How the Wisdom Panel DNA Test Really Works

Not a Dalmatian! Really!

Not a Dalmatian! Really!

If you're unsure of your dog's ancestry, you've probably thought about DNA-testing your dog. Most people are concerned about how accurate they can really be; the answer is...somewhat? A lot depends on your dog. I'm here to tell you about my experience testing my dog's DNA with the Wisdom Panel 2.0 kit and why, for me, it was totally worth the money.

Let me start by saying that DNA-testing is the only reliable way of discerning your dog's ancestry--read my post, My Dog Is Not a Dalmatian: Why Breed (Mis)Identification Matters, to learn why even a dog expert most likely can't tell your dog's breed composition based on his appearance and behavior.

Is it worth $70-80 to find out what breed your dog is? Maybe. You might want to know, for example, if your dog has any German Shepherd in him, as many German Shepherds are predisposed to hip dysplasia. Not that you can prevent it, but at least you can be better prepared by administering supplements and limiting his weight gain. It was important to me to find out if my dog, Chance, was really a Dalmatian, because all purebred Dalmatians have a genetic mutation that makes them prone to developing urinary stones--and if he had any Dalmatian in his ancestry, it would be possible for him to carry this gene, too.

 
 

The most recent Wisdom Panel DNA versions (2.5 and 3.0) test not only for breeds but also for a specific mutation in the MDR1 gene that is common to "herding breeds, some sighthound breeds, and many mixed breed dogs." According to the Wisdom Panel website, this mutation "makes them defective in their ability to limit the absorption and distribution of many drugs... As a result, dogs with the MDR1-mutation may have severe adverse reactions to some common drugs." Testing for this mutation specifically could be a life-saver for your dog.

Scientific research has proven, however, that a dog's breed does not necessarily predict their behavior, so it's unlikely that a DNA-test will provide you any insight into how to train your dog, or predict what sports he might be good at. Those decisions should all be based on the individual dog's personality and aptitude.

Several articles have been written about the various tests on the market and their accuracy--two good recent ones are Colleen Kane's "What you should know before you test your dog's DNA" for Fortune and Gemma Tarlach's "Can Doggie DNA Tests Decode Your Mutt's Makeup?" for The Crux, a Discover Magazine blog.

My personal experience is with the Wisdom Panel test; when I did it, the latest version was 2.0. Now, there are two more recent at-home kits available, as well as a veterinarian-administered blood-test version and a test designed specifically for dog breeders. Judging by this product comparison chart on the website, the newer tests offer 50+ more breeds in the database than when I did it, as well as the MDR1 mutation test and wolf/coyote detection! I am not gonna lie--I kinda want to redo the test for those last two reasons...

Finding Chance's Roots: Round 1

A real Dalmatian.

A real Dalmatian.

I really wanted to test Chance because I wanted to know if he was actually a Dalmatian. Dalmatians are prone to a very serious health problem--no purebred Dalmatian carries the gene for normal uric acid excretion, which makes many Dalmatians likely to develop urinary stones. So I went slightly nuts trying to avoid feeding Chance anything that contained organ meats, which can exacerbate the condition.

However, Chance also developed crazy allergies. Which further limited the foods I thought I could feed him. I was spending hours reading ingredient lists online.

So finally I decided to DNA-test, settling the issue once and for all. (I have since learned that doing this test alone does not, in fact, settle the issue. More on that in a bit.)

I used Wisdom Panel 2.0, the latest Wisdom Panel product at the time. I swabbed his check, registered his sample on the website, and popped the kit in the mail. It was all super-easy.

About a month later, I got an email notifying me that Chance's results were ready!

I opened my PDF report and...

Was immediately crushed.

patchwork dogs dog blog new dog guide carrie wiita

WHAT?! No he's not!

I frantically started scrolling. I next came across his "family tree:"

patchwork dogs dog blog new dog guide carrie wiita

See the little asterisks next to each of the identified dog breeds? See the fine print at the bottom?? "Breed detected, however at a lower confidence. Such results are not included in accuracy calculations."

WHAT THE....???? WHAT IS THAT SUPPOSED TO MEAN???

I was confused. And felt cheated. I felt like I'd wasted $80.

Seeing as MOST of Chance's "family tree" was made up of "mixed breed" dog-tags, I was indeed "interested in the Mixed Breed ancestry," as the bottom of the page chirped quizzically at me. So I kept scrolling, past the breed info pages on Great Danes, Shetland Sheepdogs, and Long-Haired Dachshunds (none of which remotely resembled my dog), until I landed on this page:

patchwork dogs dog blog new dog guide carrie wiita

A-HA!!!!! THERE WAS THE DALMATIAN! At the very bottom. But I was confused by the text at the top. One or more of these breeds could be in Chance's background--or none of them could be. It's not likely that ALL would be present.

