down cross right results icon premium content video video hollow icon audio lifeNews icon-comment tick starFilled betSlip hot icon-liveCommentary refresh spinner arrow-down

Professor Emmeline Hill answers our readers' questions about genetics

The eminent equine geneticist responds to your enquiries

Professor Emmeline Hill: leads research and development at Plusvital
1 of 1

World renowned equine geneticist Professor Emmeline Hill generously gave her time to answer our readers' questions.

Published below are her responses, with some questions shortened for the reason of space.

To find out more about Professor Hill's work leading the research and development team at Plusvital, and the range of equine genetic tests the company offers, click here.

Assuming Horse A is TT and Horse B is CT, and they are half-brothers, why does Horse A get faster and more precocious stock?

Also, how do stallions – like Sageburg – who cannot be TT get National Hunt runners?

Alistair Fairclough

The ‘speed gene’ contributes to distance aptitude. Our research has shown that CC horses are typically more precocious than CT and TT horses. A CT horse can produce all three types while a TT horse can only produce CT and TT types. Generally CT stallions produce more precocious progeny than TT stallions, but of course it will depend on the mares that a horse gets.

Precocity, however, is also associated with genes that influence behaviour (e.g. the HTR1A gene, which is a serotonin receptor gene) and there are hundreds, if not thousands, of other genes that contribute to all of the other anatomical, physiological and metabolic traits that are required for the development of an athlete.

It is important to note that most traits are not ‘simple’ traits (influenced by a single gene), but rather they are ‘complex’ genetic traits with contributions from multiple genes, their interactions and the environment in which a horse is managed.

We analysed the ‘speed gene’ in National Hunt winners and found that there were just as many CTs as TTs. The only difference we found was that all of the three-mile chase winners we tested were TT. A non-TT stallion can produce CT and TT progeny and therefore could produce NH runners.

Can you explain how a horse's colour comes about; how does a chestnut mare have a bay foal and vice versa?

In particular, how does grey comes about - can two greys have a chestnut etc?

Sam Murphy

Coat colour in horses is influenced by a large number of genes. The chestnut coat colour arises if a horse inherits two copies of the recessive version of the ‘red’ coat colour gene. Therefore, both parents must be carriers of the recessive version of the gene.

That means they could both be bay (but carriers of the ‘red’ gene version) or they may be chestnut themselves. The ‘grey’ gene on the other hand is dominant. This means that a grey horse must have at least one parent that is grey. If two grey horses have only one copy of the ‘grey’ gene and are carriers of the ‘red’ gene then there is a one in four chance that these two grey horses could produce a chestnut foal.

Is there any evidence to suggest that wind problems (as in soft palate displacement and laryngeal weakness/paralysis) are passed on through genetics – and if so would the increase in inbreeding be likely to worsen this scenario?

There seems to be plenty of anecdotal evidence that certain sires have a tendency to produce offspring with breathing issues – just wondering if this can be backed up by scientific evidence.

Debby Ewing

Recurrent laryngeal neuropathy (RLN) is considered a complex genetic trait, and a genetic predisposition for RLN is supported by moderate (32–46 per cent) heritability estimates.

However, the underlying genes and mechanisms of inheritance have not yet been identified. There is some evidence for an association with the major gene controlling height (LCORL), but it is not clear if this is causing the disease or associated with the incidence of disease. It is possible that there are multiple genes that contribute to wind problems and these genes may vary within families.

Stallion performance at stud is not always consistent over a long career. Many people put this down to quality of mares, perhaps influenced by the 'random' emergence of top performers over the years.

Is there also a genetic influence whereby gene expression changes over the years and that this could also be an influence on progeny performance. If so, does Plusvital monitor changes in gene expression or is this fixed when a stallion enters stud?

Jim Smith

There is evidence from studies in humans and other mammalian species that paternal ageing can affect offspring disease susceptibility. This has been attributed to epigenetic changes in the sperm.

Therefore it is a reasonable hypothesis that similar mechanisms could affect genes in horses that contribute to traits of importance given the evidence in other mammalian species with ageing.

Plusvital’s genetic tests evaluate the static DNA of the horse. Monitoring gene expression changes would require a large-scale systematic study to identify epigenetic marks that may be predictive of outcomes.

What's the probability that potential defects in a mare will be inherited by her offspring? As an example, a mare who broke her pelvis at two years old may suggest a weakness in that area – how likely is it that this will be inherited?

