genomic BLUP (ssGBLUP) for the estimation of genetic values (Amaya et al 2020)
| Livestock Research for Rural Development 34 (12) 2022 | LRRD Search | LRRD Misssion | Guide for preparation of papers | LRRD Newsletter | Citation of this paper |
Genomic selection changed the dairy cattle breeding and selection schemes. Obtaining genetic values at early ages reduce the generation interval and a significant increase in the annual genetic gain rate. Cuban dairy cattle genetic selection and improvement programs are aimed at increasing the productive potential of native breeds resistant to tropical environment and are based on genetic evaluations considering phenotypic, productive and pedigree records. The objective of this work was to expose the challenges and opportunities of Cuban cattle breeding to implement a dairy cattle genomic selection program. Having a well-established conventional genetic evaluation system, data collection trained personnel and appropriate collaboration between research centers, livestock and insemination companies are strengths to successfully implement genomic selection in Cuba. Genomic studies would contribute to reduce costs and periods to identify and select outstanding individuals in bovine populations, increase the annual genetic gain rate, and allow a better genetic characterization of current dairy herds. The selection of animals to be genotyped, the creation of genotyping laboratories and the use of economically accessible matrices are some of the main challenges to be overcome. Implementing genomic methodology will enhance country's strategic plans to achieve food safety in the production of milk and dairy products.
Key words: dairy cattle, developing countries, genomic selection, genotyping
Animal breeding is the science, art and business of improving organisms for the benefit of humans. As a science, breeding is based on theoretical and empirical knowledge of genetics. As an art, it requires subjective judgments in the design and implementation of a breeding program. Finally, as a business, it requires investments of time and money in different resources such as: technicians, equipment and materials (Paccapelo, 2015). Meuwissen et al (2001) were the first to introduce the techniques known as genomic selection (GS) by prediction of total genetic value using genome-wide dense marker maps in order to estimate the effects of more than 50,000 marker haplotypes simultaneously from a limited number of phenotypic records. Their work proposed the incorporation of numerous molecular markers into the statistical models used to estimate the genetic value of an individual. The authors implemented a simple but powerful idea: expressing phenotypes over all available markers using a linear model. The introduction of genomic selection substantially changed the breeding and selection schemes for dairy cattle. In addition, obtaining genetic values at very early ages allowed more immediate selection decisions to be made, rather than the extended time required by progeny testing, thereby reducing the generation interval and thus leading to a significant increase in the annual rate of genetic gain (Mrode et al 2019, Misztal et al 2020).
Developed countries where, fundamentally, genomic selection programs have been implemented in dairy cattle, whose success is not only due to the availability of economic resources, but also because they have strong genetic evaluation systems, well-developed breeding structures and breeding companies that contribute to these programs (Schöpke and Swalve, 2016; Mrode et al 2019). In developing countries, the situation is very different and it is characterized by the lack of pedigree and performance records, as well as the non-existence of conventional genetic evaluation systems. In some countries such as Brazil, the existence of breed associations has allowed the establishment of pedigree data records and genetic evaluation systems (Boison et al 2017), but there is still a lack of breeding structures such as artificial insemination (AI) companies, to advance breed improvement programs (Mrode et al 2019).
In Cuba, genetic selection and improvement programs in dairy cattle have been aimed at increasing the dairy production potential of the cattle existing in the country and, in parallel, developing an adequate exploitation for tropical conditions. For this purpose, groups of animals identified as native breeds resistant to the hot and humid tropical environment, such as Criollo de Cuba (Bos taurus) and Zebu Cubano (Bos indicus), have been maintained. In addition, crossbreeding has combined the hardiness of these resistant cattle with the productive qualities of imported breeds such as the Holstein.
In Cuba, breeding and genetic improvement programs are based on genetic evaluations that consider phenotypic, productive and reproductive records, pedigree and progeny testing. In addition, with the use of the Best Linear Unbiased Predictor (BLUP) methodology (Henderson, 1975), Estimated breeding values (EBVs) are obtained and subsequently used to select the animals with the highest genetic merit within the different breeds that are part of the Cuban livestock (Uffo et al 2012; Hernández et al 2020). It was not until the beginning of this century that the first studies using the Marker Assisted Selection (MAS) approachment were carried out (Uffo and Martínez, 2002; Uffo, 2003). Although genetic selection programs based on the genomic selection have been applied worldwide for more than a decade with overwhelming results in terms of genetic gain and increases in milk production and quality; in Cuba, the first approaches to this topic were carried out in 2018 (FOAR, 2018). The objective of this work was to expose the challenges and opportunities of the Cuban livestock for the implementation of a genomic selection program in dairy cattle.
In cattle, most economically important traits are controlled by many genes, each with a small effect. Therefore, a large amount of data is needed to estimate effects accurately, as well as markers densely distributed in the genome to ensure that the association between marker-QTL linkage persists across families (Cole et al 2009).
