BioPet DNA Breed Identification Test - Frequently Asked Questions

Common Questions

  • 1. How was the test developed?
  • 2. How does the test actually work?
  • 3. How many breeds does your test recognize and how did you choose those breeds?
  • 4. How accurate is your test?
  • 5. I have read that the number of markers is critical for accurate breed identification, is that true?
  • 6. Is the size of my dog controlled by the same markers as the breed?
  • 7. Do you perform any quality control tests in your assay?
  • 8. Is blood the best sample for breed identification?
  • 9. Why is using a cheek swab better than using a blood sample?
  • 10. But a cheek swab is also easier to use right?
  • 11. So why would anyone use a blood sample rather than a cheek swab?
  • 12. I have been told that cheek swabs do not work and often fail. Is that true?

    Answers

    1. How was the test developed?
    Research into the Dog Genome has been ongoing for several years. It began with a hope that a better understanding of the dog genome would help the fight against human cancer as it had been shown that there were many similarities between canine and human cancers. Completion of the Dog Genome Project in 2005 has enabled scientists to discover segments of the DNA molecule that produce differences between various breeds of dogs, such as the shape of the tail, size or color. (reference: The Dog Genome: Survey Sequencing and Comparative Analysis Ewen F. Kirkness, Vineet Bafna, Aaron L. Halpern, Samuel Levy, Karin Remington, Douglas B. Rusch, Arthur L. Delcher, Mihai Pop, Wei Wang, Claire M. Fraser, and J. Craig Venter Science 26 September 2003 301: 1898-1903)     Return to top

    2. How does the test actually work?
    When a sample is received by the lab it is put into a solution that takes the cheek cells off the swab and releases the DNA. Once you have the DNA in solution, it is mixed with a number of chemicals and copied by a process called PCR (Polymerase Chain Reaction). This basically makes millions and millions of copies of the bits of the DNA that are needed for analysis, and we call those pieces of DNA “markers". Once we have the amplified DNA markers, we analyze the information from those markers and compare it to our reference database. The comparison database is really the key to the whole process. During the comparison process, the results for the DNA markers from your mixed breed dog is compared with DNA marker information from thousands and thousands of pure bred dogs. This matching process is performed on a computer through a sophisticated program that repeats the analysis thousands of times to assure an accurate result.     Return to top

    3. How many breeds does your test recognize and how did you choose those breeds?
    The BioPet Breed ID test can recognize 63 breeds. Those 63 breeds represent about 93% of the dog DNA that is in the US according to historical trends in breed popularity. The breeds that we detect are the most common breeds you see, such as Labrador Retriever, German Shepherd Dog, Yorkshire Terrier or Chihuahua. These common breeds are the ones most likely to be present in a mixed breed dog or in the increasingly common designer dogs. The rare breeds that are found in the US are unlikely to be present in a normal mixed breed, so breeds like Pharaoh Hound or Kuvasz or Wirehaired Pointing will not be seen in many mixed breed dogs. As the cost of performing the Breed ID test goes up as you add more breeds, we decided to include just the most common breeds to keep the price down. Why pay 50% more, when for most people the breeds that make up their dog are the common ones? For more than 99% of the samples we have received, we have been able to match the samples to these 63 breeds.     Return to top

    4. How accurate is your test?
    To state what our accuracy is, we need to know what the correct answer is, and for most mixed breed dogs nobody knows exactly which breeds are present. It is possible to quote a statistical value (called positive predictive value or PPV) for pure bred dogs (where we have a PPV>99%) or for first generation mixed breed dogs (where we have a PPV>95%), but as the vast majority of mixed breed dogs are neither pure bred nor first generation mixed breed, those figures are inappropriate. The figure that we can tell you with certainty is that less than 3% of owners complain about their results, and normally the problem has been caused by contamination of the sample during collection, transit or handling.     Return to top

    5. I have read that the number of markers is critical for accurate breed identification, is that true?
    That statement is almost correct. The number of markers is important, but it is not the only important factor. We also need to consider the amount of information that each marker will give us. For example, if you have a choice between using 10 markers that will only identify one breed each and one marker that will help to identify 15 breeds, it is obvious that you would choose the marker that helps to identify the 15. Just quoting the number of markers that are used does not really help you know the precision of the test. You must also know how much information each marker will give you, therefore basing a comparison on the number of markers used in the test is not valid. We can however tell you that in our test we compare well over 500 data points when performing the breed identification.     Return to top

