OT: for the dog people—and maybe genetics or stats folks—NYT: Has Dog Breeding…

[unterhausen]

Our most recent dog was feral, picked up as a puppy by a rescue in Georgia. He was bothering construction workers for food, which is a behavior that continues. He would probably be mad if he knew I make fun of him for being a chihuahua/Australian cattle dog mix. I resolved long ago that we would never have anything but a rescue. It's unfortunate that there are plenty of those.

My wife got connected with a beagle rescue. I help walk dogs on occasion. The number of dogs coming and going is amazing. And the reasons they are there can be pretty sad or infuriating.

[benb]

Originally Posted by ltwtsculler91

Sure, but dogs aren't humans. Compare it to livestock breeding or racehorses and it all looks really similar...

There is responsible breeding from the genetic and health testing that is done beforehand to ensure that there will be no known hereditary conditions passed down. This is how good breeders work, they work with a vet who's trained and make sure that the pair will not produce a genetic defect.

The only way they can call it responsible is by redefining the word responsible and ignoring biology. They are inbreeding these dogs so badly they are creating new genetic problems that aren't even known yet, so obviously the tests are not sufficient. And they have also redefined away defects into features. If every pug has a breathing problem saying it's not a problem doesn't make it not a problem. Do vets do full-genome sequencing on dogs to try and find a safe match? If they were doing that they would probably never be able to actually find a match of the same breed.

The most inbred populations of humans in the world do not even approach the amount of inbreeding of a deliberate mixed-breed dog. The mixed breed dog is still biologically considered dangerously inbred.

Given that there is no way you can possible respsonsibly breed a "pre-bred" dog that is already at a 0.25 or worse coefficient of inbreeding because there is no other dog of that breed that can realistically be considered a safe mate!

That is the whole basis of the scientific argument.

LOL.. great quote, this is the equivalent of a human saying, "So what I got my sister pregnant, we know it's safe cause neither one of us has gotten thyroid cancer."

[EB]

Wait until you guys find out what happens to livestock.

[RWL2222]

Originally Posted by kgreene10

I used the know the author. She is a serious academic with a fantastic publication record for both academic and non-academic audiences. I like the famous quote about statistics because it’s fun and provocative. Yet it could be taken to imply malfeasance, which, without much more information, I would have a hard time believing. If re-interpreted in light of thinking over the last 15 years or so, it could be taken as a critique of frequentism. I’m with you there.

I clipped the first part of the famous quote ("There are three kinds of lies: ...") and am not implying malfeasance on the author's part.

I had to look it up, but what's happened over the last 15 years wrt frequentism?

[benb]

Originally Posted by EB

Wait until you guys find out what happens to livestock.

There is absolutely no shortage of people screaming about how dangerous this is with both livestock and agricultural (food) plants.

Heck we are also doing a really great job of breeding antibiotic resistant bacteria!

[Ken Robb]

Originally Posted by ltwtsculler91

I could go on about this all day, as we have two Havanese and my wife is on the board of the NY area Havanese club and deeply involved in the national one. She's shown our 3-year old in Conformation (Champion, shown at Westminster, won best puppy at their national show in year 1) and has done agility with our 13 year old since 2015 (Masters level dog, general grumpy old man).

Overall, this is like rim versus disc that both sides will run around throwing out arguments that are more based on emotion than anything and the mixed folks have had a field day since these "Doodles" became popular and expensive. At the end of the day, a RESPONSIBLY BRED pure-breed dog is the best way to guarantee that you're going to get a healthy, well mannered animal. Unfortunately, many folks breeding both are not responsible so you need to do your homework and find a good breeder through a breed club.

Nando is being bred now, and has a few puppies, but all is closely examined for inbreeding and dog health which mixed breeders generally don't do. People love to talk about "inbreeding" in dogs, but the whole process is much closer to racehorses or livestock breeding than popular articles like this make it out to be. Sure there are plenty of irresponsible breeders, and some of the bad clubs will over select for certain traits in the show ring (like German Sheppards or Bulldogs right now) but there is usually big controversy in the breeding world about that and folks shift it back.

Here's some pics of ours, because a dog thread needs them!

Both of our boys recently:



Paco, our 13 year old lovable grump



Nando, our 3 year old little Champion who's now starting agility. No longer in full coat because that was 6 hours a week of work


Now in chill mode:

hmmm, we have a couple of Havanese dogs in the 'hood and their owners think Annie looks like a Havanese too. She surely does look like your dogs.

[kgreene10]

Originally Posted by RWL2222

I clipped the first part of the famous quote ("There are three kinds of lies: ...") and am not implying malfeasance on the author's part.

I had to look it up, but what's happened over the last 15 years wrt frequentism?

I do love the quote.

The Bayesian critique of frequentism has gained steam over the last while. Perhaps it’s been more than 15 years in some fields. In social sciences, where my knowledge lies, it’s probably been about that long.

