1st Page

Animal Frontiers - From the Editors

Animal breeding in the genomics era


This article in

  1. Vol. 6 No. 1, p. 4-5
    Published: January 5, 2016

    * Corresponding author(s):
Request Permissions

  1. Noelia Ibanez-Escriche ** and
  2. Henner Simianer *
  1. * Division of Genetics and Genomics, The Roslin Institute, Royal (Dick) School of Veterinary Studies, University of Edinburgh, Roslin, UK and Department of Animal Breeding, The Institute for Research and Technology in Food and Agriculture (IRTA), Catalonia, Spain
     Animal Breeding and Genetics Group, Department of Animal Sciences, Georg-August-University, Goettingen, Germany

It has been 15 yr now since the genomic revolution in animal breeding was triggered by the publication of the seminal paper by Theo Meuwissen, Ben Hayes, and Mike Goddard (Meuwissen et al., 2001). As pointed out in the review provided by those three authors to the current issue (Meuwissen et al., 2016), there were mainly three cornerstones that needed to be in place to make genomic selection practically feasible:

  • ■ A suitable methodological framework, as presented in the abovementioned paper;

  • ■ A dense map of single-nucleotide polymorphism (SNP) markers, as a by-product of the initiatives to develop reference genomes for all major farm animal species; and

  • ■ A technology to perform high-throughput genotyping for large (10,000 or more) sets of SNPs at reasonable costs.

Dairy cattle breeding was the first field for which all of these conditions were met. Additionally, dairy cattle made the perfect case for a pilot application since genomic selection provided a realistic alternative to progeny testing of young bulls, which was both costly and time consuming, and thus made dairy cattle breeding programs expensive and limited the achievable genetic gain per year due to extremely long generation intervals. The potential for a huge improvement was nicely formulated by Schaeffer (2006), who gave a further motivation to publish the basic concept for genomic selection programs quickly by stating that it would then be available to everyone so that no one could patent those ideas.

A representation of a single nucleotide polymorphism. (source: Splettstoesser).


By the end of the first decade of this century, genomic selection was up and running in most of the major dairy cattle breeding programs around the world. Implementations varied in some details, and genomically selected young bulls were used at different rates. Overall, genomic selection in dairy cattle can already be considered as a success story that increased genetic progress significantly, not only for productivity, but also in functional trait complexes linked to health, animal welfare, and environmental impact. Having said this, it should be remembered that we are still in the initial phase of using these technologies, and further improvements, both conceptual and technological, are still to be expected.

Implementations in all other major farm animal species followed quickly; however, it is likely the advantage of genomic selection will not be as high in those species as it is in dairy cattle breeding. In species such as pigs and poultry, this is mainly due to the short generation intervals and accurate conventional breeding values. Also, the genotyping costs relative to the actual value of a selection candidate are rather unfavorable in those species. In other cases, such as beef cattle and horse breeding, the structure of breeding programs, e.g., with limited use of artificial insemination or a lack of consistent performance testing for relevant traits, limits the potential for a successful implementation of genomic selection. But still, improvements of genetic progress per year in the range of 10% can be expected in all species, and hardly anybody would question that 21st century animal breeding programs in all species will be built around phenotypic, pedigree, and genomic information.

As a by-product, genomic selection programs provide high-density genotypes for often huge numbers of animals, and in most cases, very detailed phenotype data are available for the same set of individuals. Mike Coffey’s often quoted statement “In the era of genomics, phenotype is king” stresses the crucial importance of high quality phenotypic data. High-throughput genotyping will only unfold its full potential if it is paralleled by high-throughput phenotyping as, e.g., in the context of “precision livestock farming,” and if the range of trait complexes for which high-precision phenotypes are available is extended to areas such as health, behavior, resource efficiency, and environmental impact.

Arguably, the only other species for which data in comparable quantities and qualities are available is the human species, and even there, certain phenotypes—such as exact daily feed consumption and weight gain for thousands of related individuals as can be recorded in pig and poultry—are impossible to achieve. Animal breeders have a real asset here, and the animal breeding community does not yet appear to have sufficiently realized the (scientific and commercial) potential of the availability of such data.

