DNA Testing


Which are better, DNA tests or more traditional screening tests?

It would depend on the disease, but for a breeder’s purposes DNA tests are the “gold standard” because they tell us exactly which gene versions the dog has.  Other forms of testing only measure the phenotype, what can be observed or measured.  For things with single gene inheritance, like PRA, knowing that a dog has it from an eye exam tells us it has two copies of the recessive version of that gene.  But if the dog gets a clear eye exam we don’t know if the absence of PRA is because the dog has two normal copies, one normal and one mutant (a carrier) or has two copies but hasn’t developed the disease yet because it is a disease of middle-aged dogs.  The important information from screening exams is who DIDN’T pass.  Unfortunately, this information often isn’t available in the US though such results are more readily available in many European countries.

Which DNA and screening tests are really necessary?

Breed health screening programs, which should include applicable DNA tests, should be developed based on the issues most frequently seen in a breed as well as some less frequent but high heath impact conditions.  Given the ease of moving dogs around the world today, a breed’s world population should be considered a single entity.  (Especially so for Aussies, which were first exported fairly recently.)  If a sub-population – working lines vs. show lines, North American vs. European) or even a particular line or family – is known to have something that the balance of the breed does not, screening might be limited to that sub-population provided any of those dogs moved to different places or bred outside the line are screened or had a DNA test, if available.  For health issues without a DNA test the dog should also have no recent family history of that disease or condition.

ASHGI has developed a Testing & Screening Protocol for Aussies which can be a handy reference.  However, any dog offered at public stud should have all the screening tests and every DNA test currently available for the breed with the exception of those for which both if his parents tested clear.  A frequently used sire is the quickest way to make a rare genetic issue a common one.

What DNA tests should be mandatory?

  1. MDR1 for every Aussie.
  2. HSF4 cataract and CEA for breeding Aussies.
  3. Stud dogs offered at public stud should have all DNA health tests currently available for the breed.
  4. Other test if there is a relative within two steps of relationship that has been diagnosed with the disease or has been tested and found to have one or two copies of the mutation for a disease.

That said, many labs now offer reasonably priced test panels for Aussies.  If they fit your budget, use them.

What kind of sample should be submitted for a DNA test?

Because DNA tests are typically offered by commercial laboratories their operation is designed for rapid processing of numerous samples to keep prices down.  The lab will specify a sample type or types which it can best process for the test (s) you want.  Usually this will be a cheek swab and they will provide a kit and instructions.  Sometimes it may be whole blood or some other sample but it is important that you only send what the lab lists as acceptable and follow directions carefully.  Mishandled or contaminated samples due to customer error will result in rejection of the sample or, if the lab does not detect the problem, inaccurate results.

Could I buy a supply of cheek swabs, use a bunch on each dog, and then keep them or later use?

Depending on how long you stored the swabs and whether or not you froze them, they may or may not be usable.  After a week at room temperature the enzymes in the saliva can start degrading the DNA.  Two weeks at room temperature is about the max for processing a useful sample.  Freezing might extend the useful life of the sample, but is not recommended.  In addition, some labs want you to use swabs which they supply.  The best thing to do is swab a dog only when you can get the samples in the mail soon after.

How fast can scientists make DNA tests?

It isn’t as simple as do-study/make-test.  Once a gene is identified it doesn’t take long to set up a testing program but it can take a very long time to find a gene.  At this point it is very likely that most of the simple (one gene) inherited diseases already have tests.  From here on out, science will be trying to tease apart things that involve multiple genes, gene regulatory factors, and environmental influences.  These complex puzzles will take longer to solve or may not be something they can make a commercial test for.

If I test a puppy and it is clear, do I really need to test my bitch or the other pups?

Yes, because one puppy’s result will only tell you a little bit about the genotypes of the parents and its littermates.  At this point the DNA tests regularly used by dog owners are for genes with only two versions so you have three possible combinations of those versions:  Two normal copies of the gene, one normal and one mutation, or two mutated copies.  Your puppy will have inherited one of its copies from each parent but you can’t know which copy came from which parent.  If both copies are the same, that is one thing, but if they are different you don’t know which parent gave the mutated version without testing them.  The littermates may have the same result as the puppy you test, but only if both parents have two copies of the same gene version.  If either or both have two different versions you will get different combinations in the pups and can’t tell who got what without testing.

The best procedure is to make sure your bitch is tested and that any stud you breed her to is tested.  Then you will know what might occur in the pups.  The only time you do not need to test pups is if both parents have tested clear for the mutated version of the gene.

 Why isn’t there a push to eliminate the bad genes found with DNA tests?

In theory it is possible to completely eliminate an unwanted gene mutation in a single generation by ceasing to breed all individuals that carry it.  But this would require 100% compliance with the testing of all breeding animals, which isn’t likely to happen.  However, concerned individuals could make it a litmus test for their own programs.

