Deterministic models of breeding scheme designs that incorporate genomic selection

Potential response from selection schemes based on progeny testing and genomic selection for the Chilean dairy cattle under pastoral systems: a deterministic simulation

Recently, a selection index called Valor Económico Lechero (VEL) was developed for Chilean dairy cattle under pasture. However, a specific selection scheme has not yet been implemented. This study aimed to estimate genetic progress from selection on the VEL selection index based on selection schemes using progeny testing (PT) and genomic selection (GS). Under a PT-scheme, estimated genetic progress was 41.50, 3.44, and 2.33 kg/year for milk, fat, and protein yield, respectively. The realised genetic gain takes eight-year after the PT-scheme implementation, which may be a disincentive for implementing a PT-scheme, suggesting that importing frozen semen of proven bulls could be a preferred alternative. In this case, an option may be to conduct the genetic evaluation of those bulls using their progeny in Chile for the traits included in VEL selection index. In the case of implementing a specific selection scheme, compared to PT, a more profitable alternative might be the implementation of a GS-scheme, that would result in a faster genetic gain in the aggregate breeding value or merit for all the traits included in the selection objective (0.323-0.371 vs. 0.194 σg/year).

Simulation studies to optimize genomic selection in honey bees

Background

With the completion of a single nucleotide polymorphism (SNP) chip for honey bees, the technical basis of genomic selection is laid. However, for its application in practice, methods to estimate genomic breeding values need to be adapted to the specificities of the genetics and breeding infrastructure of this species. Drone-producing queens (DPQ) are used for mating control, and usually, they head non-phenotyped colonies that will be placed on mating stations. Breeding queens (BQ) head colonies that are intended to be phenotyped and used to produce new queens. Our aim was to evaluate different breeding program designs for the initiation of genomic selection in honey bees.

Methods

Stochastic simulations were conducted to evaluate the quality of the estimated breeding values. We developed a variation of the genomic relationship matrix to include genotypes of DPQ and tested different sizes of the reference population. The results were used to estimate genetic gain in the initial selection cycle of a genomic breeding program. This program was run over six years, and different numbers of genotyped queens per year were considered. Resources could be allocated to increase the reference population, or to perform genomic preselection of BQ and/or DPQ.

Results

Including the genotypes of 5000 phenotyped BQ increased the accuracy of predictions of breeding values by up to 173%, depending on the size of the reference population and the trait considered. To initiate a breeding program, genotyping a minimum number of 1000 queens per year is required. In this case, genetic gain was highest when genomic preselection of DPQ was coupled with the genotyping of 10–20% of the phenotyped BQ. For maximum genetic gain per used genotype, more than 2500 genotyped queens per year and preselection of all BQ and DPQ are required.

Conclusions

This study shows that the first priority in a breeding program is to genotype phenotyped BQ to obtain a sufficiently large reference population, which allows successful genomic preselection of queens. To maximize genetic gain, DPQ should be preselected, and their genotypes included in the genomic relationship matrix. We suggest, that the developed methods for genomic prediction are suitable for implementation in genomic honey bee breeding programs.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12711-021-00654-x.

Genomic Selection for Any Dairy Breeding Program via Optimized Investment in Phenotyping and Genotyping

This paper evaluates the potential of maximizing genetic gain in dairy cattle breeding by optimizing investment into phenotyping and genotyping. Conventional breeding focuses on phenotyping selection candidates or their close relatives to maximize selection accuracy for breeders and quality assurance for producers. Genomic selection decoupled phenotyping and selection and through this increased genetic gain per year compared to the conventional selection. Although genomic selection is established in well-resourced breeding programs, small populations and developing countries still struggle with the implementation. The main issues include the lack of training animals and lack of financial resources. To address this, we simulated a case-study of a small dairy population with a number of scenarios with equal available resources yet varied use of resources for phenotyping and genotyping. The conventional progeny testing scenario collected 11 phenotypic records per lactation. In genomic selection scenarios, we reduced phenotyping to between 10 and 1 phenotypic records per lactation and invested the saved resources into genotyping. We tested these scenarios at different relative prices of phenotyping to genotyping and with or without an initial training population for genomic selection. Reallocating a part of phenotyping resources for repeated milk records to genotyping increased genetic gain compared to the conventional selection scenario regardless of the amount and relative cost of phenotyping, and the availability of an initial training population. Genetic gain increased by increasing genotyping, despite reduced phenotyping. High-genotyping scenarios even saved resources. Genomic selection scenarios expectedly increased accuracy for young non-phenotyped candidate males and females, but also proven females. This study shows that breeding programs should optimize investment into phenotyping and genotyping to maximize return on investment. Our results suggest that any dairy breeding program using conventional progeny testing with repeated milk records can implement genomic selection without increasing the level of investment.

Genetic and economic benefits of foreign sire contributions to a domestic sheep industry; including an Ireland-New Zealand case study

Background

Importation of foreign genetics is a widely used genetic improvement strategy. However, even if the foreign genetic merit is currently greater than the domestic genetic merit, differences in foreign and domestic trends mean that the long-term competitiveness of an importation strategy cannot be guaranteed. Gene flow models are used to quantify the impact that a specific subpopulation, such as foreign genetics, can have over time on the genetic or economic benefit of a domestic industry.

Methods

We used a deterministic recursive gene flow model to predict the commercial performance of lambs born across various subpopulations. Numerous breeding strategies were evaluated by varying market share, proportions of rams selected for mating, genetic trend, superiority of foreign genetics over domestic genetics and frequency of importation. Specifically, an Ireland-New Zealand case study was simulated to quantify the potential gain that could be made by using foreign sire contributions (New Zealand) in a domestic sheep industry (Ireland).

Results

Genetic and economic gains were generated from alternative breeding strategies. The ‘base scenario’ (i.e. representing the current industry) predicted an average genetic merit value of €2.51 for lambs born and an annualised cumulative benefit of €45 million (m) after 20 years. Maximum genetic (€9.45 for lambs born) and economic (annualised cumulative benefit of €180 m after 20 years) benefits were achieved by implementing the ‘PRO-intense-market scenario’ which involved shifting market share away from conservative domestic breeders and reducing the proportion of rams that were selected for mating by progressive domestic breeders from the top 40% to the top 20%, without the use of any foreign genetics. The ‘PROFOR scenario’, which considered the use of foreign and progressive domestic genetics, predicted an average genetic merit value of €7.37 for lambs born and an annualised cumulative benefit of €144 m, after 20 years.

Conclusions

Our results demonstrate that there is opportunity for a domestic industry to increase industry benefits without the use of foreign genetics but through an attempt to shift the market share away from conservative domestic breeders towards progressive domestic breeders. However, the importation and use of progressive foreign genetics may be an effective method to trigger a change in behaviour of conservative domestic breeders towards the use of progressive genetics.