The concept of genomic selection proposed by Meuwissen et al (2001) is based on the homogeneous distribution of thousands of single nucleotide polymorphisms (SNPs) across the genome and the estimation of their effects on quantitative traits. With tens of thousands of SNPs, well chosen to be representative of the entire genome, it is expected that there will always be one SNP near a particular gene or quantitative trait locus (QTL) of interest. Linkage disequilibrium between one or more SNPs and a causal mutation will be substantial and can then be used to explain a significant fraction of the trait variation observed. Subsequently, obtaining complete genome sequences and the availability of SNP arrays of various densities (10,000 to 1,000,000 SNPs) have facilitated the implementation of genomic evaluation models that estimate genetic effects for chromosome segments in a population with phenotypic and genomic information, allowing to know the effects of each SNP and to estimate direct genomic values (Misztal et al 2009; Amaya et al 2020; Gutiérrez-Reinoso-Reinoso et al 2021).
Simultaneously, the increase in genomic data and changes in selection programs imply a constant updating of genetic evaluation systems. Genomic selection models have included the extended BLUP, called GBLUP (Genomic Best Linear Unbiased Prediction), Bayesian and nonparametric methods, which increase the accuracy of estimates and reduce the generation interval, thus contributing to further genetic progress. For these reasons, most of these approaches have generally shown superiority over evaluation methods based on the use of the pedigree parentage matrix that do not include genomic information (Legarra and Ducrocq 2012; Amaya et al 2020).
Early genomic evaluations contemplated a multi-step methodology (msGBLUP: multi-step genomic BLUP). However, genetic values could not be estimated for animals without genomic information in msGBLUP and generated high variation in reliabilities (VanRaden, 2008). Although this methodology the precision of the estimates for the selection of young animals, this genomic prediction also did not consider the preselection effect, underestimating genetic values for young animals (Amaya et al 2020). Therefore, Misztal et al (2009) developed the single step best linear unbiased predictor methodology (ssGBLUP: single step genomic BLUP) that incorporates genomic information and estimates genetic values for both genotyped and non-genotyped animals and increases the accuracy in genomic estimation (Amaya et al 2020). Figure 1 shows the working structure with these two methodologies.
|
| Figure 1.
Working methodology with the multi-step genomic BLUP (msGBLUP) and a single-
step genomic BLUP (ssGBLUP) for the estimation of genetic values (Amaya et al 2020) |
The first studies carried out in Cuba, based on MAS in cattle, were focused on several objectives, such as the analysis of the biodiversity of Cuban Creole cattle (Uffo, 2003), the use of microsatellites for the identification of paternities (Sanz et al 2002), the identification by polymerase chain reaction (PCR) of genes associated with milk proteins (Uffo and Martínez, 2002), studies of genetic polymorphism of milk proteins in native Cuban breeds (Uffo et al 2006), as well as the molecular diagnosis of diseases affecting cattle (Uffo and Acosta, 2009). In the last decade, through cooperation with European institutions and from Cuba, research was conducted aimed at the analysis of the genetic diversity of Cuban native breeds and their comparison with other Ibero-American bovine breeds by microsatellites and other molecular markers, which contributed to a better characterization of Cuban genofunds (Delgado et al 2012; Martínez et al 2012, Ginja et al 2013). Likewise, the maternal origin of Cuban Creole cattle was analyzed and contemporary paternal gene flow events were detected through a 240 pb fragment of the mitochondrial D-loop (mtDNA) and five microsatellites of the Y chromosome (BTY), in 36 females and 21 males respectively, concluding that the Cuban Creole cattle population has a haplotypical mitochondrial composition comparable to that of other Creole and Mediterranean breeds, a fact that agrees with its historical origin. The BTY evidenced high levels of paternal introgression of Zebu genes (Lirón et al 2011). Cuban Creole cattle were studied with the use of different molecular markers, which allowed identifying a high genetic variability within this native Cuban population (Uffo et al 2012). Other studies employed restriction enzyme digestion assays, sequencing and mini-sequencing to determine SNPs, which contributed to the simultaneous genotyping of genes coding for the six main milk proteins, growth hormone and bovine prolactin (Acosta et al 2011). In other research, using the microsatellite genotyping approach, studies were carried out to determine the genetic diversity of five native Cuban cattle breeds, results showing that a significant amount of genetic variation was maintained in local bovine populations and that all the breeds studied could be considered distinct genetic entities (Acosta et al 2012). Likewise, in other populations of livestock species of high economic interest such as water buffalo, genetic variability analyses were also carried out using microsatellites (Acosta et al 2014).
In general, these studies on the characterization of biodiversity and productive traits in Cuban cattle populations constitute the starting point and the basis for the design and development of genetic selection programs assisted by molecular markers that contribute to the productive improvement and conservation of Cuban native breeds. However, these programs continued to rely only on the results of progeny and behavioral tests, and on some methods for estimating the genetic value through the animal model BLUP, without taking into account the benefits and effectiveness of molecular or bioinformatics tools (Uffo et al 2012; Hernández et al 2015 and 2020). Although none or very few of these results were subsequently applied in the genetic selection programs of these breeds, they did favor the updating and approval in recent years of policies that promote and support the use, not only of MAS, but also of research projects that contribute to the design and implementation of future genomic selection programs in some of these Cuban bovine populations. This is the case of the tripartite FOAR-AECID-Cuba project: "Improvement of information and management tools for updating dairy policy in Cuba", based specifically on Siboney de Cuba dairy breed, focused on defining genomic selection indicators for this population, training Cuban technical staff and farmers in genomic genotyping and bioinformatics methodology, genetic information management and genomic selection statistical models, and finally establishing a genomic selection program for dairy cattle in Cuba (FOAR 2018). The development of this research made possible to identify the productive and sanitary indicators to be considered in the selection program, mainly associated to mastitis resistance and milk quality, and to train specialists, technicians and producers in microarray technology for cattle genotyping as a tool for genomic selection. Through the analysis of Siboney de Cuba databases (CIMAGT-MINAG), it was possible to identify the animals of the last generations that would provide information for the estimation of genomic relationships. Unfortunately, the still high costs of microarray analysis have so far not allowed DNA analysis of the selected candidates.