    6. Is the size of my dog controlled by the same markers as the breed?
    One interesting genetic discovery about dogs is that there is a specific size gene for the dog. This is the reason why you can have dogs with an identical appearance except different size! The Poodle family (Toy, Miniature and Standard) illustrate this point perfectly. As an extreme example if you crossed a Great Dane and a Yorkshire Terrier it would be possible to have a dog that looks like a Great Dane but is only 6 inches tall. (reference; A Single IGF1 Allele Is a Major Determinant of Small Size in Dogs Nathan B. Sutter, Carlos D. Bustamante, Kevin Chase, Melissa M. Gray, Keyan Zhao, Lan Zhu, Badri Padhukasahasram, Eric Karlins, Sean Davis, Paul G. Jones, Pascale Quignon, Gary S. Johnson, Heidi G. Parker, Neale Fretwell, Dana S. Mosher, Dennis F. Lawler, Ebenezer Satyaraj, Magnus Nordborg, K. Gordon Lark, Robert K. Wayne, and Elaine A. Ostrander Science 6 April 2007 316: 112-115)     Return to top

    7. Do you perform any quality control tests in your assay?
    Having an effective quality control system is essential for a lab to process large numbers of DNA samples correctly. To prove that our test continues to perform properly, we run a number of control dogs every day. These are dogs that we know and have access to on a regular basis so we know what their results should be. By running a sample from these dogs alongside every group of customer dogs we are able to check that our processes are working properly. If our control dogs fail or give results that were different from previous runs, we retest all the customer samples that were processed at that same time.     Return to top

    8. Is blood the best sample for breed identification?
    No, not necessarily. It has been suggested that a blood sample is the “Gold Standard” for genotyping applications. That is simply not true. For every critical application where DNA is required it is typically a cheek swab that is used. For example, if you are watching CSI on TV, the investigators take a cheek swab to match a suspects DNA to evidence found at a crime scene. Globally, DNA databases used for matching DNA have been built by using cheek swabs. Blood samples are simply not the only, nor the preferred, way to collect DNA samples for forensics labs, the FBI or any other organization that routinely collects DNA. In human paternity testing more than 95% of the labs who perform those tests use cheek cells collected on swabs. If you look in the scientific literature, more than 85% of papers published over the past 10 years involving genotyping (what we do in the breed identification process) use a cheek swab. If the use of a cheek swab is the best method for collecting samples that will be used to determine murder trials or prove who a child’s parents are, then it is also the best method that we can use for our pets.     Return to top

    9. Why is using a cheek swab better than using a blood sample?
    The use of a cheek swab has many advantages over the use of a blood sample. The DNA that is present in cells is the same regardless of the source of that cell, be it blood or cheek. It does not matter where in the body those cells actually come from, the DNA that is in them is the same. The main advantage that you have with cheek cells is that the sample is easy to collect and relatively clean. When you collect epithelial cells from the lining of the cheek there is not much present on the swab other than cheek cells and a bit of saliva. A blood sample it is a very different proposition. The most common cell type in blood, red cells, does not contain any DNA. Even worse, those very common red cells contain hemoglobin which will interfere with a critical reaction in the breed ID process. Therefore the red cells must all be removed by a time consuming and expensive process. When we look at a cheek swab, the DNA that we obtain is easy to get to, relatively pure, and doesn’t require a labor intensive purification process like blood.     Return to top

    10. But a cheek swab is also easier to use right?
    Yes it is. Almost anyone can successfully take a cheek swab from their dog and send it in. It is quick, easy and, most importantly, stress and pain free for the dog. Compare that with a blood sample, where you must take the time to drive your dog to the V place (we do not want to say Veterinarian in case your dog gets scared and runs off). When you get there, your now stressed dog must be taken in by a technician, have a needle stuck into a vain and blood drawn. What is worse, the vet will normally charge you for the privilege of sticking them with a needle!     Return to top

    11. So why would anyone use a blood sample rather than a cheek swab?
    We cannot think of any good reason which is why we use a cheek swab! We at BioPet have been involved in collecting DNA samples for analysis for more than 20 years. There is no valid technical reason why you would want to use a blood sample over a cheek swab for this application.     Return to top

    12. I have been told that cheek swabs do not work and often fail. Is that true?
    Not in our case. If we look at the samples we have received here at BioPet over the past year, about 90% of the swabs work first time. If we look at using 2 swabs, 98% of the samples give sufficient DNA to work. That is why at BioPet we include 2 swabs in each collection kit giving a better chance of getting a successful sample then others have reported with blood!     Return to top