Intensified Use of Reproductive Technologies and Reduced Dimensions of Breeding Schemes Put Genetic Diversity at Risk in Dairy Cattle Breeds

In the management of dairy cattle breeds, two recent trends have arisen that pose potential threats to genetic diversity: the use of reproductive technologies (RT) and a reduction in the number of bulls in breeding schemes. The expected outcome of these changes, in terms of both genetic gain and genetic diversity, is not trivial to predict. Here, we simulated 15 breeding schemes similar to those carried out in large French dairy cattle breeds; breeding schemes differed with respect to their dimensions, the intensity of RT use, and the type of RT involved. We found that intensive use of RT resulted in improved genetic gain, but deteriorated genetic diversity. Specifically, a reduction in the interval between generations through the use of ovum pick-up and in vitro fertilization (OPU-IVF) resulted in a large increase in the inbreeding rate both per year and per generation, suggesting that OPU-IVF could have severe adverse effects on genetic diversity. To achieve a given level of genetic gain, the scenarios that best maintained genetic diversity were those with a higher number of sires/bulls and a medium intensity of RT use or those with a higher number of female donors to compensate for the increased intensity of RT.

A review of traditional and machine learning methods applied to animal breeding

The current livestock management landscape is transitioning to a high-throughput digital era where large amounts of information captured by systems of electro-optical, acoustical, mechanical, and biosensors is stored and analyzed on a daily and hourly basis, and actionable decisions are made based on quantitative and qualitative analytic results. While traditional animal breeding prediction methods have been used with great success until recently, the deluge of information starts to create a computational and storage bottleneck that could lead to negative long-term impacts on herd management strategies if not handled properly. A plethora of machine learning approaches, successfully used in various industrial and scientific applications, made their way in the mainstream approaches for livestock breeding techniques, and current results show that such methods have the potential to match or surpass the traditional approaches, while most of the time they are more scalable from a computational and storage perspective. This article provides a succinct view on what traditional and novel prediction methods are currently used in the livestock breeding field, how successful they are, and how the future of the field looks in the new digital agriculture era.

Efficient use of genomic information for sustainable genetic improvement in small cattle populations

In this study, we compared genetic gain, genetic variation, and the efficiency of converting variation into gain under different genomic selection scenarios with truncation or optimum contribution selection in a small dairy population by simulation. Breeding programs have to maximize genetic gain but also ensure sustainability by maintaining genetic variation. Numerous studies have shown that genomic selection increases genetic gain. Although genomic selection is a well-established method, small populations still struggle with choosing the most sustainable strategy to adopt this type of selection. We developed a simulator of a dairy population and simulated a model after the Slovenian Brown Swiss population with ∼10,500 cows. We compared different truncation selection scenarios by varying (1) the method of sire selection and their use on cows or bull-dams, and (2) selection intensity and the number of years a sire is in use. Furthermore, we compared different optimum contribution selection scenarios with optimization of sire selection and their usage. We compared scenarios in terms of genetic gain, selection accuracy, generation interval, genetic and genic variance, rate of coancestry, effective population size, and conversion efficiency. The results showed that early use of genomically tested sires increased genetic gain compared with progeny testing, as expected from changes in selection accuracy and generation interval. A faster turnover of sires from year to year and higher intensity increased the genetic gain even further but increased the loss of genetic variation per year. Although maximizing intensity gave the lowest conversion efficiency, faster turnover of sires gave an intermediate conversion efficiency. The largest conversion efficiency was achieved with the simultaneous use of genomically and progeny-tested sires that were used over several years. Compared with truncation selection, optimizing sire selection and their usage increased the conversion efficiency by achieving either comparable genetic gain for a smaller loss of genetic variation or higher genetic gain for a comparable loss of genetic variation. Our results will help breeding organizations implement sustainable genomic selection.

Genotyping strategies of selection candidates in livestock breeding programmes

Benefits of genomic selection (GS) in livestock breeding operations are well known particularly where traits are sex-limited, hard to measure, have a low heritability and/or measured later in life. Sheep and beef breeders have a higher cost:benefit ratio for GS compared to dairy. Therefore, strategies for genotyping selection candidates should be explored to maximize the economic benefit of GS. The aim of the paper was to investigate, via simulation, the additional genetic gain achieved by selecting proportions of male selection candidates to be genotyped via truncation selection. A two-trait selection index was used that contained an easy and early-in-life measurement (such as post-weaning weight) as well as a hard-to-measure trait (such as intra-muscular fat). We also evaluated the optimal proportion of female selection candidates to be genotyped in breeding programmes using natural mating and/or artificial insemination (NatAI), multiple ovulation and embryo transfer (MOET) or juvenile in vitro fertilization and embryo transfer (JIVET). The final aim of the project was to investigate the total dollars spent to increase the genetic merit by one genetic standard deviation (SD) using GS and/or reproductive technologies. For NatAI and MOET breeding programmes, females were selected to have progeny by 2 years of age, while 1-month-old females were required for JIVET. Genomic testing the top 20% of male selection candidates achieved 80% of the maximum benefit from GS when selection of male candidates prior to genomic testing had an accuracy of 0.36, while 54% needed to be tested to get the same benefit when the prior selection accuracy was 0.11. To achieve 80% of the maximum benefit in female, selection required 66%, 47% and 56% of female selection candidates to be genotyped in NatAI, MOET and JIVET breeding programmes, respectively. While JIVET and MOET breeding programmes achieved the highest annual genetic gain, genotyping male selection candidates provides the most economical way to increase rates of genetic gain facilitated by genomic testing.

Optimizing female allocation to reproductive technologies considering merit, inbreeding and cost in nucleus breeding programmes with genomic selection

Female reproductive technologies such as multiple ovulation and embryo transfer (MOET) and juvenile in vitro fertilization and embryo transfer (JIVET) have been shown to accelerate genetic gain by increasing selection intensity and decreasing generation interval. Genomic selection (GS) increases the accuracy of selection of young candidates which can further accelerate genetic gain. Optimal contribution selection (OCS) is an effective method of keeping the rate of inbreeding at a sustainable level while increasing genetic merit. OCS could also be used to selectively and optimally allocate reproductive technologies in mate selection while accounting for their cost. This study uses stochastic simulation to simulate breeding programmes that use a combination of artificial insemination (AI) or natural mating (N), MOET and JIVET with GS. OCS was used to restrict inbreeding to 1.0% increase per generation and also to optimize use of reproductive technologies, considering their effect on genetic gain as well as their cost. Two Australian sheep breeding objectives were used as an example to illustrate the methodology-a terminal sire breeding objective (A) and a dual-purpose self-replacing breeding objective (B). The objective function used for optimization considered genetic merit, constrained inbreeding and cost of technologies where costs were offset by a premium paid to the seedstock breeder investing in female reproductive technologies. The premium was based on the cumulative discounted expression of genetic merit in the progeny of a commercial tier in the breeding programme multiplied by the proportion of that benefit received by the breeder. With breeding objective B, the highest premium of 64% paid to the breeder resulted in the highest allocation of reproductive technologies (4%-10% for MOET and 19%-54% for JIVET) and hence the highest annual genetic gain. Conversely, breeding objective A, which had a lower dollar value of the breeding objective and a maximum of 5% mating types for JIVET and zero for MOET were optimal, even when highest premiums were paid. This study highlights that the level of investment in breeding technologies to accelerate genetic gain depends on the investment of genetic improvement returned to the breeder per index point gain achieved. It also demonstrates that breeding programmes can be optimized including allocation of reproductive technologies at the individual animal level. Accounting for revenue to the breeder and cost of the technologies can facilitate more practical decision support for beef and sheep breeders.