The establishment of an appropriate reference population is one of the key aspects on genomic selection. For the conformation of cattle reference populations, several authors recommend the following (Mrode et al 2019, Amaya et al 2020):
- The size of the reference population should be high (between 500 and 4000 individuals) and inversely proportional to the heritability of the trait. Low heritability traits (e.g., fertility) need a larger reference population to obtain the same reliability as high heritability traits (e.g., protein content).
- The size of a genotyped population should be larger when the candidate population for selection has a lower genetic relatedness to the genotyped population, and the higher the relatedness of the reference population to the population where the prediction equations will be applied, the higher the accuracy of the genomic predictions.
- Use proven progeny from recent generation bulls instead of older bulls.
- The animals included should be closely related and from the most representative progenies of the population for which direct genetic values are to be predicted.
- Animals with higher and lower genetic merit should be included to increase the reliability of the predictions.
- Complete and detailed information on their productive and reproductive characteristics and those of their offspring must be available, which makes yield control even more important.
According to the 2014 Country Report on Animal Genetic Resources (FAO, 2014), the composition of Cuban dairy breeds was as follows (Table 1):
|
Table 1. composition of Cuban dairy breeds |
||
|
Breeds |
Females |
Males |
|
Tropical Holstein |
3897 |
37 |
|
Siboney de Cuba |
15591 |
2299 |
|
Mambí de Cuba |
381 |
2 |
|
Taíno de Cuba |
4128 |
6 |
|
Source: FAO, 2014 |
||
If the multi-step genomic BLUP (msGBLUP) methodology were used to perform genomic evaluations, it would be very difficult to design a reference population from those herds with the exception, perhaps, Siboney de Cuba, which is the most represented and also one of the breeds best adapted to the tropical climatic conditions and with adequate genealogical, reproductive and productive records, with which genetic estimations are currently performed (Hernández et al 2015 and 2020). However, it has not been possible to carry out studies with the rest of the Cuban dairy breeds due to the small size of their populations. As mentioned, the alternative to the use of these huge reference populations for genotyping, whose main challenge is economic limitations, is the application of the ssGBLUP methodology. Those procedure first described by Misztal et al (2009) combines phenotypes, pedigree and genotypes in a single evaluation and consists of combining the pedigree relationship matrix obtained in the traditional BLUP with a genomic relationship matrix. It also generates a hybrid matrix, combining pedigree and genome relationships, which allows predicting the best values for both genotyped and non-genotyped individuals (Amaya et al 2020).
This increases the accuracy and reduces the prediction bias of the genomic estimated breeding values (GEBVs) compared to those obtained from multi-step genomic predictions (Lourenco et al 2020, Amaya et al 2020). This approach is very useful for populations that cannot support the economic effort of extensive genotyping, such as populations of the Siboney de Cuba breeds, as well as for other native breeds with relatively small herds, but of economic and genetic interest such as the Mambí de Cuba and the Taíno de Cuba.
The ssGBLUP does not use a "reference population", but it does require genotyped animals; therefore, the size of this genotyped population will depend primarily on available funding and the objectives of the genomic study. Lourenco et al (2015) suggested that the genotyping strategy in ssGBLUP should target the animals of greatest importance, which are generally the oldest and with the most genealogical and phenotypic information. Consequently, the composition and size of the genotyped population are factors that affect the variance and covariance structure of the genetic values, and thus accuracy. Although populations genotyped for ssGBLUP are sometimes small, it is important to keep in mind that larger genotyped populations will contribute more and accurately to the estimation of allelic effects. There is now greater availability of animals with known genotypes for a greater number of SNPs, due to lower genotyping costs and access to more efficient statistical and computational methods used in the imputation process (Amaya et al 2020). The inclusion of elite females is another feature to be considered in future populations to be genotyped in Cuban dairy herds. Initially, reference populations included only proven bulls with genetic values that were estimated based on the phenotypes of their daughters. More recently, for studies in smaller populations such as those in Cuba, females have begun to be incorporated into these genotyping studies. Thus, some of these populations are a combination of bulls and cows (Boison et al 2017), and in some cases the majority are cows (Brown et al 2016; Silva et al 2016).
This has implications for genomic prediction accuracy, which has tended to be lower than that obtained in developed countries, given the limited response variable information when using cow records. However, the inclusion of cows has, in some cases, resulted in up to a fivefold increase in genotyped population size and increases of up to 12 % in accuracy compared to using only bulls (Mrode et al 2019). According to Amaya et al (2020), the inclusion of genotyped females is desirable because they are an important part of breeding programs as they are equally subjected to a selection process and provide phenotypic values through their own performance and progeny. In addition, they allow increasing the size of the genotyped population, which makes them an alternative to reduce biases and increase precision.