Jay Deves

Heritability is the measure of the genetic contribution to a trait. The heritability varies for different traits and must be measured for each trait of interest. Some traits are entirely due to genetics (i.e. coat colour) and some are not at all, but most fall somewhere in between.

This means that the likelihood of a mare passing on defects to her progeny will depend on 1) the heritability of the trait and 2) whether the mare is a carrier of risk alleles.

We are not at a point yet where we understand all of the genetic contributions to traits, but the international community of equine genetics researchers is focused on using new genetic technologies to identify risk markers for disease that may help with breeding decisions.

For more information see

I would like to ask how you see the future of the thoroughbred, with regards to the fact that 70 per cent or more of all stakes/Graded winners now have Northern Dancer blood, particularly Galileo and Sadler's Wells genetics, in their pedigree?

Is it a good thing to have a diamond shaped gene pool like this?

Rod Orrell

During the last four decades there has been a significant increase in inbreeding in the population which may be due to strong selection coupled with changes in breeding practices in which there has been a greater emphasis on smaller numbers of popular sire-lines in recent years. While there is evidence for a significant reduction in genetic diversity within the breed, it is not clear how much inbreeding is tolerable.

Nonetheless, populations with lowered genetic diversity can suffer from inbreeding depression, which is the accumulation of deleterious recessive mutations that increase in frequency and may be expressed in the homozygous state.

These mutations most commonly negatively affect health and fertility traits. Our current research focus is in this area and we are developing tools that breeders will be able to use to make informed breeding decisions on the basis of DNA information.

Recent studies in other species have demonstrated the considerable power of epigenetics, i.e. the influence of environment on gene expression.

This gives credence to what many traditionalists have always thought – a horse’s performance may to some extent be pre-determined by the conditions in which the dam gestates, the nursery in which the foal matures, the pre-training and training style and schedule to which it is then subjected.

Does this mean that hard-wired genetics makes up less than 50 per cent of the equation after all?

Dr Pat Sells

Depending on the trait, the inherited DNA has a varying contribution. This is very high for traits such distance, which is largely influenced by the ‘speed gene’, but is less for other traits.

Environmental stimuli can have profound effects on cell, tissue and organismal biology through epigenetic regulatory mechanisms. It is well established that a dynamic interplay exists between the static DNA sequence of the genome and the environment, with epigenomic modifications mediating the interactions between the genome, gene expression and the environment.

For instance, it is becoming increasingly apparent that epigenetic regulatory mechanisms are key features of the modification of behavioural traits as well as the skeletal muscle response to exercise. Gene × environment interactions determine how individuals with the same or different genotypes will respond to variation in the environment.

Early life experiences contribute to variation in behaviour that was previously thought to have been directly inherited. Stressful early life events have long-lasting effects into adulthood in humans but exercise is known to have a beneficial influence on stress. Therefore, early lifetime experiences very likely influence outcomes for racehorses via epigenetic modifications to the DNA.

The heritability (genetic contribution) of racing performance traits indicates that ‘hard-wired’ genetics, or the inherited DNA element, is only a part of the puzzle. The environment in which a horse is raised and managed likely plays a critical role in shaping the expression of genes that are inherited. That is why genetic tests will never be a ‘silver bullet’.

However, starting with the optimal genetic package will contribute to the chances of producing an elite racehorse.

You have recently publicised work indicating that inbreeding is increasing in the modern thoroughbred as against those a few decades ago. The implication that seems to be drawn is that this is a dangerous phenomenon affecting the diversity of the breed, and needs to be guarded against.

Rightly or wrongly I am not surprised by the finding of increased homozygosity since my researches into pedigrees indicate that the two most influential stallions of the last century, Northern Dancer and Mr Prospector, were born in the latter half of the century.

No stallion born since 1880 approached their influence. Accordingly I would suggest that the resulting inbreeding has arisen from selection of stallions and mares on racing ability and their ability to produce racing ability. If so, increased inbreeding may be an irrelevant side effect.

My question is whether, as well as testing homozygosity in a range of modern thoroughbreds vs a range of past thoroughbreds, you have also tested it in the supreme examples of the modern European thoroughbred, the stallions Galileo and Dubawi and the racehorses Frankel and Sea The Stars.

If their inbreeding co-efficients are below those of their peers then it might indicate a problem/opportunity. If not, the whole study would seem of purely academic interest. So which is it?