The cost-benefit of genomic testing of heifers and using sexed semen in pasture-based dairy herds

Recent improvements in dairy cow fertility and female reproductive technologies offer an opportunity to apply greater selection pressure to females. This means there may be greater incentive to obtain genomic breeding values for females. We modeled the impact of changes to key parameters on the net benefit from genomic testing of heifer calves with and without usage of sexed semen. This paper builds on earlier cost-benefit studies but uses parameters relevant to pasture-based systems. A deterministic model was used to evaluate the effect on net benefit due to changes in (1) reproduction rate, (2) genomic test costs, (3) availability of parent-derived breeding values (EBV_PA), and (4) replacement rate. When the use of sexed semen was included, we also considered (1) the proportion of heifers and cows mated to sexed semen, (2) decreases in conception rate in inseminations with sexed semen, and (3) the marginal return for surplus heifers. Scenarios with lower replacement rates and no availability of EBV_PA had the largest net benefits. Under current Australian parameters, the net benefit of genomic testing realized over the lifetime of genotyped heifers is expected to range from A$204 to A$1,124 per 100 cows for a herd with median reproductive performance. The cost of a genomic test, a perceived barrier to many farmers, had only a small effect on net benefit. Genomic testing alone was always more profitable than using sexed semen and genomic testing together if the only benefit considered was increased genetic gain in heifer replacements. When other benefits (i.e., the higher sale price of a surplus heifer compared with a male calf) were considered, there were combinations of parameters where net benefit from using sexed semen and genomic testing was higher than the equivalent scenario with genomic testing only. Using sexed semen alongside genomic testing is most likely to be profitable when (1) used in heifers, (2) the marginal return for selling surplus heifers (sale price minus rearing costs) is greater than A$400, and (3) conception rates of no more than 10 percentage points lower than those achieved using conventional semen can be realized. Net benefit was highly dependent on the marginal return. Demonstrating that the initial investment in genomic testing can be recouped within the lifetime of the heifers tested may assist in the development of extension messages to explain the value of genomic testing females at the herd level.

An analytical framework to derive the expected precision of genomic selection

Background

Formulae to predict the precision or accuracy of genomic estimated breeding values (GEBV) are important when modelling selection schemes. Simple versions of such formulae have been proposed in the past, based on a number of simplifying hypotheses, including absence of linkage disequilibrium and linkage between loci, a population made up of unrelated individuals, and that all genetic variability of the trait is explained by the genotyped loci. These formulae were based on approximations that were not always clear. The objective of this paper is to offer a unique framework to derive equations that predict the precision of GEBV from the size of the reference population and the heritability of and number of QTL controlling the quantitative trait.

Results

The exact formulation of the precision of GEBV involves the expectation of the inverse of a linear function of the genomic matrix, which cannot be calculated from simple algebra but can be approximated using a Taylor polynomial expansion. First order approximations performed better than the initial prediction equations published in the literature. Second order approximations produced almost perfect estimates of precision when compared to results obtained when simulating situations that agreed with the assumptions that were required to derive the precision equations. Using this proposed framework, we present several generalizations, including multi-trait genomic evaluation.

Conclusions

Although further improvements are needed to account for the complexity of practical situations, the equations proposed here can be used to derive the precision of GEBV when comparing breeding schemes a priori.

Electronic supplementary material

The online version of this article (10.1186/s12711-017-0366-6) contains supplementary material, which is available to authorized users.

Economic and genetic performance of various combinations of in vitro-produced embryo transfers and artificial insemination in a dairy herd

The objective of this study was to find the optimal proportions of pregnancies from an in vitro-produced embryo transfer (IVP-ET) system and artificial insemination (AI) so that profitability is maximized over a range of prices for embryos and surplus dairy heifer calves. An existing stochastic, dynamic dairy model with genetic merits of 12 traits was adapted for scenarios where 0 to 100% of the eligible females in the herd were impregnated, in increments of 10%, using IVP-ET (ET0 to ET100, 11 scenarios). Oocytes were collected from the top donors selected for the trait lifetime net merit (NM$) and fertilized with sexed semen to produce IVP embryos. Due to their greater conception rates, first ranked were eligible heifer recipients based on lowest number of unsuccessful inseminations or embryo transfers, and then on age. Next, eligible cow recipients were ranked based on the greatest average estimated breeding values (EBV) of the traits cow conception rate and daughter pregnancy rate. Animals that were not recipients of IVP embryos received conventional semen through AI, except that the top 50% of heifers ranked for EBV of NM$ were inseminated with sexed semen for the first 2 AI. The economically optimal proportions of IVP-ET were determined using sensitivity analysis performed for 24 price sets involving 6 different selling prices of surplus dairy heifer calves at approximately 105 d of age and 4 different prices of IVP embryos. The model was run for 15 yr after the start of the IVP-ET program for each scenario. The mean ± standard error of true breeding values of NM$ of all cows in the herd in yr 15 was greater by $603 ± 2 per cow per year for ET100 when compared with ET0. The optimal proportion of IVP-ET ranged from ET100 (for surplus dairy heifer calves sold for ≥$300 along with an additional premium based on their EBV of NM$ and a ≤$100 embryo price) to as low as ET0 (surplus dairy heifer calves sold at $300 with a $200 embryo price). For the default assumptions, the profit/cow in yr 15 was greater by $337, $215, $116, and $69 compared with ET0 when embryo prices were $50, $100, $150, and $200. The optimal use of IVP-ET was 100, 100, 62, and 36% of all breedings for these embryo prices, respectively. At the input price of $165 for an IVP embryo, the difference in the net present value of yr 15 profit between ET40 (optimal scenario) and ET0 was $33 per cow. In conclusion, some use of IVP-ET was profitable for a wide range of IVP-ET prices and values of surplus dairy heifer calves.