Aguilar et al (2010) performed the first genetic evaluation with ssGBLUP in Holstein cattle in the United States, obtaining higher accuracies. Since then, ssGBLUP has become a simpler and more accurate method for estimating genetic values. However, its implementation requires genotyping a part of the population, considering computational requirements and evaluating factors that could influence the precision of the estimates, such as the number of animals genotyped and their relationship with the population evaluated. In the last decade, ssGBLUP has become the tool of choice for genomic evaluation and selection in many livestock species, namely cattle, pigs, broilers, layers, dairy sheep and goats, meat sheep, and fish. Although ssGBLUP adds simplicity to the genomic evaluation system, its implementation involves several details and requires knowledge of the peculiarities of the method (Amaya et al 2020).
Genomic selection does not consider at any time the complete replacement of yield and pedigree controls, thus collection and disposition of phenotypes and productive data will be essential (Aranguren and Portillo, 2017).
In Cuba, there is a well-established conventional genetic evaluation system and procedures for collecting information that make up the extensive and robust databases with which genetic estimates are made. Currently in Cuban dairy breeds (Mambí de Cuba, Siboney de Cuba, Taíno de Cuba and Holstein); the data collected are related to pedigree or genealogical information, milk yields accumulated at 100, 244 and 305 days of lactation, duration of lactation, milk production per day and some reproductive traits such as age at first calving, gestation interval, calving interval, number of births per lifetime, among others (Hernández et al 2015 and 2020).
It is evident that, for the implementation of future genomic selection programs that allow high precision in the estimation of genomic values for breeding, it will be essential to start collecting other phenotypic data related to milk quality (fat, total solids, proteins, lactose) and animal health (disease diagnosis, somatic cell count, among others). Other authors also suggest the need to incorporate other types of data such as birth weight, days of lactation, udder size, among others, that could be associated with genomic data and that will undoubtedly lead to an increase in the precision of estimates of traits and indexes of interest for dairy cattle (Schöpke and Swalve, 2016; Gutiérrez-Reinoso et al 2021). Therefore, there is a need for laboratory capabilities and qualified personnel to obtain data on milk quality and other traits directly or indirectly associated with animal health.
Genotyping is one of the main obstacles in the implementation of genomic selection programs, due to the cost involved. However, Misztal et al (2020), refer that as of November 2019; more than three million Holsteins in the United States (https://queries.uscdcb.com/Genotype/cur_freq.html) and more than 700,000 American Angus had been genotyped.
To obtain the genotypes, in the case of Cuba, it will be necessary to use and/or design SNP matrices appropriate for Bos Taurus-Bos indicus dairy crossbred populations such as the Siboney de Cuba, Mambí de Cuba and Taíno de Cuba. That is to say, these genotyping matrices should include SNPs associated with productive, reproductive or health traits, defined in the genomic selection program for these populations. Several authors report that genotyping of dairy cattle can usually be performed with low and medium density SNP arrays, which are cheaper, such as: BovineBeadchip 50K, GeneSeek super genomicprofiler (SGGP-20Ki) and GeneSeekgenomicprofiler (GGP-75Ki), and with which adequate accuracy in genomic predictions is also achieved (Boison et al 2017; Aranguren and Portillo 2017; Butty et al 2019). However, since the beginning of the last decade, the design of high density arrays such as BeadChipBovineHD (http://investor.illumina.com/), which provides 500,000 loci representing 10 times more resolution than BovineBeadchip 50K, was started (Aranguren and Portillo 2017). As can be inferred, the main advantage of increasing SNP density is that the linkage disequilibrium (LD) between SNPs flanking markers and the QTLs of interest is increased, so that higher levels of LD provide better signaling of QTLs within and across families (Harris and Johnson, 2010). Other alternatives in the design of genotyping arrays is to select SNPs that best predict some traits and include them in smaller arrays, which would allow pre-selections to be made. Currently, Illumina is developing 384 SNP arrays, which includes 100 SNPs to be able to perform parentage verifications in all cattle breeds, while the rest would allow predicting direct genomic value; however, reliability gain is still under discussion (Aranguren and Portillo, 2017).
Another strategy to cost-effectively increase the genotyped population economically has been the use of imputed genotypes from arrays of various densities combined in a single genomic evaluation to reduce costs and increase accuracy. For example, the number of animals genotyped in the United States from the Holstein breed increased significantly to over 950,000 animals in 2016 with this procedure (Misztal, 2016).
On the other hand, array design for Bos taurus- Bos indicus crossbred populations has been referenced by several authors and have been effectively used to determine the genetic composition of herds and to make genomic predictions associated with milk production and quality (Mrode et al 2019; Hidalgo et al 2016; Ducrocq et al 2018).
Other authors highlight that currently and in the immediate future, genomic evaluations should consider inserting additional genes to genotyping arrays that allow characterizing new complex and low heritability traits associated with health, feed efficiency, metabolism, immunity, methane emissions, among others (Schöpke and Swalve, 2016; Gutiérrez-Reinoso et al 2021).
One of the most important aspects to consider is that the SNP genotyping process and the calculation of gEBV will initially have to be performed outside Cuba, which will increase the cost of the genotyping process, which, although it has decreased in recent years, it is still considered a major limitation for developing countries.
