August 17, 2010

Dog breeds barely differ at all genetically

Some of the most prevalent Dumb Ideas about Race that crystallized as conventional wisdom around the time of Bill Clinton's celebration ceremony for the Human Genome Project revolve around the idea that genetic differences couldn't possibly cause differing behavioral tendencies among the races because we're all 99.999% (insert as many "9s" as you feel necessary to make the point) the same!

From the New York Times:
Wide Variety of Breeds Born of Few Genes
By SINDYA N. BHANOO

Spaniels have notably floppy ears, basset hounds have extremely short legs, and St. Bernards are large and big boned. Not to mention Chihuahuas.

Humans have bred dogs to produce tremendous variety. But a new study reports that the physical variance among dog breeds is determined by differences in only about seven genetic regions.

These seven locations in the dog genome explain about 80 percent of the differences in height and weight among breeds, said Carlos Bustamante, a geneticist at Stanford University and one of the study’s authors. The findings, published in Public Library of Science-Biology, are a result of what is the largest genotyping of dogs to date, involving more than 1,000 dogs and 80 breeds.

So, in the big picture, Labrador retrievers and pit bulls don't actually differ very much genetically. They do differ genetically on a very small number of genes. But some of those few genes that differ are related to biting. Labs tend to retract their teeth and gum whatever they're handling with their mouths, while pit bulls tend to bite down and, more unusually, not let go while they whip their heads back and forth. So, that tiny, tiny fraction of the dog genome is important when deciding which breed of dog to buy depending on your needs -- e.g., pet for your toddlers or guard dog for your crack stash.

By the way, here's Malcolm Gladwell in 2006 arguing that opposition to pit bulls is racial profiling and therefore immoral and ineffective, just like racial profiling of humans must, therefore, be.

The great thing about Malcolm is that he lacks the Uh-Oh-Let's-Not-Go-There gene that makes most spouters of the conventional wisdom prudently change the subject when they find that the logic of their argument has carried them to the edge of reductio ad absurdum.

48 comments:

The Anti-Gnostic said...

I want a guard dog for my crack stash for Christmas!

Anonymous said...

Question: Do pit bulls have 18% more serum testosterone than collies ? I reckon it's more like 180%. One chews kids to pieces and the other slobbers them and loves them. Kind of like people.

Are liberals simply brain damaged, or mutants ? I'm really wondering this. I don't mean it as a snide remark. There is something wrong with their brains and their perception of the world.

Anonymous said...

If the difference between a super-affectionate, gentle, friendly yellow labrador retriever and a ferocious, intimidating, musclebound pit bull is only a few genes........


......then those "few" genes are profoundly powerful and genes should be regarded as more, not less, potent than ever before.


Yellow labs are wonderful, gentle animals that are so easy-going that you can almost take food out of their mouths and they wont get agitated. I wouldn't advise engaging a pit bull while it was eating.

Anonymous said...

http://www.livescience.com/animals/dog-skulls-evolution-100125.html

"We usually think of evolution as a slow and gradual process," said study researcher Abby Drake of the of the College of the Holy Cross in Massachusetts. "But the incredible amount of diversity in domestic dogs has originated through selective breeding in just the last few hundred years, and particularly after the modern purebred dog breeds were established in the last 150 years."

"In contrast, Carnivora, the larger family in which domestic dogs belong, dates back at least 60 million years."

"the skull shapes of domestic dogs not only varied as much as those of the whole order [of Carnivora], but that the extremes of diversity were further apart in domestic dogs than in the rest of the order"

This is basically saying that dogs are interesting because there's a lot of stuff which is constrained in the entire family of carnivores which is less or equally as constrained in dogs which have been under "artificial" selection for 150 years than it is across this entire huge family (or at least the surviving lines) with 400000x(!) its age.

Dogs are, of course, all a lot more like one another than they are any other other member of Carnivora, when we take more than any one single trait (like members of human continental populations compared with members of others) but its amazing to see how quickly a trait can diversify when it is removed from its standing selection pressures and placed into an environment with new pressures (or no pressures).

Average Joe said...

These genes likely affect behavior as well as physical form. After all, it would be useless to have a dog with the body of a fighter if he lacked the aggression to make the most of his physical form.

Anonymous said...