Bryan Mayoh

There are two ways to assess inbreeding – one is to look at the average amount of similarity (homozygosity) across the whole genome, which provides an overall measure of inbreeding, and another is to look at the length of segments of the genome that are similar (homozygous).

This latter measure allows an assessment of inbreeding that is due to old/historic positive selection, for instance from the ancestral stallions you mentioned, and inbreeding that is due to more recent co-ancestry. It is the recent co-ancestry that has the potential to cause problems since this is where harmful mutations can quickly increase in frequency if carried in popular sire-lines. In other words, there is ‘good’ inbreeding and there is ‘bad’ inbreeding.

While we cannot comment on inbreeding levels of individual horses, we have evaluated inbreeding levels in more than 300 stallions. The distribution of inbreeding levels in stallions reflects the population levels. The issue of increased inbreeding may be less driven by the level of the inbreeding in the stallion population but rather how closely related the stallions are to the mares in the population. Pedigree is no longer a good indicator of that relationship.

We are now developing genomics-based solutions that will allow breeders to understand – at a genetic level – the genetic relationship between a mare and a stallion and the likely inbreeding level of a hypothetical foal arising from a particular mating.

I'd like to ask about the 'exceptions to the rule', i.e. those that fall outside the 90 per cent prediction of optimum race distance. Has any systematic investigation been done into e.g. CC stayers or TT sprinters, assuming there are significant enough numbers?

If so, is there anything common to their circumstances? Could these also be indicative of more deep-seated genetic inheritance 'smearing' the results, which might in turn make the third, fourth and fifth generation pedigree relevant as much as the direct parents from where the C and T genes come from?

Christopher Veale

The ‘speed gene’ test assesses a genetic marker that is ‘linked’ to the actual mutation that causes the differences in myostatin protein production that influences muscle development.

In rare cases the genetic marker is not linked with the mutation and therefore variation can arise. Also, although MSTN (the ‘speed gene’) is the single most important genetic contributor to best race distance in the thoroughbred, we have identified other genetic markers that also contribute to the trait.

Metabolic requirements and physiological responses differ considerably for endurance and shorter distance intense exercise. Endurance is associated with increased mitochondrial abundance and a shift in substrate preference from carbohydrates to fatty acids; conversely, during intense sprint exercise there is an increased reliance on anaerobic metabolism and localised hypoxia may occur in muscle.

We have identified additional genes that may contribute to optimal race distance through modulation of substrate availability and the response to exercise induced hypoxia. These gene variants can then modify variation in distance that is primarily influenced by the ‘speed gene’ and this is how the outliers may arise.

Regarding the 'speed gene': is this not a misnomer? Speed is an abstract. I am not doubting the influence of CC, CT and TT copies, but is it not the case that the MSTN and PPARGC1A polymorphisms are not the only important variants relevant to performance. I seem to recall that Byron Rogers' company, Performance Genetics, did some tests that looked at a number of polymorphisms across the genome to determine the genetic component of a horse's profile.

I'm not a scientist - I'm an engineer, with a great interest in the history of the thoroughbred racehorse, so would be grateful for a comment. Perhaps the 'speed gene' should be renamed?

Tony Byles

Following our original publications that identified the ‘speed gene’, we published a paper that demonstrated a strong relationship between the myostatin (MSTN) gene and measured speed.

We measured speeds using GPS and heart rate monitoring equipment during work-day gallops in 85 horses and found that CC horses outperformed TT horses for all of the measured speed variables.

Our findings clearly indicated that the ‘speed gene’ influences measured speed in the thoroughbred. It has also been reported that the ‘speed gene’ is the most strongly selected gene of the 20,000 genes in Quarter horses. Since Quarter horses have been specifically selected for speed, this further supports our designation of MSTN as the ‘speed gene’ .

However, the ‘speed gene’ is not the only contributor to speed. More recently we measured speed variables in 294 horses and assessed the contribution from the 20,000 genes in the horse genome.

We found two other muscle related genes – MYO18A and MYO18B – that are associated with the median speeds of 2yos (measured during work-day gallops) and the increase in speed in response to training during the two-year-old training season.

We have also identified genetic markers for early two-year-old speed that are important for racing in Australia.

If a mare is tested for Genomic Breeding Value, how can this determine the value of “the number of favourable genetic markers that a horse may pass on to their foal?” as surely unless you choose the correct stallion the value must be nearly irrelevant.