Several authors have pointed out the importance of international collaboration in the application of genomic selection (Schöpke and Swalve, 2016; Mrode et al 2019). The formation of the first reference populations for Holstein dairy cattle was achieved by sharing herd genotypes from the United States, Canada, Italy and England, which enhanced the size of the reference population and led to to increased accuracy of genomic predictions (Schöpke and Swalve, 2016). In the case of developing countries, and having crossbred dairy populations, where there has been a large importation of bulls, mostly genotyped in developed countries, the willingness to share genotypes and some other relevant performance data will help to expand the reference population and, therefore, the accuracy of genomic predictions in developing countries (Mrodre et al 2019). In this sense, the situation in Cuba is deficient, as the report on conservation of animal genetic resources refers in 2014, the import of bulls or semen from developed countries was practically null; in previous years only a few hundred Holstein, Jersey and Bradford had been imported (FAO, 2014).
Furthermore, over the last few years, the collaboration and generation of research projects with foreign institutions have been strengthened, e.g. FOAR-AECID-Cuba project, mentioned above, and the research carried out with the use of new genotyping methodologies such as complete genome sequencing and analysis by genotyping arrays for genetic variability studies in the Charolais de Cuba breeds. This has allowed the identification of genes and SNPs associated with the adaptation of this breed to the tropical conditions of Cuba, such as immunity, muscle development and meat quality (Ramírez et al 2021; Rodríguez et al 2018).
In general, the main strengths to achieve the establishment of genomic selection programs in Cuba are based on in the existence of a well-established conventional genetic evaluation system that guarantees an adequate control of pedigree and population databases and productive and reproductive traits, as well as the trained personnel to collect them. Likewise, the strong collaboration established between research centers, livestock and insemination companies, which will be in charge of implementing these programs, is a guarantee of success. However, challenges such as genotyping and computational infrastructure for genomic evaluation must be overcome, as well as the collection of phenotypic data associated with milk quality, disease resistance and other productive and reproductive data that are not currently taken into account, and which are essential to achieve adequate precision in genomic estimates.
The implementation of genomic selection includes a series of stages, which imply a merged work between livestock companies, government, ministries, and research and artificial insemination centers. Such implementation at the local level requires joining strengths and knowledge in research, development and application to the sector, by all the stakeholders involved (Aranguren and Portillo, 2017).
In Cuba, the design of bovine genetic improvement programs based on genomic selection must meet the priorities of the country's livestock development, which includes the enhancement of genetic development, the establishment of a comprehensive training system and the updating and implementation of programs aimed at the preservation and rehabilitation of natural resources, including, in particular, livestock species of economic interest, which are considered indispensable for production and world food security.
The implementation of genomic selection programs will involve a huge investment in the infrastructure needed for the initial setup and the ongoing process of genotyping, data management and estimation of genomic breeding values, all of which will undoubtedly influence the improvement of genetic selection programs.
Genomic selection is one of the most important scientific advances of this century. The application of genomic analysis in dairy cattle selection provides very interesting results such as short/medium term genetic predictions, controls the inbreeding coefficient, reduces the generation interval and improves sire selection, thus generating more productive and economical genetic lines in a shorter time.
In addition, analyses with genotyping arrays and associated technologies, such as Genome-Wide Association Study (GWAS), have also made possible to determine breed composition, paternity verification, genome diversity, as well as the frequency of rare alleles, recessive genes and new mutations that could generate genetic variability in economically interesting traits. Similarly, the determination of new traits associated with disease resistance, adaptability to the environment and others of a productive and reproductive nature can also be incorporated into breeding programs that intend to use genomic selection (Gutierrez et al 2021). Likewise, genetic profiling has reduced the margin of error in paternity testing and, therefore has increased the reliability of progeny testing. Furthermore, the identification of elite young cows and bulls has become more effective, allowing them to be used earlier in the selection and replacement process for sire generation. Consequently, this has led to a greater number of animals to be considered as future sires (Aranguren and Portillo, 2017).
According to Gutierrez et al (2021), in recent years, genomic studies have allowed the application of novel genome editing techniques such as CRISPR (Clustered Regularly Interspaced Short Palindromic Repeat) for the modification of genes associated with diseases in Japanese Black cattle (Wagyu), obtaining cattle that produce human serum albumin in milk and antiviral proteins such as lysostaphin, among others. In addition, hornless and non-allergenic dairy animals have been obtained in cattle. These techniques could help dairy cattle to better adapt to specific environmental conditions or production systems, which could improve fertility, growth, health and animal welfare in herds. On the other hand, it would allow the introduction of beneficial alleles such as those for heat tolerance or disease resistance (Van Eenennaam, 2019).
From the aspects that have been addressed in this work, it can be inferred that the application in Cuba of genomic studies will provide the advantages mentioned above, related to the decrease of costs and periods to identify and select outstanding individuals in bovine populations, the increase of the annual genetic gain rate, as well as those that allow better genetic characterization of the current dairy herds. This will help the livestock industry to genetically improve its herds in a targeted and more precise way, resulting in the selection of more productive, healthy and efficient animals, which will contribute to a great economic advance for livestock producers and strengthen capacities to improve the country food security by providing more and better milk and dairy products.