>Are liberals simply brain damaged, or mutants ? I'm really wondering this<

Brain-damaged by religion. They are the adepts of a tradition (pdf) that tried to preserve religious ethics while riding the tidal wave of the rise of science. The perfect liberal combines Christian altruism with science - and short-circuits (the two are contradictory): when science says something that is "wrong wrong wrong," he discards science in an hysterical manner; and when religion crosses science (e.g., virgin birth), he discards religion in an hysterical manner (sneering, throwing bags of feces, etc.).

Such people are truly screwed up. They would be pitiable, were they not dangerous.

Anonymous said...

This article deserves a big "so what." Genes are complex molecules, and as most chemists know, small changes in complex molecules can result in vastly different properties. Simple example: many very different spices in your kitchen cupboard have the same basic structure as vanilla, the so-called the vanilloids, which include eugenol in bay leaves, zingerone in ginger, and capsaicin in chilli peppers. Changes in positions of a few atoms can make a huge difference to your tongue.

headache said...

The great thing about Malcolm is that he lacks the Uh-Oh-Let's-Not-Go-There gene that makes most spouters of the conventional wisdom prudently change the subject when they find that the logic of their argument has carried them to the edge of reductio ad absurdum.

More like he lacks the brains but makes up for that in skin color.

Kylie said...

Properly speaking, "pit bull" does not actually describe a breed of dog but is a catchall term used to refer to several of the molosser breeds.

Pit Bull Wikipedia.

Unfortunately, most people can't tell the difference between even, say, a Bulldog and an American Pit Bull Terrier as I discovered to my horror when a rumor went around my neighborhood that our bullie was a pit. It does make a difference, because some dogs that attack humans are mistakenly identified as "pit bulls".

OEBs, Staffies, ABPTs, etc. make wonderful pets though they do need proper socialization in puppyhood.

Anonymous said...

"...or guard dog for your crack stash.

By the way, here's Malcolm Gladwell..."

That was one hell of a transition. :-)

Chuck said...

Note the linked "genetics" page though:

Genetics:
"It is common knowledge that a person's appearance -- height, hair color, skin color, and eye color -- are determined by genes. Mental abilities and natural talents are also affected by heredity, as is the susceptibility to acquire certain diseases."

I guess the inference is a bit of a stretch.

Nanonymous said...

Of course it's not necessary for breeds to differ hugely genetically in order to be hugely distinct in various traits (including intelligence and temperament). E.g., one point mutation in the right gene can lead to a mental retardation and mental retardation is arguably a very big phenotype.

yet, let's read more attentively what the article is actually saying: "These seven locations in the dog genome explain about 80 percent of the differences in height and weight among breeds".

So, at least seven genes explain majority but not all of the variation in two traits. Like humans, dog breeds differ in a lot more than just size and color, so there is no doubt that dogs differ in a lot more genes, too. Indeed, the original abstract states:

we identify 51 regions of the dog genome associated with phenotypic variation among breeds in 57 traits

51 gives a large number of combinations (155111875328738228022424301646930321106325972001698611200000
0000000)

FF said...

It's odd how the people who own pit bulls seem very pragmatic about group and tribal differences, but say that it is conditioning of the dog that keeps thier kids safe, at least until the kid does get mauled.

On the other hand, liberals with a labrador maintain that social conditioning will 'fix' people, but won't go near a pitt bull because of the bad breeding.

M McK said...

There are lots of data showing that dog breeds are behaviorally different, and significant data that the behavioral differences are genetic. Localization of the regions (and soon the genes) that are different by as little as a single base pair will come soon.

There is a substantial literature about changing wild foxes into more domestic animals similar to dogs in a small number of generations. Interestingly, the behavioral changes correlate with morphological changes found in domestic canines.

The data on dogs is striking in that these breeds have largely been selected within the last few centuries. This leads to finding a small number of genes with mutations of large effect, with possible related deleterious consequences.

It may be that in human populations, behavioral and neurologically involved genes have developed over more generations leading to more alleles of genes with minor affect. 50-60K years is about 2000-2400 generations. Time for even moderate pre-existing variants to change frequency, and has been noted for notable changes with the expansion of the non-African population.