Increased genetic gains in multi-trait sheep indices using female reproductive technologies combined with optimal contribution selection and genomic breeding values

Female reproductive technologies such as multiple ovulation and embryo transfer (MOET) and juvenile in vitro fertilisation and embryo transfer (JIVET) can produce multiple offspring per mating in sheep and cattle. In breeding programs this allows for higher female selection intensity and, in the case of JIVET, a reduction in generation interval, resulting in higher rates of genetic gain. Low selection accuracy of young females entering JIVET has often dissuaded producers from using this technology. However, genomic selection (GS) could increase selection accuracy of candidates at a younger age to help increase rates of genetic gain. This increase might vary for different traits in multiple trait breeding programs depending on genetic parameters and the practicality of recording, particularly for hard to measure traits. This study used both stochastic (animals) and deterministic (GS) simulation to evaluate the effect of reproductive technologies on the genetic gain for various traits in sheep breeding programs, both with and without GS. Optimal contribution selection was used to manage inbreeding and to optimally assign reproductive technologies to individual selection candidates. Two Australian sheep industry indexes were used – a terminal sire index that focussed on growth and carcass traits (the ‘Lamb 2020’ index) and a Merino index that focuses on wool traits, bodyweight, and reproduction (MP+). We observed that breeding programs using artificial insemination or natural mating (AI/N) + MOET, compared with AI/N alone, yielded an extra 39% and 27% genetic gain for terminal and Merino indexes without GS, respectively. However, the addition of JIVET to AI/N + MOET without GS only yielded an extra 1% genetic gain for terminal index and no extra gain in the Merino index. When GS was used in breeding programs, we observed AI/N + MOET + JIVET outperformed AI/N + MOET by 21% and 33% for terminal and Merino indexes, respectively. The implementation of GS increased genetic gain where reproductive technologies were used by 9–34% in Lamb 2020 and 37–98% in MP+. Individual trait response to selection varied in each breeding program. The combination of GS and reproductive technologies allowed for greater genetic gain in both indexes especially for hard to measure traits, but had limited effect on the traits that already had a large amount of early age records.

Changes in genetic selection differentials and generation intervals in US Holstein dairy cattle as a result of genomic selection

Seven years after the introduction of genomic selection in the United States, it is now possible to evaluate the impact of this technology on the population. Selection differential(s) (SD) and generation interval(s) (GI) were characterized in a four-path selection model that included sire(s) of bulls (SB), sire(s) of cows (SC), dam(s) of bulls (DB), and dam(s) of cows (DC). Changes in SD over time were estimated for milk, fat, and protein yield; somatic cell score (SCS); productive life (PL); and daughter pregnancy rate (DPR) for the Holstein breed. In the period following implementation of genomic selection, dramatic reductions were seen in GI, especially the SB and SC paths. The SB GI reduced from ∼7 y to less than 2.5 y, and the DB GI fell from about 4 y to nearly 2.5 y. SD were relatively stable for yield traits, although modest gains were noted in recent years. The most dramatic response to genomic selection was observed for the lowly heritable traits DPR, PL, and SCS. Genetic trends changed from close to zero to large and favorable, resulting in rapid genetic improvement in fertility, lifespan, and health in a breed where these traits eroded over time. These results clearly demonstrate the positive impact of genomic selection in US dairy cattle, even though this technology has only been in use for a short time. Based on the four-path selection model, rates of genetic gain per year increased from ∼50-100% for yield traits and from threefold to fourfold for lowly heritable traits.

Random forest estimation of genomic breeding values for disease susceptibility over different disease incidences and genomic architectures in simulated cow calibration groups

A simulation study was conducted to investigate the performance of random forest (RF) and genomic BLUP (GBLUP) for genomic predictions of binary disease traits based on cow calibration groups. Training and testing sets were modified in different scenarios according to disease incidence, the quantitative-genetic background of the trait (h(2)=0.30 and h(2)=0.10), and the genomic architecture [725 quantitative trait loci (QTL) and 290 QTL, populations with high and low levels of linkage disequilibrium (LD)]. For all scenarios, 10,005 SNP (depicting a low-density 10K SNP chip) and 50,025 SNP (depicting a 50K SNP chip) were evenly spaced along 29 chromosomes. Training and testing sets included 20,000 cows (4,000 sick, 16,000 healthy, disease incidence 20%) from the last 2 generations. Initially, 4,000 sick cows were assigned to the testing set, and the remaining 16,000 healthy cows represented the training set. In the ongoing allocation schemes, the number of sick cows in the training set increased stepwise by moving 10% of the sick animals from the testing set to the training set, and vice versa. The size of the training and testing sets was kept constant. Evaluation criteria for both GBLUP and RF were the correlations between genomic breeding values and true breeding values (prediction accuracy), and the area under the receiving operating characteristic curve (AUROC). Prediction accuracy and AUROC increased for both methods and all scenarios as increasing percentages of sick cows were allocated to the training set. Highest prediction accuracies were observed for disease incidences in training sets that reflected the population disease incidence of 0.20. For this allocation scheme, the largest prediction accuracies of 0.53 for RF and of 0.51 for GBLUP, and the largest AUROC of 0.66 for RF and of 0.64 for GBLUP, were achieved using 50,025 SNP, a heritability of 0.30, and 725 QTL. Heritability decreases from 0.30 to 0.10 and QTL reduction from 725 to 290 were associated with decreasing prediction accuracy and decreasing AUROC for all scenarios. This decrease was more pronounced for RF. Also, the increase of LD had stronger effect on RF results than on GBLUP results. The highest prediction accuracy from the low LD scenario was 0.30 from RF and 0.36 from GBLUP, and increased to 0.39 for both methods in the high LD population. Random forest successfully identified important SNP in close map distance to QTL explaining a high proportion of the phenotypic trait variations.

Prediction of genetic gain in finite populations with heterogeneous predicted breeding values accuracies

The algebraic expression of the genetic selection differential (expected genetic superiority of breeders after a selection on their Predicted Breeding Values) was derived when a limited number of individuals were selected from a limited sample of candidates on the basis of their predicted genetic value, with heterogeneous reliabilities. A formula is proposed for situations in which these reliabilities can be clustered in a few classes. We show that the expected genetic selection differential increases with the number of classes, the mean reliability being constant. In the panel of cases simulated, this increase reached up to 18% of the values obtained in the homogeneous situation. We used the proposed formulae to estimate selection differentials and compared it numerically with performing simulations. In terms of speed of computation, our algebraic formulae performed better than simulations in populations of limited size.

Approximated prediction of genomic selection accuracy when reference and candidate populations are related

Background

Genomic selection is still to be evaluated and optimized in many species. Mathematical modeling of selection schemes prior to their implementation is a classical and useful tool for that purpose. These models include formalization of a number of entities including the precision of the estimated breeding value. To model genomic selection schemes, equations that predict this reliability as a function of factors such as the size of the reference population, its diversity, its genetic distance from the group of selection candidates genotyped, number of markers and strength of linkage disequilibrium are needed. The present paper aims at exploring new approximations of this reliability.

Results

Two alternative approximations are proposed for the estimation of the reliability of genomic estimated breeding values (GEBV) in the case of non-independence between candidate and reference populations. Both were derived from the Taylor series heuristic approach suggested by Goddard in 2009. A numerical exploration of their properties showed that the series were not equivalent in terms of convergence to the exact reliability, that the approximations may overestimate the precision of GEBV and that they converged towards their theoretical expectations. Formulae derived for these approximations were simple to handle in the case of independent markers. A few parameters that describe the markers’ genotypic variability (allele frequencies, linkage disequilibrium) can be estimated from genomic data corresponding to the population of interest or after making assumptions about their distribution. When markers are not in linkage equilibrium, replacing the real number of markers and QTL by the “effective number of independent loci”, as proposed earlier is a practical solution. In this paper, we considered an alternative, i.e. an “equivalent number of independent loci” which would give a GEBV reliability for unrelated individuals by considering a sub-set of independent markers that is identical to the reliability obtained by considering the full set of markers.

