype by environment G. Thus, training and validation data should be collected in different environments to ensure sound estimates of GEBV prediction accuracy. Accurate GEBV predictions offer the possibility that future elite and parental lines will be selected on GEBV rather than phenoypic data from extensive field testing. Immediate impact would be a great increase in speed of breeding cycle increasing selection gains per unit time.
Thus, GS could radically change the practice of field evaluation for breeders. Of course, regardless of the breeding method used, final field evaluations of varieties across the target environments will be needed before they are distributed to farmers. Breeding cycle time is shortened by removing phenotypic evaluation of lines before selection as parents for the next cycle.
Model training and line development cycle length will be crop and breeding program specific. In a GS breeding schema, genome-wide DNA markers are used to predict which individuals in a breeding population are most valuable as parents of the next generation of offspring.
The prediction model is additionally continuously rejuvenated as genotypical and phenotypic data from elite lines derived from the collaborating breeding programs is incorporated into the prediction models.
In this manner, new germplasm may be infused into the system at any point. As lines derived from the recently infused germplasm advance within the breeding process, their genotypical and phenotypic data may be incorporated into the prediction models. The purpose of phenotyping now is to pick the best lines from a segregating population and to judge fewer lines with larger replication in every cycle of selection.
However during a GS driven breeding cycle, the aim of phenotyping is to estimate or re-estimate marker effects. It is far from clear at this point whether or not it will be advantageous to evaluate solely the best lines or to evaluate few lines with high replication.
All these small genes together explain the vast majority of the genetic variation for most traits Yang et al. In , Hayes and Goddard predicted 50— genes affected dairy traits, which was considered a high estimate at that time. Based on current GWAS and genomic selection results, we believe that dairy traits are affected by 2,—10, genes. Many genes affecting a trait implies that individual genes have small effects, which limits the efficiency of the MAS approach.
Three breakthroughs have resulted in the current widespread use of DNA information: 1 the GS methodology Meuwissen et al. In MAS, a small number of significant markers were used, and the rest were treated as having zero effect. Hence, the assumption that all SNPs have an effect may be approximately valid, and we should change our focus from significance testing to estimating the effects of all markers.
The second generation sequencing efforts that have resulted in discovery of the genome sequence of many of the livestock species have as a by-product revealed many thousands of SNP markers.
Next, selection candidates are genotyped, and by combining their genotypes with the estimated effects, genomic EBV GEBV are estimated for the selection candidates. In traditional breeding, the elite breeding animals were as accurately as possible trait and pedigree recorded. This potential to decouple accurate recording from the elite breeding population makes it possible to completely redesign the breeding scheme, and consequently GS has resulted in a paradigm shift in animal breeding.
Our goal here is to describe the GS method in more detail for a general scientific non-geneticists audience. In addition, we will describe current and predict future impacts of GS on dairy and beef cattle, pigs, and poultry breeding. A practical advantage of the GBLUP approach is that all the traditional BLUP methods and software can still be applied: we only need to replace pedigree by genomic relationships. The pedigree relationship between two fullsibs is 0. The latter requires genomic relationship estimates to be based on a sufficiently large number of SNPs.
For livestock and relationships within a breed, 50, SNPs distributed across the entire genome seems to suffice Goddard et al. Relationships across breeds are small and require a larger number of SNPs to be used. Since usually the number of genotyped animals is smaller than 50,, the GBLUP method is computationally preferred. In the future, the number of genotyped animals is expected to increase dramatically, so it may well be that the SNP-BLUP method becomes the method of choice. However, other, non-BLUP methods, may also gain popularity as shown in the following sections.
The prior distribution of SNP effects used by the nonlinear methods makes a lot more sense biologically than assuming that all SNPs have an effect and that all effects are very small. In computer simulation studies, the nonlinear methods clearly outperform GBLUP Meuwissen and Goddard, , but in real data, nonlinear methods are somewhat superior for some traits but not all Erbe et al. This may be explained by the following: 1 there are many genes affecting the economically important traits, so that assuming all SNPs are having an effect is approximately true; 2 linkage disequilibrium the non-random association between two loci extends over large genomic distances in livestock populations, such that many SNPs are associated with a gene; and 3 the SNP density is not high enough, so that each QTL can be explained by a single SNP and so many SNPs are needed to jointly explain the QTL effect.
