- Atlas
- Veterinary, Agricultural & Food Sciences
- Animal Genetics
Animal Genetics
Animal genetics is the science of inheritance in domesticated animals and the use of that science to change them.
10 frameworks, from 1750 to today · Animal Science · Veterinary, Agricultural & Food Sciences
- Blending inheritance and breed formation 1750-1900
- Biometry 1890-1920
- Mendelian inheritance 1900-1930
- Quantitative genetics 1918-Present
- Genetic marker 1950-2010
- Progeny testing and national genetic evaluation 1950-Present
- Marker-assisted selection 1990-2010
- Genome-wide association and functional genomics 2005-Present
- Genomic selection 2009-Present
- Genome editing 2012-Present
Overview
Animal genetics is the science of inheritance in domesticated animals and the use of that science to change them. Its practical aim is to make the next generation of cattle, pigs, sheep, goats, poultry, horses, or fish better suited to human purposes—more milk, faster growth, leaner meat, stronger wool, greater disease resistance—and its intellectual aim is to understand how traits pass from parents to offspring in populations shaped by human choice. What makes the field distinctive is that it is a science of managed evolution: the breeder, not nature alone, decides which animals reproduce, and the geneticist's job is to predict what those decisions will produce. The history of animal genetics is therefore a history of learning to measure inheritance, then to model it, then to read it directly in DNA.
Breeding by eye: blending inheritance and the herd book
For most of the history of domesticated animals, selection was a craft. Breeders judged animals by appearance, pedigree, and performance, and they assumed that offspring were a blend of their parents—that mating a good milker to a good milker would produce something in between, and that desirable qualities could be diluted or concentrated like a liquid. This "blending inheritance" was the working assumption behind the great breed-formation projects of the eighteenth and nineteenth centuries. Robert Bakewell in Leicestershire in the mid-1700s used close inbreeding and systematic selection to fix traits in Longhorn cattle and Leicester sheep, and his methods spread through the livestock world. Breed societies and herd books followed—the English Thoroughbred stud book from the late 1700s, the Shorthorn herd book in the 1820s, and many others—creating written pedigrees that let breeders track ancestry and line breeding.
The problem with blending inheritance was that it could not explain why variation persisted1. If inheritance truly blended, every population should become uniform within a few generations, and selection should have nothing left to work on1. Breeders also noticed that "prepotent" animals—those that reliably stamped their qualities on offspring—behaved in ways blending could not easily account for. The craft produced remarkable results, but it had no theory of why they worked or how to predict them.
Mendel, and the rediscovery that changed breeding
Gregor Mendel's experiments on garden peas, published in 1866, showed that inheritance is particulate: traits are governed by discrete factors (what we now call genes) that pass unchanged through generations, with dominant and recessive forms23. Mendel's work was ignored for decades, then rediscovered in 1900 by several biologists at once23. In animal science the Mendelian view was taken up quickly, because it explained what breeders had seen. An animal could carry a hidden recessive allele—a version of a gene—and pass it on without showing it, which accounted for traits that "skipped" generations and for the prepotency of certain sires.
The clearest early demonstration in animals came from coat color and from inherited defects. Mendelian ratios were confirmed in poultry, in guinea pigs, and in livestock coat-color inheritance in the early twentieth century. The practical consequence was a new kind of breeding: instead of blending, one could reason about which alleles an animal carried and predict the proportions of offspring that would show a trait. This made possible the deliberate management of simple traits—recessive defects, color, some horn conditions—and it gave breeders a vocabulary of carriers and homozygotes.
Biometry versus Mendel: measuring continuous traits
Mendel's factors worked for discrete traits, but most economically important traits—milk yield, growth rate, wool weight—vary continuously, and here a rival tradition had grown up. Francis Galton and Karl Pearson in Britain developed biometrical methods: statistics of correlation and regression applied to measurements of parents and offspring. They found that continuous traits were inherited in a statistical, gradual way, and some biometricians doubted that Mendelian factors could explain them. The dispute between Mendelians and biometricians was sharp in the early 1900s.