As mentioned, the application and use of genomic selection in Cuba has a high probability of success and will depend fundamentally on the breed, its improvement and conservation programs and selection objectives. In the future, it will be necessary to assume several challenges, such as the identification and selection of animals that conform the populations to the genotype, the creation of genotyping laboratories and the development of special strategies when analyzing small populations. It will also be necessary the use or development of economically accessible matrices, the implementation of statistical methods allowing more accurate reliability calculations, the acquisition of equipment capable of processing and storing databases combining genomic and genealogical information, the training of the personnel involved, and the adequate interrelation of the entities that participate in the development and application of these genomic studies. Undoubtedly, the application of this methodology will definitely contribute to the country strategic plans to achieve adequate food security in the production of milk and dairy products.
Acosta A, Uffo O, Sanz A, Obregón D , Ronda R, Osta R , Martin-Burriel I, Rodellar C and Zaragoza P 2014 Genetic characterization of the cuban water buffalo population using microsatellite DNA markers. Buffalo Bulletin: 33(1). http://ibic.lib.ku.ac.th/e-Bulletin/IBBU201401019.pdf
Acosta A C, Sanz A, Uffo O, Ronda R, Osta R, Martin I, Rodellar C and Zaragoza P 2011 Development of an eigth-plex SNP assay associated with milk production. Archivos de Zootecnia: 60:595- 606. https://www.uco.es/ucopress/az/index.php/az/issue/view/99
Acosta AC, Uffo O, Sanz A, Ronda R, Osta R, Rodellar C and Zaragoza P 2012 Genetic diversity and differentiation of five Cuban cattle breeds using 30 microsatellite loci. Journal of Animal Breeding and Genetics 130(1), 79–86. https://onlinelibrary.wiley.com/doi/10.1111/j.1439-0388.2012.00988.x
Aguilar I, Misztal I, Johnson D, Legarra A, Tsuruta S and Lawlor T 2010 Hot topic: A unified approach to utilize phenotypic, full pedigree, and genomic information for genetic evaluation of Holstein final score. Journal of Dairy Science:93(2), 743-752. https://www.sciencedirect.com/science/article/pii/S0022030210715174
Amaya M A, Martínez S R y Cerón M M 2020 Evaluaciones genéticas en bovinos por medio del uso del mejor predictor lineal insesgado genómico en una etapa. Ciencia y Tecnología Agropecuaria:21(1), e1548. http://revista.corpoica.org.co/index.php/revista/article/view/1548
Aranguren M J A and Portillo M G 2017 Selección genómica en la ganadería bovina. Revista Ecuatoriana de Ciencia Animal:1(2). ISSN 2602-8220. http://www.revistaecuatorianadecienciaanimal.com/index.php/RECA/article/view/18
Boison S A, Utsunomiya A T H, Santos D J A, Neves H H R, Carvalheiro R, Mészáros G, Utsunomiya Y T, do Carmo A S, Verneque R S, Machado M A, Panetto J C C, Garcia J F, Sölkner J and da Silva M V G B 2017 Accuracy of genomic predictions in Gyr (Bos indicus) dairy cattle. Journal of Dairy Science:100(7), doi: 10.3168/jds.2016-11811. https://www.sciencedirect.com/science/article/pii/S0022030217304976
Brown A, Ojango J, Gibson J, Coffey M, Okeyo M and M rode R 2016 Genomic selection in a crossbred cattle population using data from the dairy genetics East Africa project. Journal of Dairy Science: 99, 7308–7312. https://www.sciencedirect.com/science/article/pii/S002203021630443X
Butty M A, Sargolzaei M, Miglior F, Stothard P, Schenkel F S, Gredler-Grandl B and Baes C F 2019 Optimizing selection of the reference population for genotype imputation from array to sequence variants. Frontiers in Genetics:10: 510. doi.org/10.3389/fgene.2019.00510. https://www.frontiersin.org/articles/10.3389/fgene.2019.00510/full
Cole J B, VanRaden P M, O’Connell J R, Tassell C P V, Sonstegard T S, Schnabel R D, Taylor J F and Wiggans G R 2019 Distribution and location of genetic effects for dairy traits. Journal of Dairy Science:92(6):2931-46. doi.org/10.3168/jds.2008-1762 https://www.sciencedirect.com/science/article/pii/S0022030209706095
Delgado J V, Martínez A M, Acosta A, Álvarez L A, Armstrong E, Camacho E, Cañón J, Cortés O, Dunner S, Landi V, Marques J R, Martín-Burriel I, Martínez O R, Martínez R D, Melucci L, Muñoz J E, Penedo M C T, Postiglioni A, Quiróz J, Rodellar C, Sponenberg P, Uffo O, Ulloa-Arvizu R, Vega-Pla J L, Villalobos A, Zambrano D, Zaragoza P, Gama L T and Ginja C 2012 Genetic characterization of Latin-American Creole cattle using microsatellite markers. Animal Genetics.43(1):2-10. DOI: 10.1111/j.1365-2052.2011.02207.x, https://onlinelibrary.wiley.com/doi/10.1111/j.1365-2052.2011.02207.x
Ducrocq V, Laloe D, Swaminathan M, Rognon X, Tixier-Boichard M and Zerjal T 2018 Genomics for ruminants in developing countries: From principles to practice. Frontiers in Genetics: 9:251. doi: 10.3389/fgene.2018.0025. https://www.frontiersin.org/articles/10.3389/fgene.2018.00251/full
FAO 2014 Conservación de los recursos zoogenéticos (RZG) en animales de granja de la república de Cuba. Informe de país. http://www.fao.org.