So what was the answer to my question? Does Chance have any Dalmatian in him??

Maybe. Very definitely there is a tiny chance that MAYBE.

That was the absolute worst non-answer I've ever gotten in my life. The only good thing about this stupid test so far was that first dog listed at top of mixed-breed-maybe's list. I had never heard of a Catahoula Leopard Dog, so I Google-image searched it:

THAT'S MY DOG. Yes my dog was all white with black spots, but as far as the body type and face shape and ears??? TOTALLY MY DOG.

There was only one last piece of information the test had to give me--Chance's predicted adult weight. Chance had been already examined by a veterinarian who determined him to be about 7 months old and "probably close" to his adult weight. Her EXACT WORDS were, "yeah, he'll probably stay around this size," which was 35 pounds.

One last scroll, and I was faced with this sentence: "The adult weight is predicted to be between 54-82 pounds."

Nope. NOPE. OBVIOUSLY WRONG. They were getting this weight because the test had erroneously assigned a Great Dane to Chance's ancestry somewhere. So I didn't even get to know how big he was going to be because the test had gotten it wrong from the start.

Faced with all this inconclusivity, I did what any frustrated consumer would do...

I wrote a strongly-worded email. (OK it felt strongly worded, but I just dug it up and it's actually quite reasonable in tone.)

 

If At First You Don't Succeed: Round 2

A Catahoula Leopard Dog.

A Catahoula Leopard Dog.

In my email, I explained that my dog looked nothing like the three breeds the test had found, that if anything he maaaaybe looked like the Catahoula Leopard Dog listed on the Mixed-Breeds Ancestry page, and "I'm sure this has something to do with the Great Dane detected, but at 7 months Chance weighs 35 lbs. There is no way he will hit 50 lbs, let alone 80 lbs" (actual quote from the actual email). I also included two photos so they could understand my deep outrage. And then I asked how I could go about getting a do-over.

Not long after, Wisdom Panel customer support responded. They thanked me for sending the photos: "Reviewing the photo, in combination with the results, will help [our review team] understand whether a computer algorithm-generated breed signature mismatch may have occurred or it will allow them to point out the various traits of the breeds identified in your report."

So, either the computer messed up, or the "review team" was going to tell me how my dog looked like a Great Dane/Shetland Sheepdog/Long-Haired Dachschund.

A week later, I received another email, this one with the subject line "Escalated Case #xxxxx." They notified me that they were escalating my concerns to "the science team" for "review," and that "the science team will be doing a manual re-analysis of the raw DNA data and will reach out to you to discuss the findings. The escalation process takes approximately 21 business days."

Another A-HA!!! So it was a "computer algorithm-generated breed signature mismatch" after all!!

In the ensuing three weeks, I sorta forgot about the whole thing. (And Chance kept putting on weight.)

Then one day a very nice vet tech called. She let me know that she was with Wisdom Panel and that she'd like to go over my results with me.

The conversation that followed was absolutely 100% worth $80,even though the upshot of the conversation was that she was very sorry, but she could not really tell me what breeds were in my dog.

Seriously.

Let me explain.

 

Wherein Carrie Learned More About Dog Genetics Than She Ever Thought Possible

One of Chance's ancestors??

One of Chance's ancestors??

The tech's first question was, "I just need to ask--where did you get your dog???"

I explained that I adopted Chance in Los Angeles, but that the rescue had found him in Tijuana, Mexico.

"Ahhhhh!" she exclaimed, relieved. "That makes SO much sense."

Currently, the Wisdom Panel website does caution that "if your dog was imported from a country other than Canada, the UK, Australia, Germany, or mainland U.S., or you suspect that your dog’s ancestors are from outside these countries, his breed ancestry may not be well represented in our database."

However, my vet tech explained that, in Chance's case, it wasn't so much that Wisdom Panel couldn't identify some exotic Mexican breeds in Chance's background. Rather, because there is a very low spay/neuter rate in Mexico, the genetic pool of the Mexican dog population is hugely muddied. It is very, very hard to find a purebred--which is how Wisdom Panel is able to identify ancestral breeds in a given dog. In the US, on the other hand, spay/neuter programs and the persevering popularity of purebred dogs has led to a "clearer" gene pool--meaning that US dogs generally have at least some purebred ancestors in the past three generations. Mexican dogs? Notsomuch.

"The fact is," she explained, "because it's been so long since there's been a purebred in Chance's family tree, it's really hard for the computer to accurately identify any breeds."