Conclusions

This paper is a further step towards the development of deterministic models that describe breeding plans based on the use of genomic information. Such deterministic models carry low computational burden, which allows design optimization through intensive numerical exploration.

Electronic supplementary material

The online version of this article (doi:10.1186/s12711-016-0183-3) contains supplementary material, which is available to authorized users.

Increased genetic gains in sheep, beef and dairy breeding programs from using female reproductive technologies combined with optimal contribution selection and genomic breeding values

Background

Female reproductive technologies such as multiple ovulation and embryo transfer (MOET) and juvenile in vitro embryo production and embryo transfer (JIVET) can boost rates of genetic gain but they can also increase rates of inbreeding. Inbreeding can be managed using the principles of optimal contribution selection (OCS), which maximizes genetic gain while placing a penalty on the rate of inbreeding. We evaluated the potential benefits and synergies that exist between genomic selection (GS) and reproductive technologies under OCS for sheep and cattle breeding programs.

Methods

Various breeding program scenarios were simulated stochastically including: (1) a sheep breeding program for the selection of a single trait that could be measured either early or late in life; (2) a beef breeding program with an early or late trait; and (3) a dairy breeding program with a sex limited trait. OCS was applied using a range of penalties (severe to no penalty) on co-ancestry of selection candidates, with the possibility of using multiple ovulation and embryo transfer (MOET) and/or juvenile in vitro embryo production and embryo transfer (JIVET) for females. Each breeding program was simulated with and without genomic selection.

Results

All breeding programs could be penalized to result in an inbreeding rate of 1 % increase per generation. The addition of MOET to artificial insemination or natural breeding (AI/N), without the use of GS yielded an extra 25 to 60 % genetic gain. The further addition of JIVET did not yield an extra genetic gain. When GS was used, MOET and MOET + JIVET programs increased rates of genetic gain by 38 to 76 % and 51 to 81 % compared to AI/N, respectively.

Conclusions

Large increases in genetic gain were found across species when female reproductive technologies combined with genomic selection were applied and inbreeding was managed, especially for breeding programs that focus on the selection of traits measured late in life or that are sex-limited. Optimal contribution selection was an effective tool to optimally allocate different combinations of reproductive technologies. Applying a range of penalties to co-ancestry of selection candidates allows a comprehensive exploration of the inbreeding vs. genetic gain space.

Recent developments in cattle, pig, sheep and horse breeding - a review

The aim of this review was to summarize new genetic approaches and techniques in the breeding of cattle, pigs, sheep and horses. Often production and reproductive traits are treated separately in genetic evaluations, but advantages may accrue to their joint evaluation. A good example is the system in pig breeding. Simplified breeding objectives are generally no longer appropriate and consequently becoming increasingly complex. The goal of selection for improved animal performance is to increase the profit of the production system; therefore, economic selection indices are now used in most livestock breeding programmes. Recent developments in dairy cattle breeding have focused on the incorporation of molecular information into genetic evaluations and on increasing the importance of longevity and health in breeding objectives to maximize the change in profit. For a genetic evaluation of meat yield (beef, pig, sheep), several types of information can be used, including data from performance test stations, records from progeny tests and measurements taken at slaughter. The standard genetic evaluation method of evaluation of growth or milk production has been the multi-trait animal model, but a test-day model with random regression is becoming the new standard, in sheep as well. Reviews of molecular genetics and pedigree analyses for performance traits in horses are described. Genome – wide selection is becoming a world standard for dairy cattle, and for other farm animals it is under development.

Genetic variances of SNP loci for milk yield in dairy cattle

Regression coefficients and genetic variances for 40,890 single nucleotide polymorphisms (SNPs) for milk yield were calculated using mixed model equations, with deregressed proof (DRP) as the dependent variable. Bulls were genotyped using the Illumina BovineSNP50 v2 BeadChip and SNPs were edited according the minor allele frequency (MAF) and high incidence of missing genotype. Evaluation was conducted in two rounds. In the preliminary round, the direct genetic values (DGVs) of all genotyped bulls (2,904) were computed and the absolute difference between the DGV and the input DRP of each bull was investigated. Bulls with an absolute difference greater than the mean absolute difference plus two standard deviations were eliminated from the data set prior to the final analysis (2,766 bulls remaining). SNP regression coefficients from the final analysis had a mean absolute value of 0.506 kg and a standard deviation of 0.409 kg. The SNP with the highest regression coefficient and genetic variance was ARSBFGLNGS4939 on chromosome 14. This SNP is located within the gene DGAT1 (diacylglycerol O-acyltransferase 1). Other SNPs with high regression coefficients and genetic variance are localised in proximity to DGAT1. The mean genetic variance of an individual SNP was 0.170, with a standard deviation of 0.384 and a mean heterozygosity of 0.372. The sum of genetic variances of all SNPs was only 6,968.8, probably because of the existence of genetic covariances between loci. The largest sum of genetic variances was on chromosome 14 (498.4, 7.15 % of the total). After the final analysis, the correlation between the DGV and the input DRP was 0.951 for all bulls. The variance of the predicted DGV was 98.11 % of the variance of the input estimated breeding value (EBV) and 63.65 % of the variance of the DRP.

Assessing the impact of natural service bulls and genotype by environment interactions on genetic gain and inbreeding in organic dairy cattle genomic breeding programs

The objective of the present study was to compare genetic gain and inbreeding coefficients of dairy cattle in organic breeding program designs by applying stochastic simulations. Evaluated breeding strategies were: (i) selecting bulls from conventional breeding programs, and taking into account genotype by environment (G×E) interactions, (ii) selecting genotyped bulls within the organic environment for artificial insemination (AI) programs and (iii) selecting genotyped natural service bulls within organic herds. The simulated conventional population comprised 148 800 cows from 2976 herds with an average herd size of 50 cows per herd, and 1200 cows were assigned to 60 organic herds. In a young bull program, selection criteria of young bulls in both production systems (conventional and organic) were either 'conventional' estimated breeding values (EBV) or genomic estimated breeding values (GEBV) for two traits with low (h 2=0.05) and moderate heritability (h 2=0.30). GEBV were calculated for different accuracies (r mg), and G×E interactions were considered by modifying originally simulated true breeding values in the range from r g=0.5 to 1.0. For both traits (h 2=0.05 and 0.30) and r mg⩾0.8, genomic selection of bulls directly in the organic population and using selected bulls via AI revealed higher genetic gain than selecting young bulls in the larger conventional population based on EBV; also without the existence of G×E interactions. Only for pronounced G×E interactions (r g=0.5), and for highly accurate GEBV for natural service bulls (r mg>0.9), results suggests the use of genotyped organic natural service bulls instead of implementing an AI program. Inbreeding coefficients of selected bulls and their offspring were generally lower when basing selection decisions for young bulls on GEBV compared with selection strategies based on pedigree indices.