BUT…..doing so can have unwanted side effects:

  • If the mutated version of the gene is very common (MDR1 or the HSF4 cataract mutation in Aussies) and all dogs that have it are eliminated, you will eliminate a major portion of the breed overnight. The remaining dogs may or may not have the traits you want and will certainly have genes for a few things that you don’t want.  Those things might prove to be far worse than the target of the purge and likely won’t have a handy DNA to help identify them.
  • If the condition is of minor health impact, as with yellow coat color, the zealous removal of all carrier individuals could eliminate otherwise valuable dogs.
  • If the condition is polygenic, the target gene is only part of the picture. This is true for risk factor genes like the degenerative myelopathy (DM) mutation in Aussies.  Eliminating all dogs that have it might reduce the (rare) incidence of this admittedly nasty but you would also be eliminating significant  numbers of dogs who would never have developed an active xase.

The goal is not to produce affected puppies. DNA tests enable a breeder to do that, or at least considerably reduce the risk of doing so.  Carrying a defective gene version should be considered a fault and preference should be given to offspring that do not themselves have the mutation.  Over multiple generations one could phase out use of dogs with the mutation, thereby eliminating the unwanted gene while preserving good qualities and genetic diversity.  How long this would take would depend on the frequency of the gene in the breed population.  It will take a number of generations to work MDR1 out of Aussies because half of them have it at present (2013).  But the HSF4 test, which was released in 2007, almost certainly played an important role in reducing the number of cataracts diagnosed by half a decade later.

If we would equate genetic health issues to faults and treat them as such when making mating selections we’d do much better in reducing their frequency.  Some health issues are severe faults (epilepsy, cancer) and others are minor (MDR1 or CEA) and the things with greater impact on health and quality of life should be prioritized.   All other things being equal, preference should be given to the dog that does not have or carry this “fault.”  Another important thing to remember:  Every dog has some bad genes and we only have tests for a few.  You are several steps ahead of the game on those you can test for.

How can a dog have a clear DNA test and still get the disease?    

Depending on which test it was, any of several things may have happened.  Some diseases (progressive retinal atrophy, neuronal ceroid lipofuscinosis) have several different genetic forms.  If you used the test for one and the dog has another it could appear that the test didn’t work when it actually did, but with a different form of the disease.

If the diagnosis was based on exam findings, as it is with most eye diseases, it is possible that the exam diagnosis is incorrect.  CEA and PRA can both be misdiagnosed upon occasion.  In this case, it is the examiner who is in error.

Finally, if the DNA sample was not handled properly (i.e. mixed up with that of another dog) you might have an incorrect test result because what you got does not belong to your dog.  This is far more likely to happen when the sample is collected and packaged for shipping than at the lab.

Can DNA be used to confirm a diagnosis?  

Yes.  If there is no definitive diagnostic test or diagnosis is based on observation, a DNA test can be used to confirm.  Among tests available for Aussies that this can be done with are any of the available eye disease tests and the tests for hereditary cobalamin malabsorbtion, and degenerative myelopathy test.

What DNA tests are available for Aussies?

Currently (2019) Aussies can be tested for the tests listed below.  For details on how to interpret test results go to the FAQ secion for that disease.

Canine Multifocal Retinopathy (CMR1 or MR1) a form of retinal dysplasia
Coat Color – black/liver
Coat Color – dilute
Coat Color – merle
Coat Color – yellow
Collie Eye Anomaly (CEA)
Cone Degeneration (CD)
Degenerative Myelopathy (DM)
Hemophilia (both types)
Hereditary Cobalamin Malabsorbtion (HCM)
Hyperuricosuria (urate bladder stones)
Multi-Drug Resistance 1 (MDR1)
Neuronal Ceroid Lipofuscinosis, types 6 and 8 (NCL)
Progressive Retinal Atrophy (prce form)
Von Willebrand’s Disease (VWD)

Why don’t they do double-blind studies before they make DNA tests available to the public?

Double-blind studies are not applicable to DNA tests.
The purpose of a double-blind study is to remove bias from the testing of drugs and some other medical treatments.  Some subjects in a double-blind study will get the treatment while others receive a placebo.  Those receiving the placebo are termed “controls.”  This helps control for the “placebo effect” that can occur when people think they (or their dogs!) are receiving something that will help them.  In a double-blind study, neither the patient nor the person administering the treatment knows what the patient is receiving, hence the name “double-blind.”

Animals aren’t prone to the placebo effect and if owner feedback is not part of the data a single blind approach – where the person administering the treatment and/or recording results – doesn’t know if it’s real or placebo.  This will minimize the effects of perceptual errors made by the people dealing with the animals.  If you really like an individual animal, you might want it to get better so might make biased assumptions about what you were observing.

Since DNA tests are used to determine the presence of gene versions already identified in other research and sometimes to confirm a tentative diagnosis already made by the treating veterinarian, the placebo effect isn’t an issue.  Therefore there is no need to do a double-blind study on them before they are made commercially available.  However, current DNA research does make use of “controls.”  Genome searches will be done on a set of dogs that have the disease under study and on another set of unrelated dogs who do not (the controls) in order to emphasize genetic differences between the two groups.