So, if a mare is sent to six different stallions of varying stud fees and pedigrees and only one of them is a winner at Listed level, the Genomic Breeding Value is useless unless you have done the same test for every stallion and can suggest the most compatible stallion on genetics!

As if you have a mare with a class 1 value, a breeder could still make a mess of things by going to a “unsuitable stallion” as far as genetics are concerned?

Steve MacDonald

The Genomic Breeding Value (GBV) (Plusvital Premium Pack) is an evaluation of the additive genetic contribution to performance potential. It represents the potential to pass on favourable genetic variants to progeny.

The GBV is different from the Genomic Racing Value (GRV) since for breeding, each individual passes on just one copy of its genetic material (genome) to its progeny. The GRV and GBV scores therefore will usually be different for an individual. The breeding value scores the potential genetic contribution to offspring and is independent of the contribution of the other parent.

We are now developing tests that look at the interaction between the mare and stallion DNA that will provide breeders with information on the genetic relationship between a mare and a stallion and the likely inbreeding level of a hypothetical foal. Mare reproductive health is also an area we are researching which will lead to tests to identify suitable stallions to improve mare fertility.

If a mare is covered by the same stallion say five times and produces a group horse, a couple of moderate handicappers and two who are very poor, can genetic tests identify which of the family are more likely to be talented at the yearling stage?

And if each of the winners won over different distances, could genetics have identified the trip each of the individuals would be better over? (i.e. the speed gene test)

As in breeding and buying, I wouldn’t pay over the odds for a yearling which is a full-brother or sister to a very smart horse, but it would not put me off buying a full-brother or sister to a moderate horse if I liked the conformation and individual.

Steve MacDonald

The value of genetic testing is for exactly this purpose, to identify the gene variants that an individual has actually inherited, rather than using pedigree information to try to best guess the genes inherited from ancestors in a pedigree.

Full-siblings have identical pedigree pages, but their genetics will in all cases be different. Breeding is a genetic lottery. There are 32 pairs of chromosomes in the horse genome. During the generation of sperm and egg cells just one chromosome from each parental pair is passed on. Which one of the chromosome pairs is passed on is completely random, making each combination unique.

The number of possible chromosome combinations from a single mare × stallion mating is huge – two to the power of 32 = 4 billion!

Genetic tests can identify the combination of genes that has been uniquely inherited by that individual. It is important to consider that it is the combination of genes that is critical. No one gene alone is sufficient to guarantee success, but each favourable gene that a horse inherits will provide marginal gains such that the greater the number of favourable gene versions that a horse inherits, the greater likelihood of success.

Our established prediction algorithms have identified the combinations of genes that contribute to various racecourse outcomes such as the chance of having a racecourse start as atwo-year-old or three-year-old (Raced V Unraced test), the preference for racing on dirt or turf surfaces (Dirt v Turf test) and the chance of being an elite (Group) race performer (Elite Performance test).

In all cases these are predictions based on statistical over and underrepresentation and are not ‘diagnostic’ for a particular outcome.

There is no black and white answer for any racing performance trait, but genetic information can be used to improve the odds.

It is sometimes said that in order to win the Derby a horse needs a combination of speed and stamina.

Others say that there needs to be stamina on both sides of the pedigree because it's an extreme test. How do you see it and what is the ideal profile of the Derby winner if you could pick a model profile? For example do you prefer stamina on the dam side and speed on the stallion side - any refinements like some speed on the dam side?

Mark Sumner

CT and TT type horses are best suited to the Derby distance.

However, our data has not shown that the ‘speed gene’ type of the mare/stallion influences the potential for elite performance in the different distance ranges.

Read our Life in Lockdown Q&As with industry figures

Richard Venn: 'The French are in a good position to get back racing sooner'

Tim Kent: 'It's difficult to plan when we don't know when racing will resume'

Russell Ferris: 'Weatherbys had contingency plans that we activated at once'

Grant and Tom Pritchard-Gordon: 'Inglis Easter has kept us busy since January'

Peter Hockenhull: 'The social side of meeting and chatting to breeders is gone'

Polly Bonnor: 'We've fulfilled every feed order, including all our exports'

Richard Lancaster: 'We're fortunate that some Shadwell staff live on site'


Starting with the optimal genetic package will contribute to the chances of producing an elite racehorse
E.W. Terms