Incorporating heifer feed efficiency in the Australian selection index using genomic selection

The economic benefit of expanding the Australian Profit Ranking (APR) index to include residual feed intake (RFI) was evaluated using a multitrait selection index. This required the estimation of genetic parameters for RFI and genetic correlations using single nucleotide polymorphism data (genomic) correlations with other traits. Heritabilities of RFI, dry matter intake (DMI), and all the traits in the APR (milk, fat, and protein yields; somatic cell count; fertility; survival; milking speed; and temperament), and genomic correlations between these traits were estimated using a Bayesian framework, using data from 843 growing Holstein heifers with phenotypes for DMI and RFI, and bulls with records for the other traits. Heritability estimates of DMI and RFI were 0.44 and 0.33, respectively, and the genomic correlation between them was 0.03 and nonsignificant. The genomic correlations between the feed-efficiency traits and milk yield traits were also close to zero, ranging between -0.11 and 0.10. Positive genomic correlations were found for DMI with stature (0.16) and with overall type (0.14), suggesting that taller cows eat more as heifers. One issue was that the genomic correlation estimates for RFI with calving interval (ClvI) and with body condition score were both unfavorable (-0.13 and 0.71 respectively), suggesting an antagonism between feed efficiency and fertility. However, because of the relatively small numbers of animals in this study, a large 95% probability interval existed for the genomic correlation between RFI and ClvI (-0.66, 0.36). Given these parameters, and a genetic correlation between heifer and lactating cow RFI of 0.67, inclusion of RFI in the APR index would reduce RFI by 1.76 kg/cow per year. Including RFI in the APR would result in the national Australian Holstein herd consuming 1.73 × 10(6) kg less feed, which is worth 0.55 million Australian dollars (A$) per year and is 3% greater than is currently possible to achieve. Other traits contributing to profitability, such as milk production and fertility, will also improve through selection on this index; for example, ClvI would be reduced by 0.53 d/cow per year, which is 96% of the gain for this trait that is achieved without RFI in the APR.

Comparison of heritabilities of dairy traits in Australian Holstein-Friesian cattle from genomic and pedigree data and implications for genomic evaluations

The reliability of genomic evaluations depends on the proportion of genetic variation explained by the DNA markers. In this study, we have estimated the proportion of variance in daughter trait deviations (DTDs) of dairy bulls explained by 45 993 genome wide single-nucleotide polymorphism (SNP) markers for 29 traits in Australian Holstein-Friesian dairy cattle. We compare these proportions to the proportion of variance in DTDs explained by the additive relationship matrix derived from the pedigree, as well as the sum of variance explained by both pedigree and marker information when these were fitted simultaneously. The proportion of genetic variance in DTDs relative to the total genetic variance (the total genetic variance explained by the genomic relationships and pedigree relationships when both were fitted simultaneously) varied from 32% for fertility to approximately 80% for milk yield traits. When fitting genomic and pedigree relationships simultaneously, the variance unexplained (i.e. the residual variance) in DTDs of the total variance for most traits was reduced compared to fitting either individually, suggesting that there is not complete overlap between the effects. The proportion of genetic variance accounted by the genomic relationships can be used to modify the blending equations used to calculate genomic estimated breeding value (GEBV) from direct genomic breeding value (DGV) and parent average. Our results, from a validation population of young dairy bulls with DTD, suggest that this modification can improve the reliability of GEBV by up to 5%.

Potential benefits of genomic selection on genetic gain of small ruminant breeding programs

In conventional small ruminant breeding programs, only pedigree and phenotype records are used to make selection decisions but prospects of including genomic information are now under consideration. The objective of this study was to assess the potential benefits of genomic selection on the genetic gain in French sheep and goat breeding designs of today. Traditional and genomic scenarios were modeled with deterministic methods for 3 breeding programs. The models included decisional variables related to male selection candidates, progeny testing capacity, and economic weights that were optimized to maximize annual genetic gain (AGG) of i) a meat sheep breeding program that improved a meat trait of heritability (h(2)) = 0.30 and a maternal trait of h(2) = 0.09 and ii) dairy sheep and goat breeding programs that improved a milk trait of h(2) = 0.30. Values of ±0.20 of genetic correlation between meat and maternal traits were considered to study their effects on AGG. The Bulmer effect was accounted for and the results presented here are the averages of AGG after 10 generations of selection. Results showed that current traditional breeding programs provide an AGG of 0.095 genetic standard deviation (σa) for meat and 0.061 σa for maternal trait in meat breed and 0.147 σa and 0.120 σa in sheep and goat dairy breeds, respectively. By optimizing decisional variables, the AGG with traditional selection methods increased to 0.139 σa for meat and 0.096 σa for maternal traits in meat breeding programs and to 0.174 σa and 0.183 σa in dairy sheep and goat breeding programs, respectively. With a medium-sized reference population (nref) of 2,000 individuals, the best genomic scenarios gave an AGG that was 17.9% greater than with traditional selection methods with optimized values of decisional variables for combined meat and maternal traits in meat sheep, 51.7% in dairy sheep, and 26.2% in dairy goats. The superiority of genomic schemes increased with the size of the reference population and genomic selection gave the best results when nref > 1,000 individuals for dairy breeds and nref > 2,000 individuals for meat breed. Genetic correlation between meat and maternal traits had a large impact on the genetic gain of both traits. Changes in AGG due to correlation were greatest for low heritable maternal traits. As a general rule, AGG was increased both by optimizing selection designs and including genomic information.

Genotyping-by-sequencing (GBS): a novel, efficient and cost-effective genotyping method for cattle using next-generation sequencing

High-throughput genotyping methods have increased the analytical power to study complex traits but high cost has remained a barrier for large scale use in animal improvement. We have adapted genotyping-by-sequencing (GBS) used in plants for genotyping 47 animals representing 7 taurine and indicine breeds of cattle from the US and Africa. Genomic DNA was digested with different enzymes, ligated to adapters containing one of 48 unique bar codes and sequenced by the Illumina HiSeq 2000. PstI was the best enzyme producing 1.4 million unique reads per animal and initially identifying a total of 63,697 SNPs. After removal of SNPs with call rates of less than 70%, 51,414 SNPs were detected throughout all autosomes with an average distance of 48.1 kb, and 1,143 SNPs on the X chromosome at an average distance of 130.3 kb, as well as 191 on unmapped contigs. If we consider only the SNPs with call rates of 90% and over, we identified 39,751 on autosomes, 850 on the X chromosome and 124 on unmapped contigs. Of these SNPs, 28,843 were not tightly linked to other SNPs. Average marker density per autosome was highly correlated with chromosome size (coefficient of correlation  = −0.798, r²  = 0.637) with higher density in smaller chromosomes. Average SNP call rate was 86.5% for all loci, with 53.0% of the loci having call rates >90% and the average minor allele frequency being 0.212. Average observed heterozygosity ranged from 0.046–0.294 among individuals, and from 0.064–0.197 among breeds, with Brangus showing the highest diversity as expected. GBS technique is novel, flexible, sufficiently high-throughput, and capable of providing acceptable marker density for genomic selection or genome-wide association studies at roughly one third of the cost of currently available genotyping technologies.

