A Tale of Two Brothers

PUBLISHED ON November 12, 2019 by jessica hekman

Recently the Darwin’s Ark team was privileged to sequence two internet celebrity dogs, Mr. Woo and his “brother” TWooie. Woo’s adventures have been documented in living color for years by his owner, affectionately known as the Food Lady, at her photo blog Wootube. Several years after the Food Lady acquired Mr. Woo, she encountered a dog of the same age, from the same region, who looked and acted so much like him that she thought they were very likely to be related. She brought the dog home and named him TWooie. The two are now inseparable and known as the WooTWoo.

The WooTWoo

(Twooie is on the left, and Wootie is on the right.)

We were very pleased to get the chance to weigh in on these two – brothers genetically or just in spirit? And what breeds did go into those pudgy balls of cuteness? Surely some Australian Shepherd, maybe some Shetland Sheepdog?

Our results for Mr. Woo:

Mr. Woo's results

The important results here are that he’s around 16.6% Golden Retriever, 16% Alaskan Malamute, 13.7% Shetland Sheepdog, and 8% Collie. A great-grandparent contributes about 12.5%, so this probably means Woo has three purebred great-grandparents, a Golden, a Malamute, and a Sheltie. He might have a fourth – a Collie – or that might be a great-great grandparent. The rest is just unknown. Yes, we do list 10.2% “others,” and the Food Lady can get that list. However, those breeds are all at under 5%, which means it’s such a small contribution that we don’t trust it; it could be computer error as easily as something real. We provide that info because we hate to withhold data, but we definitely don’t put any stock in it. Then there’s 35.2% unknown. We are finding a large number of dogs with a lot of “unknown,” probably meaning there’s just a whole lot of bits of different breeds at too small a percentage for us to actually pick them out.

So, Golden, Malamute, and Sheltie, huh? I bet you can see the Golden in his lovely orange coloring. And the Sheltie – along with the Collie, that’s the herding breed that we can easily see. But Malamute? Really? This is the kind of finding that makes people say “there must be something wrong with the results.”

I believe in the Malamute, though. Just one great-grandparent contributing some Malamute genetics can easily be lost in the wash. Remember, the genome is huge – so just because you can’t see any Malamute shape or color doesn’t mean there isn’t something else Malamute in this mixy-mix dog. There may be Malamute liver function or Malamute immune system or who knows what!

And Malamutes are big and Mr. Woo is – well, he’s not big when you measure straight up, although I have reason to suspect the Food Lady wishes it were easier to keep him smaller in the other direction. But I still believe in the Malamute. There are probably a few dozen genes that work together to affect a dog’s size. In the breeds that we’ve successfully picked out of this mix, we have the one really big one, one or two large ones, and the petite Sheltie. But there’s still about a third of Woo’s genome not spoken for, and he could easily have gene variants for small size in there. Moreover, Goldens and Collies carry some genetic variants for small size – if all of their genes were set to the “large” size variant then they’d be, well, Malamute-sized. So Woo happened to get the small size variants from those great-grandparents, maybe, and a bunch from his Sheltie ancestor, and not so many big ones from the one lone Malamute in there. (Who contributed… fluffiness? An impressively flamboyant tail?)

Our results for TWooie:

TWoo's results

The important results here are that TWoo is around 10% Australian Shepherd and around 13% Collie. This probably means he has two purebred great-grandparents, an Aussie and a Collie. Then there’s 59.2% unknown. That’s an impressive amount of unknown, but it’s not an uncommon amount – a lot of mixed-breed dogs look like this. So it looks like except for those possible two purebred great-grandparents, TWooie hasn’t had any purebreds in his ancestry for quite some time. It takes a lot of different contributions to make a dog as awesome as this one.

So, are they related?

Looking at the different breed results, it seems unlikely that these two are related – but astute readers of the Food Lady’s blog pointed out that it’s possible they share a mother and different fathers, and mom made up the entire “unknown” category. (In fact, this would mean they could even come from the same litter.) So we dug deeper.

We don’t normally do relationship analysis on the dogs we sequence, but I really wanted to know the answer for these two, because I’ve been following their saga since I was in vet school. (New posts to this blog were the highlights of otherwise stressful and exhausting weeks.)

Methods for genomics nerds: I used the PLINK2 software’s “genome” function to ask “how related are these two dogs?” I analyzed them in conjunction with 1800 other dogs from our data set, so that the program was able to take into account common allele frequencies in the population. I pruned for linkage disequilibrium. I got a Pi-hat of 0, and I checked the entire population to make sure that this wasn’t a fluke – indeed, most dogs got Pi-hat of 0 (unrelated) but a few got over 90% (sample duplicates in the dataset).

It was a sad day for me when the analysis said pretty clearly that they’re not related at all. I wanted to believe in this story, and they are my favorite dog brothers. But as the Food Lady and I discussed on private chat, they are clearly still soul brothers, and no one can take that away from them. They are and will always be the WooTwoo.

Want to talk with other people about this story? Post about it on the Darwin's Ark's forums!