The notion of a confidence interval — say a 95% CI — implies that, over repeated tests of the same hypothesis, 5% of the tests will yield estimates that do not capture the population parameter (ie, the true value). In this context, p-hacking — testing multiple models with small changes until you obtain the desired result — can lead to incorrect results.

There are ways to mitigate this vulnerability, such as pre-registering hypotheses and statistical model specifications for the proposed tests. Even that is imperfect.

I think the bottom line is that nothing in science is determinative. The hope is that vigorous debate among competing researchers can arc toward truth.

[Erikg]

I think we need a 'post pics of your dog thread', so cute!

[Alistair]

Originally Posted by ltwtsculler91

Sure, but dogs aren't humans. Compare it to livestock breeding or racehorses and it all looks really similar...

It does, but that’s not always a good thing.

We are selecting soeme dogs for traits that are negatives - snub nosed dogs that can’t breath well and overheat easy, dogs that can’t reliably give birth without C-section or other intervention, show dogs that are less “useful” than their ancestors (GSD with short rear legs and a sloping back).

We tolerate livestock breeding like this because they’re food. And we largely don’t see the sausage being made. It’s kind of wild we encourage some pretty obvious “defects” in household companions in the name of aesthetics.

My current best buddy is a poodle-jack Russell mix. Or something like that, we don’t know for sure.

[edgerat]

I have owned/trained/bred high level bird dogs for half my life. When you breed for performance, you are generally bettering the breed. When you solely breed for looks, "you" and I are going to have problems. When you no longer care about the animal's ability to ambulate (Bassett, etc) you are doing harm, in my opinion. All that said, there are still two things that kill dogs, cancer and heart failure. Some bloodlines have more cancer than others and I have avoided those for that reason. The AKC is like most large organizations these days, they have become a large business and they no longer appear to care about the well-being of the dogs in their registry.

[coffeecherrypie]

Originally Posted by christian

The stats seem pretty solid to me. I've linked the actual study below. The net seems to be that there has been so much inbreeding across many many generations that the percentage of shared alleles in purebred dogs is astonishingly high. Basically, there is little diversity added to the pool, ever.

https://cgejournal.biomedcentral.com...75-021-00111-4

The average inbreeding based on genotype across 227 breeds was Fadj = 0.249 (95% CI 0.235–0.263). There were significant differences in morbidity between breeds with low and high inbreeding (H = 16.49, P = 0.0004). There was also a significant difference in morbidity between brachycephalic breeds and non-brachycephalic breeds (P = 0.0048) and between functionally distinct groups of breeds (H = 14.95 P < 0.0001). Morbidity was modeled using robust regression analysis and both body weight (P < 0.0001) and inbreeding (P = 0.013) were significant (r2 = 0.77). Smaller less inbred breeds were healthier than larger more inbred breeds.

Wow, that is wild, F=0.25 within breeds. That means that *within* a breed, the average parents are as closely related as brother and sister would be in a non-inbred population.

[RWL2222]

'An animal whisperer and her feisty mutt Miles are among the hopefuls at the dog show’s agility competition.'

Gift Link: https://www.nytimes.com/2024/05/10/s...smid=url-share

[christian]

Originally Posted by coffeecherrypie

Wow, that is wild, F=0.25 within breeds. That means that *within* a breed, the average parents are as closely related as brother and sister would be in a non-inbred population.

It's shocking, but also not surprising when you think about it. If you consider that most modern dog breeds stem from the late 1800s with chokepoints at the World Wars, it's not inconceivable that the initial stock of some dogs only amounted to a couple dozen dogs and that almost all subsequent dogs of that breed are bred from descendants of those ancestors. For instance, Felix Dobermann began breeding Dobermanns in 1880 and died in 1894, when Otto Goeller became the pre-eminent breeder. I think it's safe to assume the total number of Dobermanns in 1900 was fewer than 500, and nearly all had parentage from those two breeders. At that point, it doesn't really matter how you interbreed the subsequent generations... without outcrosses, you're going to have this result.

[Likes2ridefar]

Originally Posted by .RJ

They pick you at the shelter, easy.

Everyone I know with purebred dogs has health issues, and mutts generally have pretty few.

Same, buying an expensive pure bred dog that often have oddities from inbreeding makes no sense to me when there are plenty of dogs looking for good homes. I see it nothing more as vanity unless after a breed as a tool. All my parents Shetland sheep dogs had issues but they sure looked pretty.

My first and current dog rescued from Mexico (her mother was brought here pregnant) ended up looking like a border collie that a breeder even insisted it was after we told her she isn’t. Dna tests say she is a mutt.

[benb]

It would be really interesting if some of the breeders actually started and tried to make a new breed that minimized that coefficient of inbreeding and tried to get it down below 0.03-0.05 while still trying to have some semblance of still being the same breed.