Accelerating improvement of livestock with genomic selection

Three recent breakthroughs have resulted in the current widespread use of DNA information: the genomic selection (GS) methodology, which is a form of marker-assisted selection on a genome-wide scale, and the discovery of large numbers of single-nucleotide markers and cost effective methods to genotype them. GS estimates the effect of thousands of DNA markers simultaneously. Nonlinear estimation methods yield higher accuracy, especially for traits with major genes. The marker effects are estimated in a genotyped and phenotyped training population and are used for the estimation of breeding values of selection candidates by combining their genotypes with the estimated marker effects. The benefits of GS are greatest when selection is for traits that are not themselves recorded on the selection candidates before they can be selected. In the future, genome sequence data may replace SNP genotypes as markers. This could increase GS accuracy because the causative mutations should be included in the data.

Hot topic: Effect of breeding strategies using genomic information on fitness and health

A complex deterministic approach was used to model the breeding goal and breeding structure for the Austrian Fleckvieh (dual-purpose Simmental) breed. The reference breeding goal corresponded to the current total merit index (TMI-R), where dairy traits have a relative weight of 37.9% and fitness traits of 43.7% (beef traits 16.5%; milkability 2%). The breeding program was characterized by 280,000 cows under performance recording, 3,200 bull dams, 100 test bulls with a test capacity of 25%, and 15 proven bulls and 8 bull sires per year. The annual monetary genetic gain (AMGG) was generated mainly by increases in milk fat and milk protein yield (80.6%) and only to a small extent by fitness traits (6.6%). The inclusion of direct health traits (early reproductive disorders, cystic ovaries, and mastitis) with their economic weights increased the relative AMGG for fitness traits from 6.6 to 11.2%. The presently slightly negative AMGG for fertility index and udder health changed in a positive direction. Increasing the weight on the direct health traits by 50% resulted in a further shift toward fitness and health. The effect of strategies using genomic information in a total merit index (TMI) with varying weights on fitness and health traits was also analyzed. The conventional progeny-testing scheme was defined as the reference breeding program. A breeding program was considered to be genomically enhanced (GS50) when 50% of inseminations of herdbook cows and of bull dams were from young bulls with a genomic TMI, and a second program (GS100) did not rely on progeny-tested bulls at all. For GS50, a clear shift of the relative gain in AMGG toward fitness and health traits was observed for all 3 TMI scenarios, as a result of larger progeny groups and a shorter generation interval. For GS100, where no gene flow from progeny-tested bulls was assumed, the genetic gain per generation was lower for the fertility and udder health index but higher per year. The results based on natural genetic gain per year showed that no positive genetic response for fertility and udder health index were achieved for TMI-R (without the inclusion of direct health traits) in GS50 and GS100. The direction of the genetic trend was determined by the weights given to fertility and udder health indices within the TMI. When appropriate weights generated a clear positive trend, GS50 and GS100 reinforced this trend.

Strategies and utility of imputed SNP genotypes for genomic analysis in dairy cattle

Background

We investigated strategies and factors affecting accuracy of imputing genotypes from lower-density SNP panels (Illumina 3K, 7K, Affymetrix 15K and 25K, and evenly spaced subsets) up to one medium (Illumina 50K) and one high-density (Illumina 800K) SNP panel. We also evaluated the utility of imputed genotypes on the accuracy of genomic selection using Australian Holstein-Friesian cattle data from 2727 and 845 animals genotyped with 50K and 800K SNP chip, respectively. Animals were divided into reference and test sets (genotyped with higher and lower density SNP panels, respectively) for evaluating the accuracies of imputation. For the accuracy of genomic selection, a comparison of direct genetic values (DGV) was made by dividing the data into training and validation sets under a range of imputation scenarios.

Results

Of the three methods compared for imputation, IMPUTE2 outperformed Beagle and fastPhase for almost all scenarios. Higher SNP densities in the test animals, larger reference sets and higher relatedness between test and reference animals increased the accuracy of imputation. 50K specific genotypes were imputed with moderate allelic error rates from 15K (2.85%) and 25K (2.75%) genotypes. Using IMPUTE2, SNP genotypes up to 800K were imputed with low allelic error rate (0.79% genome-wide) from 50K genotypes, and with moderate error rate from 3K (4.78%) and 7K (2.00%) genotypes. The error rate of imputing up to 800K from 3K or 7K was further reduced when an additional middle tier of 50K genotypes was incorporated in a 3-tiered framework. Accuracies of DGV for five production traits using imputed 50K genotypes were close to those obtained with the actual 50K genotypes and higher compared to using 3K or 7K genotypes. The loss in accuracy of DGV was small when most of the training animals also had imputed (50K) genotypes. Additional gains in DGV accuracies were small when SNP densities increased from 50K to imputed 800K.

Conclusion

Population-based genotype imputation can be used to predict and combine genotypes from different low, medium and high-density SNP chips with a high level of accuracy. Imputing genotypes from low-density SNP panels to at least 50K SNP density increases the accuracy of genomic selection.

Predicted accuracy of and response to genomic selection for new traits in dairy cattle

Genomic selection relaxes the requirement of traditional selection tools to have phenotypic measurements on close relatives of all selection candidates. This opens up possibilities to select for traits that are difficult or expensive to measure. The objectives of this paper were to predict accuracy of and response to genomic selection for a new trait, considering that only a cow reference population of moderate size was available for the new trait, and that selection simultaneously targeted an index and this new trait. Accuracy for and response to selection were deterministically evaluated for three different breeding goals. Single trait selection for the new trait based only on a limited cow reference population of up to 10 000 cows, showed that maximum genetic responses of 0.20 and 0.28 genetic standard deviation (s.d.) per year can be achieved for traits with a heritability of 0.05 and 0.30, respectively. Adding information from the index based on a reference population of 5000 bulls, and assuming a genetic correlation of 0.5, increased genetic response for both heritability levels by up to 0.14 genetic s.d. per year. The scenario with simultaneous selection for the new trait and the index, yielded a substantially lower response for the new trait, especially when the genetic correlation with the index was negative. Despite the lower response for the index, whenever the new trait had considerable economic value, including the cow reference population considerably improved the genetic response for the new trait. For scenarios with a zero or negative genetic correlation with the index and equal economic value for the index and the new trait, a reference population of 2000 cows increased genetic response for the new trait with at least 0.10 and 0.20 genetic s.d. per year, for heritability levels of 0.05 and 0.30, respectively. We conclude that for new traits with a very small or positive genetic correlation with the index, and a high positive economic value, considerable genetic response can already be achieved based on a cow reference population with only 2000 records, even when the reliability of individual genomic breeding values is much lower than currently accepted in dairy cattle breeding programs. New traits may generally have a negative genetic correlation with the index and a small positive economic value. For such new traits, cow reference populations of at least 10 000 cows may be required to achieve acceptable levels of genetic response for the new trait and for the whole breeding goal.