The combination of explanations 1 and 2 , i. In the latter case, the causative mutations are expected to be present in the sequence data, and thus, GS can act on the causative mutations directly, instead of having to rely on LD between markers and causative mutations. However, these mutations are hidden among many millions of SNPs with no effect.
Second, current WGS data are not very accurate, either due to imperfect genotype calling, the extensive reliance on SNP imputation see next section , or structural genomic variations, which are difficult to assess by short reads of sequences. Third, long-range LD may be extensive in the reference population animals, causing large chromosomal segments or haplotypes to be common.
Consequently, there will be many combinations of SNPs that explain the effect of the haplotype as well as the causal mutations.
Each statistical method will chose a combination of SNP effects that best fits its prior assumptions, but they may all give the same prediction of the haplotype effect. However, if the range of the LD is reduced, e. Another problem is that present-day computers struggle to store and handle these massive amounts of data, especially if WGS is to be collected on many animals.
Despite current issues with the efficient use of WGS data, it is expected that WGS data will be the future's genotype data because, if the sequencing costs continue to fall, WGS may become the most effective genotyping method Gorjanc et al.
After SNP-chip genotyping, some of the genotypes will be missing. This is solved by a process called genotype imputation. Based on the known genotypes of the animals, the haplotype that the animal carries is recognized since the same haplotype was also observed in other animals. Thus, the missing genotype can be read from the genotype of these other animals, which carry the same haplotype.
Imputation methods can also be used in combination with sparse, but cheap, SNP chips. Key ancestors are genotyped with the dense, but expensive, chip to identify the haplotypes in the population. Next, large numbers of descendants are genotyped with a sparse, cheap SNP chip. The sparse chip has enough SNPs to recognize which of the haplotypes the animal carries.
Since the haplotypes are known at high density, the missing genotypes can be imputed. The same strategy is employed to obtain WGS data on many animals: the 1,bull-genome project Daetwyler et al. Next, many animals are genotyped with SNP chips, the bovine haplotypes that they carry are recognized, and their WGS data are imputed.
Another option is to sequence the descendants at low coverage. In this case, the low coverage sequence should be just enough to recognize the haplotypes Gorjanc et al. The 1, bull genomes project demonstrated that accurate imputation of sequence genotypes was possible for SNPs and other variants with high minor allele frequency.
For SNPs with low minor allele frequency however, accuracy of imputation was poor. Druet et al. In genomic selection, many probably most animals are not genotyped, but we need to include their phenotypic information in the breeding value estimation.
At least, traditional selection would use such information. One way to do this is by multiple-step GS: in step 1, pseudo-phenotypes are calculated for the genotyped animals where the pseudo-phenotype of animal i includes information records on its ungenotyped relatives; in step 2, genomic prediction is performed using the pseudo-records and their genotypes; and in step 3, the traditional EBV and GEBV are combined into a total EBV e. As an example of a pseudo-record, the average production of the daughters of a bull can be used.
Here, the bull is genotyped but not phenotyped whereas his daughters are phenotyped but not genotyped. Since the data are handled in multiple steps, this method is clearly suboptimal. However, in practice, good GS accuracies have been achieved using this method. An obvious idea is to replace pedigree with genomic relationships where available and retain the pedigree relationships where we do not have genomic relationships.
However, if genotyping shows that e. The correct relationship matrix can be obtained by starting with the genotyped animals and then using the pedigree to calculate relationships involving ungenotyped descendants of these genotyped animals, i. The same idea can also be used up the pedigree, i. A shortcoming of the single-step method is that it so far does not work for nonlinear estimation although some solutions to single-step nonlinear estimation have been proposed in the literature Liu et al.
In most studies, increases in reliability due to single step, over a pure genomic model, are small e. A more important feature of single-step models may be that they can account for pre-selection of young genotyped bulls, which could otherwise cause bias in the GEBV Vitezica et al. Until recently, the requirement that the G matrix must be inverted directly limited the size of the dataset to which ssBLUP could be applied.
For the future, there is a clear need for a single-step method that uses a nonlinear statistical method on sequence level data. The accuracy of genomic prediction in dairy cattle exceeds 0. These high accuracies reflect the large reference populations for each breed that have been assembled to enable genomic predictions and the fact that many of the animals in the reference populations are progeny-tested bulls with highly accurate phenotypes from average daughter performance. In addition, the GEBV are often used to predict close relatives of animals in the reference population.
The high accuracies of genomic prediction and relatively low cost of obtaining the genomic predictions from low-density genotyping followed by imputation, has resulted in very large numbers of selection candidates being genotyped. Worldwide, approximately 2 million dairy cattle have now been genotyped for the purposes of genomic prediction.