The data already available provide a basis for tackling many other research questions beyond the field of genomic selection. High-throughput genotypes in combination with detailed phenotypes can be used for powerful genome-wide association studies, identifying regions of the genome that are directly associated with variability in phenotypes. This will help to better understand the genetic background of performance traits as well as functional traits and diseases. Another emerging field is the detection of selection signatures by screening the genome for unusual patterns that cannot be attributed to neutral mechanisms such as genetic drift. Understanding the mechanisms of selection on a genomic basis may, in the long term, help to design more efficient breeding programs (see the review of Qanbari and Simianer, 2014). Genomic data can also be used for designing more efficient crossbreeding strategies, for assessing and conserving genetic diversity, and for managing production. One may even think of genotype-based “individualized management” or “individualized feeding” in farm animals as an analog to “individualized medicine” in the human genetics field.

What are the perspectives and challenges in the years ahead? A clear trend is that genotyping (especially with low-density genotyping arrays) is going to be used more widely, and it is very likely, that a major proportion—if not all—of the breeding animals will be genotyped in the future. On the other hand, whole-genome sequencing is becoming available on a larger scale, as e.g., in the 1,000-bull genomes project (Daetwyler et al., 2014), and a strategy of sequencing an (often small) number of key ancestors followed by imputing the sequence to all genotyped individuals will provide an unprecedented data basis for genomic studies. While the benefit of a sequence- rather than genotype-based prediction appears limited (for details, Meuwissen et al., 2016 in this issue), working with sequences will be extremely helpful in many other instances, e.g., in identifying causal variants for certain phenotypes (Pausch et al., 2014).

Developments in genotyping and sequencing are accompanied by rapid evolvement of technological innovations in related fields: whole-genome expression studies, transcriptomics, and metabolomics provide insight into the complex biological processes linking genotype and phenotype; epigenetics focus on the interaction of the genome with the environment; and metagenomics study the genetic dynamics of the microbiome of an organism, most relevantly in the rumen of cattle and small ruminants. We are only at the beginning of generating large-scale data in those fields, but integrating those results with genomic data will strongly improve our biological understanding of genetic causes and mechanisms underlying relevant phenotypes, allowing more knowledge-based breeding strategies to be developed.

Population-wide genotyping in combination with imputation of sequence information and high-throughput phenotyping will lead to an explosion of the data volume to be handled and poses enormous challenges to bioinformatics and biostatistics in animal breeding. Significant investments in hardware, software, and brains, both in research and industry, will be required to develop concepts, systems, and tools for efficient data management as well as advanced statistical methods, and thus make optimal use of the new wealth of information.

Noelia Ibanez-Escriche is a senior researcher with more than 13 years of experience in quantitative genetics and animal breeding. Her Ph.D. was conducted at the University Polytechnic of Valencia, Spain and the Institute of Agricultural Science, Denmark. She was employed as a researcher at the Agricultural Research Institute of Catalonia (IRTA), Spain from 2006 to 2015. In 2008, she worked as a post-doctoral visitor at Iowa State University, USA. In 2015, she obtained a Marie Skłodowska-Curie grant to work as a senior researcher at the division of Genetics and Genomics of the Roslin Institute at Edinburgh University, UK. She has made major contributions to the development of Bayesian methods to analyze complex traits and improve the genetic/genomic selection approaches of crossbred animals. Dr. Ibanez has been leading several national and international research projects. She has published more than 100 peer-reviewed papers and conference papers, some of them as invited speaker. She has been the secretary of the genetics commission of EAAP since 2011.

Henner Simianer is Professor of Animal Breeding and Genetics at Georg-August-University in Goettingen, Germany. Coming from a quantitative genetics background, he has made major contributions to the optimum design of animal breeding programs in most farm animal species. His current research focuses on a better understanding of the dynamics of farm animal genomes under selection. The massive high-throughput genotyping and whole-genome sequence data that have become available in the course of the “genomic revolution” provide an excellent basis to unravel footprints of selection in the genome, and first results seem to confirm the high genetic complexity of most traits of interest to breeders. Recently, he widened his interest beyond animal breeding, trying to reveal methodological synergies between animal and plant breeding. He was chair of the program committee for the 2010 World Congress on Animal Genetics in Livestock Production (WCGALP) in Leipzig and president of the Genetics Commission of EAAP from 2010 to 2015.




Be the first to comment.

Please log in to post a comment.
*Society members, certified professionals, and authors are permitted to comment.