BRAHMAN NEWS JUNE 2008 Issue #159
Of all the inherited characteristics in cattle, only a few are controlled by a single gene sequence, for example, some diseases such as Pompe’s disease, and simple traits such as horn/poll. The expression of single-gene traits is independent of environment.
Most of the commercially important production traits, such as growth and carcase traits, are influenced by many genes. These traits are known as mutli-gene (or poly-genic) traits. The animal’s phenotype (what we see or measure) is the sum expression of the multiple genes, and is influenced by the production environment in which the animal is run, eg; level of nutrition.
Breedplan EBVs are calculated from phenotypic differences between individual animals, and reflect the sum effect of all the genes that influence the trait in a given environment.
For many traits of interest to cattle breeders, phenotypic measurements are only available when the animal is yearling age or older. Some traits, such as marbling, cannot be measured directly in the live animal so we measure another trait that is genetically related to the target trait, such as % intramuscular fat as measured by ultra-sound.
Gene markers identify the chromosome location or proximity of individual genes that influence a trait. In isolation, they may only account for a small amount of the phenotypic variation expressed for the trait.
A significant advantage of gene markers is that they can identify genetic merit for traits that are hard to measure on the live animal, such as marbling; they can be measured early in life, eg; at birth; and they are not reliant on phenotypic expression for measurement.
Gene marker information can be included with phenotypic information to calculate more highly accurate Marker-Assisted EBVs, or by providing marker-only EBVs for difficult to measure traits such as NFI where phenotypic measurement is difficult or costly.
Single-gene trait markers - Markers for single gene traits such as diseases, red/black coat colour (Angus) mysostatin (enhanced muscling) in Angus and Limousins, and horned/polled (if/when available) are valuable selection/culling tools and can easily be incorporated into a selection program.
Multi-gene trait markers - Development of multi-gene trait markers to date has been based on the assumption that these traits are influenced by a small number of genes that have a large effect on expression of the trait, together with a large number of genes that collectively only have a small effect.
Current gene marker technology has been directed towards finding those genes of large effect, however, with the release of the bovine genome in 2006, the “few genes of major effect” theory appears to be an oversimplified explanation of the biology of multi-gene traits.
Firstly, with the exception of Tenderness markers, which appear to code for specific enzymes, the “large effect” genes haven’t eventuated and most traits need a large number of markers to account for any significant phenotypic effect.
Secondly, it appears that marker effects may be specific to a population (breed or genotype) and/or production system, and it cannot be assumed that the effect will be consistent when changing to different populations and/or production systems.
At present, 4 markers each for Marbling, Net Feed Intake (NFI) and Tenderness are being marketed in Australia by Catapult Genetics. Favourable copies of each marker are reported using a star system, with a maximum of 2 stars per marker (one for each allele of the target gene). For example, with the GeneStar Marbling test using four markers, an animal may have up to 8 stars – the greater number of stars, the greater number of favourable markers identified for that trait.
To be of value as a selection tool, breeders need to know what difference in trait performance is explained by the presence or absence of the markers - does their presence or absence influence phenotypic performance, and by how much?
Whilst the size of the marker effect is an important aspect in deciding whether or not a gene marker is of value, so is the frequency of the marker in the population – if frequency of the favourable marker is extremely high or extremely low in a population, then opportunity to select better animals is very limited.
For an example, a marker with a frequency of 98% means that only 2% of the population is available for alternate selection. Mid-range marker frequency allows more selection scope than extreme range marker frequency.
In the case of extreme frequencies, the cost-benefit of finding those few animals needs to be considered.
Thus, both marker effect and marker frequency is important information when deciding to invest in marker technology. A breeder needs to know both sides of the equation before making an investment decision.
Validation of the effect of markers have generally been conducted by the discoverer or commercializer where the results may not be available for public or scientific scrutiny.
Whilst it would be desirable for all marker effects to be validated against independent reference populations, the high cost of doing so for a large number of traits makes this unlikely.
Industry will most likely need to rely on validation based on “proof of effect” from the analysis of phenotype information being recorded on marker identified animals, (as few phenotypes will be routinely collected for difficult-to-measure traits such as Tenderness or NFI, validation will need to be planned in purposely set up and recorded research herds).
The discovery and commercialization of gene markers is ongoing, and it can be expected that commercialisers will be releasing an increased number of markers over the next 12 months – additional markers for traits that currently have markers available, and markers for a range of new traits that have not previously had markers available.
Because of the very small effect of most individual markers (requiring a large number of markers required to account for a reasonable effect), marker ratings, such as stars, will become increasingly clumsy to use and interpret. The best use of markers is to have them incorporated into the relevant EBV, if one is available.
Where a significant marker effect is found and validated, this information can be included in the calculation of Breedplan EBVs using procedures currently being developed by the AGBU.
These EBVs will be known as Marker Assisted EBVs (MA-EBV). The first MA- EBV, for Tenderness, is expected to be released by mid-year. For selection purposes, the fact they are marker assisted doesn’t affect how they are used – they should be used the same as conventional EBVs, but they will have a higher accuracy at a younger age.
Markers with unknown or no significant effect will not influence the calculation of marker-assisted EBVs, as they would get zero estimated effect in the prediction equation.
Well researched and validated gene marker information can significantly increase the accuracy and prediction value of Breedplan EBVs by the calculation of higher accuracy Marker Assisted EBVs. It should not be viewed as an alternate technology.
If validated markers are available for traits of interest, and you are satisfied as to the cost-benefit of investing in those markers, then they are a useful add-on to your performance recording program.
Investment in gene markers that have not been independently validated and published as to their effect on phenotypic performance and frequency within a population must be considered a risk proposition.
If not convinced that currently available DNA marker technology offers a cost-benefit to your herd’s breeding program, it is a good idea to adopt a “wait, but be prepared” policy of collecting and storing tail hairs of important animals in the pedigree (eg. sires).
Those hair samples (which must been clean, dry and have the follicle attached) can then be tested at some future time when the effectiveness and cost-benefit of the tests are validated to your satisfaction.
When buying bulls, the primary genetic selection tool should be Breedplan EBVs. Any phenotypic benefit derived from the marker will be reflected in the EBV.
Marker Assisted EBVs incorporating DNA marker information will result in EBVs of higher accuracy.
Marker ratings, eg star ratings, may or may not be useful information, depending upon:
A - the relative value of a target trait to your production and marketing goals. For example, marbling has a high relative value for the Jap B3 market, but no relative value for the EU market. Similarly, the Tenderness genes are of importance to the Bos indicus breeds but less so to British breeds.
This information allows you to put some $$ significance to increased trait performance.
B - the effect that presence/absence of the marker will have on the phenotypic performance of your stock, under your production environment. For example, what increase in performance can you expect for each copy of a favourable marker? A quick A x B calculation brings the cost-benefit into focus.
C - the frequency of the marker in the breed/population your working with. This indicates your chance of improving trait performance user marker selection. For example, if the frequency is extremely high or extremely low, the majority of animals will have similar marker profiles and selection opportunity is limited.
Bull buyers need to know the answers to the above questions before they can reasonably justify paying more for bulls on the basis of marker ratings.
Remember, DNA tests enhance EBVs, not replace them.
Where available, marker assisted EBVs are the most effective method of presenting marker information to clients in a sale catalogue.
Providing marker ratings, eg. stars, of validated markers as an adjunct to MA-EBVs may be useful to reinforce the breadth of your performance recording strategy. Validated information on the size of marker effect should be provided as a footnote, along with current breed EBV averages.
Providing marker ratings on unvalidated markers, or markers known to have no significant effect is uninformative and will be misleading to clients.