Some things appear to be very recent, including skin and eye color in European and Asian populations appear to be mediated by a small number of altered genes with high affect, but with different loci in each population.

On the other hand, their appear to be alternative alleles in monoamine oxidases in both African and out of African populations, for which there is some difference in behavior depending on allele a man has. In this case it appears that the frequency of the different alleles varies with populations indicating selection for one or the other depending on local conditions, both environmental and social.

Peter A said...

"Are liberals simply brain damaged, or mutants ? I'm really wondering this."

Libertarians share the same the crazy belief - to an even greater extreme. People have a vested interest in believing anyone can achieve anything if they just try. Makes them feel good.

Anonymous said...

So, that tiny, tiny fraction of the dog genome is important when deciding which breed of dog to buy depending on your needs -- e.g., pet for your toddlers or guard dog for your crack stash

Classic Sailer

BTW I fn hate people that own pit bulls- if it's a white dude it's a virtual certainty they have a barbed wire tattoo or something as stupid. I lump pit bull owners and terrorists into the same bucket.

Mohammed Atta= dumbass guy at the dog park

They really are imbeciles though- and inevitably once a year the pit bull attacks a lab or retriever and it's an ugly scene where they take no responsibility.


Dan in DC

Anonymous said...

to anonymous comment #2 - liberalism is a mental illness- they want a fn mosque at ground zero.

BTW talking to a devout muslim last week and he asked me my religion and i said catholic but I used to be a muslim until I converted..... he was fuming


Dan in DC

Anonymous said...

Steve, i read that bit from 2006 on Gladwell- you mentioned 1 person was killed bay a lab over the last 20 years- what you neglected to mention was that the person was licked to death.

dan in Dc

Anonymous said...

> People have a vested interest in believing anyone can achieve anything if they just try. Makes them feel good.<

Exactly. It makes them feel morally superior to people who don't achieve x. By their theory, such people are moral villains who chose to fail. Whereas they themselves are moral heroes who chose to succeed. The failures are put in the category of evildoers, of the immoral, of the willfully wrong; so they get to be blamed and scapegoated without natural limit. ("Pygmies could be NBA players if they would only try, so they must not be trying - what low character these people have!") The feeling that exercising moral judgment gives to people is the malignant warmth of prideful self-righteousness, which they wield as a flame against the winter wind of the fact that "choice" and "moral character" have a limited role in anyone's success or failure.

As a commenter on another thread here said: "If the problem is finally 'proved' to be genetic, then you immediately lose white guilt, because none of these problems would ever have been Ol' Whitey's fault in the first place, and you also burst the bubble of negro vindictiveness, because you can no longer blame Ol' Man Racism for all of the world's ills."

A great deal of moral righteousness and other forms of hysteria would be cut off at the root, if we understood humans scientifically. But until this happens, let's keep burning witches!

Melykin said...

Average Joe wrote:
"These genes likely affect behavior as well as physical form. After all, it would be useless to have a dog with the body of a fighter if he lacked the aggression to make the most of his physical form.
=============================
Well, not useless. Just not dangerous--just like my tiny Chihuahua who tends towards agression.

Some pit bulls do lack aggression, and make excellent pets.

Anonymous said...

***This article deserves a big "so what."***

You completely missed the point of this blog post. What part of the meme "races are 99.99% genetically the same, therefore there are no meaningful differences between the races" did you not understand?

Dogs demonstrate that you can have a species with almost no genetic variation at all (as defined simply by percentage of genetic difference) and yet have huge differences in appearance, intelligence, and behavior, based on a few genetic changes in a very limited number of locations.

Sort of like people. Thus the "we're 99.99% the same so race is irrelevant" argument is shown to be the scientifically illiterate shibboleth we knew it to be all along.

So there's you're "so what", pal.

Anonymous said...

Had a bullmastiff. She had a lot of the same bloodlines as what people call pitbulls. You could take food from her mouth and my kids tried to insert their heads as often as she would let them. Contrast that with similar looking dogs like those ones that killed that woman in San Fran.

eh said...

Not about dogs, but...

A couple of sites I had not seen before:

Racial Reality Blog

Anthroscape Human Biodiversity Forum

Anonymous said...

"You completely missed the point of this blog post. What part of the meme "races are 99.99% genetically the same, therefore there are no meaningful differences between the races" did you not understand?"