Ginja C, Gama L T, Cortes Ó, Delgado J V, Dunner S, García D, Landi V, Martín-Burriel I, Martínez-Martínez A, Penedo M C T, Rodellar C, Zaragoza P, Cañon J and BioBovisConsortium 2013 Analysis of conservation priorities of Iberoamerican cattle based on autosomal microsatellite markers. Genetics Selection Evolution: 45:35. https://gsejournal.biomedcentral.com/articles/10.1186/1297-9686-45-35
Gutierrez Reinoso M A, Aponte P M and Garcia-Herreros M 2021 Genomic Analysis, Progress and Future Perspectives in Dairy Cattle Selection: A Review. Animals:11, 599. https://doi.org/10.3390/ ani11030599. https://www.mdpi.com/2076-2615/11/3/599
Harris B L and Johnson D L 2010 The impact of high density SNP chips on genomic evaluation in dairy cattle INTERBULL BULLETIN NO. 42. Riga, Latvia, May 31 - June 4, 40. https://journal.interbull.org/index.php/ib/article/view/1513
Henderson C 1975 Best Linear Unbiased Estimation and Prediction under a Selection Model. Biometrics:31(2): 423-447. https://pubmed.ncbi.nlm.nih.gov/1174616/
Hernández A, Ponce de León R y García S M 2015 Genetic evaluation of multibreed populations of milking bovines. Cuban Journal of Agriculture Science: 49(4): 459-463. https://www.cjascience.com/index.php/CJAS/issue/view/31
Hernández R A, Ponce-de-León-Sentí R, González-Prieto S, Fernández-Naranjo A, Prada-Gutiérrez-Reinoso N y Ramírez-Vázquez R 2020 Estrategias de selección para la mejora genética de las razas Siboney de Cuba y Mambí de Cuba. Anales de la Academia de Ciencias de Cuba:11(1). http://www.revistaccuba.cu/index.php/revacc/article/view/907.
Hidalgo D C, Swaminathan M, Deshpande P, Boussaha M, Saintilan R, Laloe D and Ducrocq V 2016 Interest and limits of the bovine 50K chip to study Bos taurus x Bos indicus crossbred animals in an Indian bull stud. In Proceedings EAAP Meeting (Belfast). https://meetings.eaap.org/wp-content/uploads/2022/01/EAAP2016BA.pdf
Legarra A Y and Ducrocq V 2012 Computational strategies for national integration of phenotypic, genomic, and pedigree data in a single-step best linear unbiased prediction. Journal of Dairy Science: 95(8), 4629-4645. doi:10.3168/jds.2011-4982. https://www.sciencedirect.com/science/article/pii/S0022030212004535
Lirón J P, Acosta A, Rogberg-Muñoz A, Uffo O, Posik DM, García J, Peral García P and Giovambattista G 2011 Origin of Cuban Creole Cattle inferred by patri and matrilineage. Archivos de Zootecnia:60:1171-80. https://www.uco.es/ucopress/az/index.php/az/issue/view/90
Lourenco D, Legarra A, Tsuruta S, Masuda Y, Aguilar I and Misztal I 2020 Single-Step Genomic Evaluations from Theory to Practice: Using SNP Chips and Sequence Data in BLUPF90. Genes:11, 790: 3-32. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7397237/pdf/genes-11-00790.pdf
Lourenco DAL, Tsuruta S, Fragomeni B O, Masuda Y, Aguilar I, Legarra A and Misztal I 2015 Genetic evaluation using single-step genomic best linear unbiased predictor in American Angus. Journal of Dairy Science:93(6), 2653-2662. https://pubmed.ncbi.nlm.nih.gov/35031806/
Martínez A M, Gama L T, Cañon J, Ginja C, Delgado J V, Dunner S, Landi V, Martín-Burriel I, Penedo M C T, Rodellar T, Vega-Pla J L, Acosta A, Álvarez L A, Camacho E, Cortés O, Marques S, Martínez S, Martínez R D, Martínez-Velázquez G, Muñoz J E, Postiglioni A, Quiroz J, Sponenberg P, Uffo O, Villalobos A, Zambrano D and Zaragoza P 2012 Genetic Footprints of Iberian Cattle in America 500 Years after the Arrival of Columbus. PLoS ONE: 7(11): e49066. doi:10.1371/journal.pone.0049066. https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0049066
Meuwissen THE, Hayes B and Goddard M E 2001 Prediction of total genetic value using genome-wide dense marker maps. Genetics:157:1819-29. DOI: 10.1093/genetics/157.4.1819 https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1461589/pdf/11290733.pdf
Misztal I, Lourenco D and Legarra A 2020 Current status of genomic evaluation. Journal of Animal Science: 98(4):1–14. DOI: 10.1093/jas/skaa101, https://academic.oup.com/jas/article-pdf/98/4/skaa101/33495645/skaa101.pdf
Misztal I 2016 Inexpensive computation of the inverse of the genomic relationship matrix in populations with small effective population size. Genetics. 202(2), 401-409. doi: 10.1534/genetics.115.182089 https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4788224/pdf/401.pdf
Misztal, I Legarra, A Aguilar I 2009 Computing procedures for genetic evaluation including phenotypic, full pedigree, and genomic information. Journal of Dairy Science:92(9), 4648-4655. doi:10.3168/jds.2009-2064. https://www.sciencedirect.com/science/article/pii/S0022030209707921
Mrode R, Ojango J M K, Okeyo A M and Mwacharo J M 2019 Genomic Selection and Use of Molecular Tools in Breeding Programs for Indigenous and Crossbred Cattle in Developing Countries: Current Status and Future Prospects. Frontiers in Genetics: 9:694. doi: 10.3389/fgene.2018.00694. https://www.frontiersin.org/articles/10.3389/fgene.2018.00694/full
Paccapelo M V 2015 Modelos de Selección Genómica para Caracteres Cuantitativos basados en Marcadores Moleculares aplicados al mejoramiento de maíz. (Tesis). Córdoba: Universidad Nacional de Córdoba. https://1library.co/document/download/q016rn3z
Proyecto Cuba – FOAR –AECID 2018 Mejoramiento de la información y las herramientas de gestión para la actualización de la política genética lechera de Cuba”.