The reconciliation came from R. A. Fisher's 1918 paper showing that many Mendelian factors, each of small effect, together produce the continuous variation the biometricians measured1. This was the foundation of quantitative genetics: the idea that a continuous trait is controlled by many genes of small effect plus environmental variation. In animal science this mattered immediately, because it meant that the statistical tools of the biometricians and the particulate theory of the Mendelians were describing the same underlying system.
Quantitative genetics and breeding value
The next step was to turn this into a predictive theory of breeding. The key concept is breeding value: the value of an animal as a parent, measured as twice the deviation of its offspring from the population mean. An animal's own performance is a noisy signal of its breeding value, because environment and chance also affect it. Quantitative genetics, developed through the mid-twentieth century by Fisher, J. B. S. Haldane, Sewall Wright, and in animal breeding especially by Jay L. Lush at Iowa State, provided the framework for separating genetic from environmental effects.
Lush's 1937 book Animal Breeding Plans and his later work introduced the idea that selection should target breeding value estimated from records, not appearance. The tools were heritability (the fraction of variation in a trait due to genetic differences) and repeatability, and the method was selection index—combining information from an animal's own record, its relatives' records, and its progeny into a single score. The concrete case that showed the power of this approach was dairy cattle: milk records collected through milk-recording schemes let breeders estimate the genetic merit of bulls from their daughters' production, and selection on these estimates produced steady genetic gain where selection on appearance had not.
Progeny testing and national genetic evaluation
Because a dairy bull's own milk yield tells nothing about his daughters, the only reliable measure of his breeding value is the performance of his offspring. Progeny testing—waiting for a bull's daughters to calve and record their milk—became the backbone of dairy breeding in the mid-twentieth century. It was slow and expensive, but it worked, and it drove the rise of artificial insemination, which let a single tested bull sire thousands of daughters across a country.
The problem was scale and comparability: records from different herds, years, and environments could not be compared directly. The solution was national genetic evaluation—centralized statistical systems that combined records across herds, corrected for environmental effects (herd, year, season, age), and estimated breeding values for all animals on a common scale. These systems, built from the 1950s onward in dairy and later in beef, pigs, and poultry, used mixed-model statistics (best linear unbiased prediction, or BLUP) to handle the unbalanced, interconnected data of real populations. The result was that a breeder anywhere could compare an animal's genetic merit with the national population, and selection decisions became coordinated across an industry.
Molecular markers: reading the genome before sequencing
Through the twentieth century, genetics was inferred from pedigrees and performance; the genes themselves were invisible. That changed as molecular biology advanced. Blood groups and protein polymorphisms in the mid-twentieth century gave the first directly observable genetic markers—variants that could be typed in the laboratory and traced through families. They were useful for parentage testing and for mapping, but they were few and mostly not connected to economic traits.
The real breakthrough came with DNA markers: restriction fragment length polymorphisms, then microsatellites (short repeated DNA sequences that vary in length between individuals), then single nucleotide polymorphisms (SNPs—single-letter DNA differences). These gave dense, reliable markers across the genome. The first practical use in animals was parentage verification and traceability, but the larger prize was mapping: finding the chromosomal locations of genes affecting traits by tracking markers through pedigrees. Linkage mapping in reference families—large half-sib families from AI sires—let researchers locate regions of the genome associated with milk yield, growth, and disease resistance.
Major genes and marker-assisted selection
Some traits are controlled by a single gene of large effect, and these were the first targets. The halothane gene in pigs, which caused stress susceptibility and pale soft exudative meat, was mapped and later identified as a mutation in the ryanodine receptor gene. The double-muscling gene in cattle (myostatin) was another. The Booroola gene in sheep, which increases ovulation rate, was mapped in the 1990s. Identifying such major genes let breeders select directly for or against them with a DNA test, rather than waiting for the trait to appear.
