Originally Posted by pstinard  Misty asked me some questions about the meaning of qualitative vs. quantitative when talking about DNA analysis of dog foods, so I thought I would expand on it a little.
Qualitative results are results that are not measured, or are somewhat subjective. Examples include: "The painting has some blue in it." "The stone is smooth." "The people look happy." "The dog food has some soybean DNA in it."
Quantitative results are measurements: "The board is 2 inches X 4 inches X 10 feet." "Bella weighs seven pounds." "The dog food contains 5% soybean protein, by weight."
Back in the 1980's, when I was doing DNA analysis and cloning of genes from corn plants, we had to extract DNA from a lot of plant samples. A plant cell is basically a tiny water balloon filled with carefully separated DNA, proteins, enzymes, carbohydrates, fats, etc. Once the cell bursts, everything mixes together. The enzymes that degrade DNA (called nucleases) are no longer separated from the DNA, and the DNA breaks down rapidly at room temperature. When we did DNA extractions, we had to take fresh samples, quick freeze them in liquid nitrogen, grind them while they were still frozen solid, add buffers that inactivate the nucleases, and do a series of procedures to purify the DNA. Everything had to be kept on ice until the nucleases were inactivated or the DNA would be hopelessly degraded and useless. And once we had extracted the DNA, we had to handle it carefully, because stirring the DNA sample would shatter the DNA further, kind of like breaking up dried ramen noodles. I used to joke that if you even looked cross-eyed at the DNA, it would degrade.
Flash forward to modern DNA technology, where you can extract tiny quantities of DNA from 1,000 year old bones, for example. Most of the DNA is degraded and useless, but through a technique called PCR, small pieces of intact DNA can be amplified and analyzed. That is much the same situation with doing DNA analysis of dog food. Once the meats, grains, soybeans, etc. are ground up, the nucleases are breaking down the DNA. PCR can detect small pieces of DNA that are still intact, but if the ground meat has been sitting around too long (think of a package of hamburger sitting around at room temperature all day), it might not have ANY useful DNA in it at all.
That's why I was saying that the results of DNA analysis of dog food are qualitative, not quantitative, even if the results are presented as percentages. You can sometimes detect DNA from a source that is not on the dog food label, but you might not always be able to detect it. In other words, they can tell you that a venison limited ingredient dog food has some soybean DNA in it that shouldn't be there, but they can't tell you how much, and they can't always tell you that it's there, even if some soybeans slipped into the batch--the soybean DNA might have degraded so much that they couldn't detect it. To state this more elegantly, "The absence of evidence is not the evidence of absence."
So yes, all of the research studies that found DNA in dog food samples that shouldn't be there should be of concern to people who own dogs with food allergies to those contaminating ingredients. We don't know how much contamination is there, and we can't be certain that it will cause an allergic reaction, but it's a very real possibility. And just because a test result came back showing NO contamination doesn't mean that there is no contamination--it just means that none was detected in those particular samples using those particular testing methods.
The best way to do an elimination diet is to make it yourself in consultation with a veterinary nutritionist. The second best is to choose a commercial company with a good track record and excellent quality control. |