Similar numbers of animals have been genotyped by other countries combined, including , in France alone Boichard, personal communication. Implementing genomic selection in dairy cattle has resulted in increased genetic gain, which has now been demonstrated by genetic trend analysis in a number of countries. For example, in Canada, the rate of genetic gain has approximately doubled since genomic selection was introduced VanDoormal, personal communication.
There is also some suggestion that genomic selection has increased the rate of inbreeding per year Schenkel, Maximizing genetic gain from genomic selection while constraining the rate of inbreeding will therefore be an important topic for future research.
Interestingly, the majority of the genotyped animals in many countries are now heifer calves. While genotyping young bull calves results in the greatest genetic gain, genotyping is now sufficiently cheap that genotyping heifer calves for the purposes of choosing which heifers to retain in the herd is profitable Pryce and Hayes ; Weigel et al.
The genotypes of the heifers can also be used when choosing bulls to which to mate them so that inbreeding of the resulting calf can be minimized. When the selected heifers enter the herd and have herd recording data, they can be used in the reference population for genomic prediction Wiggans et al. When the aim is to increase the size of the reference population to improve accuracy of genomic prediction, genotyping mature cows with good phenotypic records can help— Kemper et al.
This relatively large increase probably reflects the smaller bull reference sets 4, and 1, for Holsteins and Jerseys, respectively compared with some of the populations above. In some beef breeds, genomic selection is now applied on a large scale.
In general, however, accuracies of genomic predictions in beef cattle have been lower than in dairy cattle. For instance, in their review, Van Eenennaam et al. The lower accuracy is because the reference populations are of higher quality in dairy cattle. In beef cattle, the reference population contains fewer animals within a breed, and these animals have not been progeny tested.
In addition, the target population and validation animals may be less closely related to the reference population in beef cattle than in dairy cattle. To compensate for the small number of reference animals within a breed, it is not uncommon to use a multi-breed reference population. Bolormaa et al. When Akanno et al. If the target breed is not included in the reference population, the accuracy is very low.
These disappointing results for prediction across breeds are not unexpected. De Roos et al. Therefore, when using a 50k SNP panel, information from another breed is not expected to increase accuracy. Even if high density SNPs are used, information from another breed is much less useful than information from the target breed because animals of different breeds share much smaller chromosome segments than animals of the same breed. When BLUP is used to predict breeding values, the variance for a chromosome segment is assumed to be proportional to its length or number of SNPs , and so small segments have lower variance and are estimated less accurately than larger segments.
The situation is improved a little by using Bayesian methods that allow some SNPs and therefore some segments to have a larger effect than others. The value of combining breeds in a reference population depends to some extent on QTL segregating in multiple breeds.
However, Bolormaa et al. Genomic selection has not been adopted as widely in beef as in dairy cattle breeding.
This is partly because the accuracy is lower, but also because the economic advantages are not as great. Genomic selection is most advantageous for traits that are difficult to select for traditionally.
It is less advantageous in beef than dairy because progeny testing is not needed for traits that can be measured on selection candidates at a young age such as growth rate. However, several important traits in beef cattle are difficult to select for such as feed conversion efficiency and beef quality. Because these traits are also expensive to record, it is costly to set up a large training population and there are no large companies that could justify this cost for their own breeding program.
For these traits, a multi-breed training population and nonlinear analysis based on high-density SNPs or genome sequence data may be the best approach.
Despite these difficulties, genomic selection has been implemented in beef cattle. There are two ways in which this can be done. First, the genotypes can be provided to the organization that calculates EBVs who then calculate the prediction equation.
Both methods are in operation. The advantage of the first method is that the full dataset of phenotypes and genotypes can be used to derive the prediction equation and the DNA information can be fully integrated through one-step genomic prediction methods e. In pig breeding, the most important selection step is the selection of elite boars in the nucleus herd. This may be at the boar test station in the case of cooperating pig breeders.
The implementation of GS in pig breeding is therefore mainly directed at traits whose recording is invasive such as slaughter quality, maternal traits that cannot be recorded on the boars, and crossbred performance, which cannot be recorded on the purebred animals.
With respect to the maternal traits, female sibs of the test boars are raised in nucleus herds, but their maternal trait recordings become available after the selection of the boars. However, GS for maternal traits can be based on aunts of the test boars.
0コメント