This was the era of marker-assisted selection: using a marker linked to a known gene to select for the desirable allele4. It worked well for major genes, but for the many small-effect genes behind quantitative traits, markers explained too little variation to be worth much. The approach was also limited by the cost of genotyping and by the difficulty of finding markers tightly linked to the causal variants. Marker-assisted selection therefore delivered real gains for a handful of traits but did not transform breeding as a whole.
Genome-wide association and functional genomics
The sequencing of the human genome and then of livestock genomes in the 2000s changed the scale of what was possible. The chicken genome was sequenced in 2004, the cow in 2009, the pig in 2012, and the sheep in 2014. With a reference genome and cheap SNP chips that could type tens of thousands of markers at once, researchers could ask which markers across the whole genome were associated with a trait—genome-wide association studies (GWAS). Instead of tracking a few markers through families, GWAS compared many animals and found statistical associations between markers and phenotypes5.
GWAS located many regions associated with production, health, and reproduction traits, and functional genomics—the study of what genes do and when—began to connect those regions to actual biology. But GWAS also revealed a hard truth: most traits are affected by very many variants, each of tiny effect, scattered across the genome. A marker that is statistically associated with a trait in one population may not be the causal variant, and its effect may not transfer to another breed or environment. This limited the value of marker-assisted selection for complex traits and set the stage for a different approach.
Genomic selection
The idea that transformed animal breeding in the 2010s was genomic selection, proposed by Theo Meuwissen, Ben Hayes, and Mike Goddard in 2001. Instead of finding individual genes, genomic selection uses all the markers at once to predict an animal's breeding value6. A reference population of animals with both genotypes and phenotypes is used to estimate the effect of each marker on the trait; those effects are then summed across an animal's genotype to produce a genomic estimated breeding value (GEBV)6. The prediction captures the combined effect of many small variants without needing to know which are causal.
The method became practical when SNP chips made dense genotyping cheap. Dairy cattle led the way: in 2009 several countries introduced genomic evaluations for Holsteins, and the effect was dramatic. A young bull could be evaluated from a DNA sample at birth, without waiting years for daughters to calve, which cut the generation interval and roughly doubled the rate of genetic gain in some dairy populations. Genomic selection spread to beef cattle, pigs, poultry, sheep, and aquaculture, and it is now the dominant method for genetic evaluation in many species. Its limits are real: prediction accuracy depends on the reference population being close to the animals being predicted, and accuracy falls when breeds or environments differ.
Gene editing and the present
The newest turn is gene editing—precise alteration of an animal's DNA using tools such as CRISPR-Cas9, developed in 2012. Unlike earlier genetic modification, which inserted foreign genes, editing can make small changes at a chosen site, including changes that mimic naturally occurring variants. In livestock, editing has been used experimentally to produce polled (hornless) cattle, to improve disease resistance (for example, resistance to porcine reproductive and respiratory syndrome), and to alter production traits. The science is advancing faster than regulation and public acceptance, and edited animals face different rules in different countries.
Today animal genetics rests on the same foundations laid across this history: Mendelian inheritance, quantitative genetics and breeding value theory, national genetic evaluation, and molecular genomics. Genomic selection is the working method in most major livestock species, with GWAS and functional genomics supplying the biological understanding behind the predictions. The open questions are substantial: how to predict across breeds and environments, how to include traits that are hard to measure (disease resistance, welfare, emissions), how to manage the loss of genetic diversity that intense selection can cause, and how gene editing will be governed and accepted. The field's central problem—how to choose which animals reproduce so that the next generation is better—remains the same; what has changed is how precisely the breeder can read the inheritance that selection acts on.
References
- Population genetics (History), Wikipedia
- Mendelian inheritance (introduction, History), Wikipedia
- Gregor Mendel, Wikipedia
- Marker-assisted selection, Wikipedia
- Genome-wide association study, Wikipedia
- Genomic selection, Wikipedia
The frameworks of Animal Genetics
Each framework is a way of working the field has used: when it held, and the question it answers.