Ramírez ALC, Rocha D, Ramos Onsins S E, Leno-Colorado J, Olivier-Bouchez M C, Rodríguez-Valera Yl, Pérez-Enciso M, and Ramayo-Caldas Y 2021 Whole genome sequencing reveals insights into the adaptation of French Charolais cattle to Cuban tropical conditions. Genetics Selection and Evolution:53:3. https://doi.org/10.1186/s12711‑020‑00597‑9 https://gsejournal.biomedcentral.com/articles/10.1186/s12711-020-00597-9
Rodríguez V Y, Renand G, Naves M, Fonseca J Y, Moreno-Robance T I, Ramos Onsins S , Rocha D and Ramayo-Caldas Y 2018 Genetic diversity and selection signatures of the beef ‘Charolais de Cuba’ breed. Scientific Reports:8:11005. DOI:10.1038/s41598-018-29453-z, https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6054659/pdf/41598_2018_Article_29453.pdf
Sanz A, Uffo O, Miranda I and Martínez S 2002 Empleo de los microsatélites para determinar paternidad en bovinos cubanos. Revista de Salud Animal:24(3): 166
Schöpke K and Swalve H H 2016 Review: Opportunities and challenges for small populations of dairy cattle in the era of genomics. Animal:10:6, pp 1050–1060. https://linkinghub.elsevier.com/retrieve/pii/S1751731116000410
Silva R M O, Fragomeni B O, Lourenco D A L, Magalhães A F B, Irano N, Carvalheiro R, Canesin R C, Mercadante M E Z, Boligone A A, Baldi F S, Misztalt I and Albuquerque L G 2016 Accuracies of genomic prediction of feed efficiency traits using different prediction and validation methods in an experimental Nelore cattle population. Journal of Animal Science:94, 3613–3623. DOI: 10.2527/jas.2016-0401, https://academic.oup.com/jas/article-lookup/doi/10.2527/jas.2016-0401
Uffo O and Martínez S 2002 Amplificación por PCR de los genes que codifican para la α-lactoalbúminaβ-lactoglobulina y la κ-caseína bovinas en una recordista y parte de su descendencia. Revista Salud Animal:24(1): 22-26
Uffo O and Acosta A 2009 Molecular diagnosis and control strategies for the relevant genetic diseases of cattle. Review. Biotecnología Aplicada:26:204-208. https://elfosscientiae.cigb.edu.cu/PDFs/Biotecnol%20Apl/2009/26/3/BA002603RV204-208.pdf
Uffo O, Martín-Burriel I, Martínez S, Ronda R, Osta R, Rodellar C, Zaragoza P 2006 Caracterización genética de seis proteínas lácteas en tres razas bovinas cubanas. Boletín AGRI-FAO:(39):15-24. https://www.fao.org/docrep/pdf/010/a0806t/a0806t02.pdf
Uffo O 2003 Aplicación de los marcadores moleculares al estudio de la biodiversidad del ganado bovino cubano. (Tesis doctoral). CENSA, La Habana, Cuba.
Uffo O, Acosta A, Martínez S, Ronda R 2012 Genetic characterization of Cuban Creole cattle using molecular tools. Biotecnología Aplicada:29:127-128. https://elfosscientiae.cigb.edu.cu/PDFs/Biotecnol%20Apl/2012/29/2/BA002902RP127-128.pdf
Van Eenennaam AL 2019 Application of genome editing in farm animals: Cattle. Transgenic Research; 28:93–100. doi.org/10.1007/s11248-019-00141-6, https://link.springer.com/article/10.1007/s11248-019-00141-6
VanRaden P M, Tassell C P V, Wiggans G R, Sonstegard T S, Schnabel R D, Taylor J F, Schenkel F S 2009 Reliability of genomic predictions for North American Holstein bulls. Journal of Dairy Science:92(1):16-24. doi:10.3168/jds.2008-1514, https://www.sciencedirect.com/science/article/pii/S0022030209703054