Now this is the part where I really wish I'd recorded this conversation--I did start scribbling notes, but I've lost them somewhere along the way. She told me, essentially, that when a sample is first analyzed, the computer goes through and is able to identify the presence of four main genetic groups (the Wisdom Panel website now lists 11 genetic groups, so this may be another upgrade). Chance's results indicated the presence of three of the groups (I've forgotten which ones), and the absence of the "Ancient" group. So there were no Tibetan Mastiffs in his lineage, at least.

And this is where my science and memory get shaky. I believe she said that, once the computer determines which groups are present, it then more or less "guesses" what breeds are showing up from each group, based on statistical probability. In Chance's case, the tech said, the computer determined (based on genetic markers, I guess??) that there was a statistical probability that there was Great Dane in Chance's background--but just like there's a statistical probability that if you flip a coin twice one will be heads and one will be tails, it's totally possiblethat BOTH flips will be heads and NEITHER will be tails.

She said that usually, the computer is pretty accurate. But they acknowledge that sometimes it's just not. That's where people come in.

With subjective human analysis, a tech is able to look at a dog's results, compare that to pictures of the dog, and make a better assessment of what is actually going on in a given dog's background. The computer can look at Chance's sample and say "probably Great Dane," but a tech can look at Chance and say "ehhhh...probably not."

She agreed that Chance looked more Catahoula Leopard Dog than anything, but that didn't mean that he IS COMPRISED OF more Catahoula Leopard Dog than anything. The genetic lottery just happened to favor traits that look more Catahoula Leopard Dog-y. He could be equal parts of fifty different breeds.

I was fascinated, but also sorta bummed. This was all really cool--many dogs will get incredibly accurate DNA test results!! Just...not my dog.

"OK, so..." I said, disappointment evident in my voice, "there's no way you can tell me if he's Dalmatian or not? I was really worried about that mutation that gives them urinary stones, and I've been trying to avoid feeding him organ meat..."

"Well..." she said, hesitantly. "I really doubt he's Dalmatian. As for the mutation... We're not really supposed to do this, but I can check and see if he has that genetic mutation for you."

And now you see why I haven't mentioned my heroic vet tech's name. But Wisdom Panel, if you're listening, HONESTLY THIS WOMAN DESERVES A DAMN RAISE BECAUSE SHE TOTALLY SAVED WHAT OTHERWISE WOULD HAVE BEEN A SUPER-BAD CUSTOMER SERVICE EXPERIENCE.

We also then talked about Chance's coat color (he's actually a BLACK dog with tan points and extreme white spotting), and whether he might be prone to the deafness that frequently affects majority-white dogs (she thought not, because of the amount of pigmentation on his ears). Seriously, it was one of the coolest dog experiences I've ever had.

She contacted me a few days later to let me know that she'd checked for the genetic mutation, and Chance was in the clear! It was such a relief.

 

Conclusion: Probably Worth It

My purebred Tijuana Street Dog.

My purebred Tijuana Street Dog.

For me, the Wisdom Panel test was helpful in pointing me in the direction of the breed that my dog most closely resembles, and was sort of helpful in telling me that though he looks a lot like a Catahoula Leopard Dog, he's nowhere close to being a purebred.

The true value for me was in finding out that I don't need to worry about that specific mutation--a service which is not technically supposed to be included. If the vet tech hadn't checked on that genetic mutation, I don't know if I would have felt like my $80 was well-spent.

Now that the test has been upgraded, I think it's probably worth it for the MDR1 test alone. And when it comes to breeds, you'll most likely have one of two outcomes. EITHER your dog is like mine, and you'll find out that he's so mixed that he's practically his own breed (how I like to look at it), OR you'll actually get a fairly accurate breakdown of what breeds make up your mutt.

My advice to you is that if you feel like the results aren't quite right, or you have any questions about the results, DO NOT BE SHY about contacting Wisdom Panel. It seems like they expect some customers to be totally satisfied with their first round of results, but they're definitely prepared to more closely look at the results for all those customers NOT satisfied.

Unless you really don't care at all to know what breeds comprise your precious pooch--and I actually think that's quite commendable--I definitely recommend the Wisdom Panel DNA-test!

Oh, and one last note, to reward you for making it this far. Chance is now two years old, and tips the scales at 62 pounds--exactly where Wisdom Panel said he'd end up. Even without any Great Dane.

 

Have you DNA-tested your dog? Tell us about your experience in the comments section below, or join the discussion over on the PATCHWORK DOGS Facebook page (and share a photo of your patchwork dog)!


patchwork dogs dog blog new dog guide carrie wiita

ABOUT THE AUTHOR

Carrie Wiita is an actorbloggercheap-wine enthusiast, and dog-parent to Chance, a two-year-old rescue. Secretly she wishes Chance does have some Great Dane and he grows to the size of a small pony BECAUSE MORE TO LOVE.