- 1750-1900
Blending inheritance and breed formation
How do offspring resemble their parents?
- 1890-1920
- 1900-1930
- 1918-Present
- 1950-2010
- 1950-Present
Progeny testing and national genetic evaluation
How can genetic merit be compared across herds?
- 1990-2010
Marker-assisted selection
Can markers predict which animals carry useful genes?
- 2005-Present
Genome-wide association and functional genomics
Which genome variants affect traits?
- 2009-Present
- 2012-Present
How the ideas connect
Each arrow runs from an earlier idea to the later one it shaped. The list below says what passed between them.
Mendelian inheritance rival of Biometry
After 1900 Mendelians and biometricians competed over whether inheritance was particulate or continuous, with the Mendelians arguing that discrete factors, not blending, explained heredity. The dispute was sharp in the early 1900s before the two views were reconciled.
Carried by William Bateson; Karl Pearson.
Quantitative genetics influenced by Biometry
R. A. Fisher's 1918 paper used the biometricians' statistical machinery of correlation and variance, showing that many Mendelian factors of small effect produce the continuous variation they had measured. This became the foundation of quantitative genetics.
Carried by R. A. Fisher; The Correlation between Relatives on the Supposition of Mendelian Inheritance (1918).
Quantitative genetics influenced by Mendelian inheritance
Fisher's reconciliation took the Mendelian premise of discrete factors and showed that many such factors, each of small effect, sum to continuous variation. Quantitative genetics is built on that particulate model of inheritance.
Carried by R. A. Fisher; The Correlation between Relatives on the Supposition of Mendelian Inheritance (1918).
Progeny testing and national genetic evaluation grew out of Quantitative genetics
National genetic evaluation applied Lush's quantitative-genetic concepts of breeding value and heritability to progeny-test records, using mixed-model statistics to estimate breeding values across herds. The evaluation systems are a direct practical branch of the breeding-value theory.
Carried by Jay L. Lush; Animal Breeding Plans (1937).
Genomic selection influenced by Quantitative genetics
Genomic selection is quantitative genetics applied to genomic data: breeding values are estimated from genome-wide markers using the same infinitesimal model and mixed-model machinery. The prediction of genetic merit remains the goal.
Marker-assisted selection grew out of Genetic marker
Marker-assisted selection grew directly out of DNA marker technology: once microsatellites and SNPs allowed linkage mapping, breeders used markers linked to major genes and QTL to select animals. The halothane gene in pigs was an early target.
Genome-wide association and functional genomics grew out of Genetic marker
Genome-wide association and functional genomics extended marker technology from sparse markers and linkage mapping to dense SNP panels scanned across the whole genome. The marker typing methods and mapping logic of the earlier period carried directly into GWAS.
Genomic selection influenced by Genome-wide association and functional genomics
Genomic selection took the dense genome-wide marker data and statistical association methods of GWAS and turned them into prediction: all markers are fitted simultaneously to estimate genomic breeding values. The approach was demonstrated in dairy cattle and other livestock.
Genetic marker influenced by Quantitative genetics
Quantitative genetics supplied the statistical framework of linkage mapping and the idea that traits are controlled by many loci, which motivated the search for directly observable markers. Marker work in animal genetics was framed around mapping quantitative trait loci.
Genomic selection replaced Marker-assisted selection
Genomic selection replaced marker-assisted selection as the working method: instead of selecting on a few significant markers, it uses dense genome-wide markers to predict breeding values. The shift was made possible by cheap SNP genotyping and was framed as overcoming MAS's limited success with complex traits.
Beyond the field
Where Animal Genetics's ideas came from in other fields, and what they led to there: the strongest 12 of 79 links.
Blending inheritance and breed formation replaced by Mendelian inheritance in Genetics
Mendelism replaced the older blending view of inheritance used in breed formation, showing that hereditary factors segregate rather than blending irreversibly.