Novel strategies to minimize progeny inbreeding while maximizing genetic gain using genomic information

In this study, 3 strategies for controlling progeny inbreeding in mating plans were compared. The strategies used information from pedigree inbreeding coefficients, genomic relationships, or shared runs of homozygosity. The strategies were compared for the reduction in genetic gain and progeny inbreeding that would be expected from selected matings, and for the decrease of homozygosity of deleterious recessive alleles. Using real pedigree, genotype [43,115 single nucleotide polymorphism (SNP) markers], and estimated breeding value data from Holstein cattle, mating plans were derived for herds of 300 cows with 20 sires available for mating, replicated 50 times. Each of the 300 individuals allocated as dams were matched to 1 of 20 sires to maximize genetic merit minus the penalty for estimated progeny inbreeding, and given the restriction that the sire could not be mated to more than 10% of the cows. The strategy that used a genomic relationship matrix (GRM) was the most effective in reducing average progeny inbreeding; this strategy also resulted in fewer homozygous SNP out of 1,000 low-frequency SNP compared with the strategy using pedigree information. In the future, large numbers of cattle may be genotyped for low-density SNP panels. A GRM constructed using 3,123 SNP produced results similar to a GRM constructed using the full 43,115 SNP. These results demonstrate that using GRM information, a 1% reduction in progeny inbreeding (valued at around $5 per cow) can be made with very little compromise in the overall breeding objective. These results and the availability of low-cost, low-density genotyping make it attractive to apply mating plans that use genomic information in commercial dairy herds.

Design of a bovine low-density SNP array optimized for imputation

The Illumina BovineLD BeadChip was designed to support imputation to higher density genotypes in dairy and beef breeds by including single-nucleotide polymorphisms (SNPs) that had a high minor allele frequency as well as uniform spacing across the genome except at the ends of the chromosome where densities were increased. The chip also includes SNPs on the Y chromosome and mitochondrial DNA loci that are useful for determining subspecies classification and certain paternal and maternal breed lineages. The total number of SNPs was 6,909. Accuracy of imputation to Illumina BovineSNP50 genotypes using the BovineLD chip was over 97% for most dairy and beef populations. The BovineLD imputations were about 3 percentage points more accurate than those from the Illumina GoldenGate Bovine3K BeadChip across multiple populations. The improvement was greatest when neither parent was genotyped. The minor allele frequencies were similar across taurine beef and dairy breeds as was the proportion of SNPs that were polymorphic. The new BovineLD chip should facilitate low-cost genomic selection in taurine beef and dairy cattle.

Genomic selection in French dairy cattle

Genomic selection is implemented in French Holstein, Montbéliarde, and Normande breeds (70%, 16% and 12% of French dairy cows). A characteristic of the model for genomic evaluation is the use of haplotypes instead of single-nucleotide polymorphisms (SNPs), so as to maximise linkage disequilibrium between markers and quantitative trait loci (QTLs). For each trait, a QTL-BLUP model (i.e. a best linear unbiased prediction model including QTL random effects) includes 300–700 trait-dependent chromosomal regions selected either by linkage disequilibrium and linkage analysis or by elastic net. This model requires an important effort to phase genotypes, detect QTLs, select SNPs, but was found to be the most efficient one among all tested ones. QTLs are defined within breed and many of them were found to be breed specific. Reference populations include 1800 and 1400 bulls in Montbéliarde and Normande breeds. In Holstein, the very large reference population of 18 300 bulls originates from the EuroGenomics consortium. Since 2008, ~65 000 animals have been genotyped for selection by Labogena with the 50k chip. Bulls genomic estimated breeding values (GEBVs) were made official in June 2009. In 2010, the market share of the young bulls reached 30% and is expected to increase rapidly. Advertising actions have been undertaken to recommend a time-restricted use of young bulls with a limited number of doses. In January 2011, genomic selection was opened to all farmers for females. Current developments focus on the extension of the method to a multi-breed context, to use all reference populations simultaneously in genomic evaluation.

Designing dairy cattle breeding schemes under genomic selection: a review of international research

High rates of genetic gain can be achieved through (1) accurate predictions of breeding values (2) high intensities of selection and (3) shorter generation intervals. Reliabilities of ~60% are currently achievable using genomic selection in dairy cattle. This breakthrough means that selection of animals can happen at a very early age (i.e. as soon as a DNA sample is available) and has opened opportunities to radically redesign breeding schemes. Most research over the past decade has focussed on the feasibility of genomic selection, especially how to increase the accuracy of genomic breeding values. More recently, how to apply genomic technology to breeding schemes has generated a lot of interest. Some of this research remains the intellectual property of breeding companies, but there are examples in the public domain. Here we review published research into breeding scheme design using genomic selection and evaluate which designs appear to be promising (in terms of rates of genetic gain) and those that may have unfavourable side-effects (i.e. increasing the rate of inbreeding). The schemes range from fairly conservative designs where bulls are screened genomically to reduce numbers entering progeny testing, to schemes where very large numbers of bull calves are screened and used as sires as soon as they reach sexual maturity. More radical schemes that incorporate the use of reproductive technologies (in juveniles) and genomic selection in nucleus herds are also described. The models used are either deterministic and more recently tend to be stochastic, simulating populations of cattle. A key driver of the rate of genetic gain is the generation interval, which could range from being similar to that in conventional testing (~5 years), down to as little as 1.5 years. Generally, the rate of genetic gain is between 12% and 100% more than in conventional progeny testing, while the rate of inbreeding tends to be lower per generation than in progeny testing because Mendelian sampling terms can be estimated more accurately. However, short generation intervals can lead to higher rates of inbreeding per year in genomic breeding programs.

Use of female information in dairy cattle genomic breeding programs

Genomic selection has the potential to increase the accuracy of selection and, therefore, genetic gain, as well as reducing the rate of inbreeding, yet few studies have evaluated the potential benefit of the contribution of females in genomic selection programs. The objective of this study was to determine the effect on genetic gain, accuracy of selection, generation interval, and inbreeding, of including female genotypes in a genomic selection breeding program. A population of approximately 3,500 females and 500 males born annually was simulated and split into an elite and commercial tier representation of the Irish national herd. Several alternative breeding schemes were evaluated to quantify the potential benefit of female genomic information within dairy breeding schemes. Results showed that the inclusion of female phenotypic and genomic information can lead to a 3-fold increase in the rate of genetic gain compared with a traditional BLUP breeding program and decrease the generation interval of the males by 3.8 yr, while maintaining a reasonable rate of inbreeding. The accuracy of the selected males was increased by 73% in the final 3 yr of the genomic schemes compared with the traditional BLUP scheme. The results of this study have several implications for national breeding schemes. Although an investment in genotyping a large population of animals is required, these costs can be offset by the greater genetic gain achievable through the increased accuracy of selection and decreased generation intervals associated with genomic selection.

Genomic selection in livestock populations

Most traits of economic importance in livestock are either quantitative or complex. Despite considerable efforts, there has been only limited success in identifying the polymorphisms that cause variation in these traits. Nevertheless, selection based on estimated breeding values (BVs), calculated from data on phenotypic performance and pedigree has been very successful. Genomic tools, such as single nucleotide polymorphism (SNP) chips, have led to a new method of selection called 'genomic selection' in which dense SNP genotypes covering the genome are used to predict the BV. In this review we consider the statistical methodology for estimating BVs from SNP data, factors affecting the accuracy, the long-term response to genomic selection and the design of breeding programmes including the management of inbreeding.