Carried by Gregor Mendel.
Mendelian inheritance influenced Quantitative Genetics in Genetics
Mendelian inheritance provided the discrete factors whose combined action quantitative genetics models as continuous variation.
Mendelian inheritance led to Molecular Genetics in Genetics
Molecular genetics took over Mendelian inheritance's central problem, the transmission of discrete hereditary factors, and redefined the factor as a DNA sequence after 1953.
Carried by James D. Watson; Francis Crick.
Mendelian inheritance influenced Mendelian randomization in Nutritional Epidemiology
Mendelian randomization rests on Mendel's laws of inheritance, particularly the random segregation of alleles at conception, which provides the instrument's independence from confounding. The method's validity argument is built directly on this Mendelian principle.
Carried by Gregor Mendel.
Mendelian inheritance led to Mendelian–Biometrician synthesis in Population Genetics
The Mendelian-biometrician synthesis grew directly out of Mendelian inheritance by showing that many Mendelian genes of small effect yield continuous variation. Fisher's 1918 paper is the founding statement of this reconciliation.
Carried by R. A. Fisher.
Mendelian inheritance influenced Population Genetics in Genetics
Mendelian inheritance supplied the particulate basis of heredity on which population genetics built its frequency models.
Mendelian inheritance influenced Quantitative genetics in Plant Breeding
Fisher's 1918 reconciliation of Mendelian inheritance with continuous variation founded quantitative genetics; the biometrician-Mendelian dispute over inheritance was the problem the theory resolved.
Carried by Ronald A. Fisher.
Genetic marker influenced Marker-assisted selection in Plant Breeding
Genetic markers, especially isozymes and later DNA markers, are the essential instrument of marker-assisted selection; without mapped markers linked to traits the method does not exist. The marker technology developed in animal and plant genetics was taken up directly by MAS.
Genetic marker influenced Marker-assisted selection in Pomology
Marker-assisted selection depends on genetic markers, whose development from the mid-twentieth century onward provided the tools the method uses. The concept of a heritable marker linked to a trait passed directly into MAS.
Progeny testing and national genetic evaluation influenced by Quantitative Genetics and Breeding Value Selection in Animal Production Systems
Quantitative genetics and breeding value selection provided the theory of breeding values and selection response on which progeny testing and national genetic evaluation were built.
Carried by Ronald Fisher; Jay Lush.
Progeny testing and national genetic evaluation replaced by Reproductive Genomics and Genomic Selection in Theriogenology
Genomic selection replaced and extended progeny testing and national genetic evaluation, using the same breeding-value framework but estimating it from genomic markers. The evaluation infrastructure and selection goals came from progeny testing.
Marker-assisted selection led to Genomic selection in Dairy Science
Genomic selection grew directly out of marker-assisted selection: it generalises MAS from a few significant markers to dense genome-wide marker panels, using all markers simultaneously to estimate breeding values. Meuwissen, Hayes and Goddard's 2001 paper framed this as an extension of marker-based prediction.
Carried by Theo Meuwissen; Ben Hayes; Mike Goddard.

Animal Genetics is a planet in Nousfarer
Its city has a tower for each of these 10 frameworks. Land, walk in, and climb one concept at a time.
Land on Animal GeneticsRelated fields
Fields its ideas are linked to
- Genetics 13 links
- Plant Breeding 8 links
- Animal Production Systems 7 links
- Animal Science 7 links
- Genetic Epidemiology 4 links
- Dairy Science 3 links
- Meat Science 3 links
- Poultry Science 3 links
- Molecular Biology 3 links
- Medical Genetics 3 links
- Aquaculture Genetics 2 links
- Genomics 2 links
More in Animal Science
- Animal Nutrition
- Animal Physiology
- Animal Production Systems
- Animal Science
- Animal Welfare
- Dairy Science
- Meat Science
- Poultry Science
- Reproductive Physiology
See where Animal Genetics sits on the Map of Human Knowledge