Sigh. The rude nerd strikes again. It's why I contribute as Anonymous. I fully understood the point of this blog post as a faithful iSteve reader. I was simply showing lack of surprise at the first example that comes to *everyone's* mind and adduced a non-genetic example of how simple changes in a complex molecule results in very different properties.

rob said...

More like he lacks the brains but makes up for that in skin color.

Headache, that's the silliest thing about Gladwell I have ever read. The afro got him where he is. It's all about his poofy hair.

Anonymous said...

The money quote in the Gladwell article is down at the end:

"the city could easily have prevented the second attack with the right kind of generalization—a generalization based not on breed but on the known and meaningful connection between dangerous dogs and negligent owners. But that would have required someone to track down Shridev CafĂ©, and check to see whether he had bought muzzles, and someone to send the dogs to be neutered after the first attack, and an animal-control law that insured that those whose dogs attack small children forfeit their right to have a dog. It would have required, that is, a more exacting set of generalizations to be more exactingly applied."

In other words, we should classify people based on their impulsiveness, their future-time orientation, their predisposition to violence, and their demonstrated intellectual capabilities or lack thereof.  Those whose traits indicate they are more likely to be dangerous should be watched more carefully and dealt with immediately when they get out of line.

But Gladwell would call this "profiling".

Svigor said...

These genes likely affect behavior as well as physical form. After all, it would be useless to have a dog with the body of a fighter if he lacked the aggression to make the most of his physical form.

Half the breed names designate roles and behaviors; they don't call them shepherds and retrievers over morphology. The behavioral genetic differences of dog breeds are readily apparent to anyone who works with more than one.

Svigor said...

ABPTs, etc. make wonderful pets though they do need proper socialization in puppyhood.

In my experience, most of the dogs called pit bulls are APBTs, and they are fantastic family pets. The thing you have to watch for is dog-dog aggression, not dog-human. They make crappy guard dogs because they've had the dog-human aggression bred out.

You're far more likely to be attacked by a Cocker Spaniel. Not that you'll mind as much...

SG said...

This doesn't surprise me at all.

I have an interest in neural networks, especially evolved neural networks. I recommend Steve look into this field.

What is shown is that a very small bunch of numbers can exhibit very sophisticated behavior such as playing chess (see Blondie 24).

My guess it that it a typical evolved chess playing net would be small enough to fit and run on a calculator from the 1970s.

(This of course is not taking into account the horse power needed to evolve such a net in the first place)

Anyway, the point is an embarrassingly complicated logic can be encoded into an embarrassingly small space.

Which brings us quite correctly to genetic difference.

Anonymous said...

I wrote about this in a comment before. I guess it bears repeating.

People act as if genetics is an analog process whereas of course it is digital.

If genetics were analog then the "mixing of the blood" would yield intermediates in proportion to their relative contribution to the "blood" of the progeny. That was pretty much how everyone looked at it before Mendel.

But genetics is largely a discrete or digital process. There are only four base pairs that matter and those are complimentary so that they always pair together. The result is that genetic coding is "on" or "off".

My computer science students never could get their heads around this simple fact. They were prepared to accept that music could be encoded on a CD or DVD with On-Off coding but recoiled from the notion that they themselves were in any way digital.

My Charlie (a toy spaniel) last week was frustrated when he couldn't sniff the butt of an Irish Wolfhound we met on our walk. It was too high. If this article is correct that only means that Charlie's height switch is off while the Wolfhound's is on. That's an over simplification but essentially accurate. Thought of this way dog variation is not very surprising.

Albertosaurus

Peter A said...

you also burst the bubble of negro vindictiveness, because you can no longer blame Ol' Man Racism for all of the world's ills.

I doubt it, honestly. You're assuming people with sub 100 IQs are even going to understand the argument. We still have 50% of the population that doesn't accept evolution, what chance does HBD have in penetrating beyond a small circle of the population? Hell, for all I know most of our key decision makers may already accept HBD as a fact of life and are simply using affirmative action and other racial games as ploys to control populations. Sometimes it seems that way.

Matt said...

Steve, while not exactly on topic, this may be consonant to your interests:

The Genetical Evolution of Chimp Culture

"Are differences in chimp behaviour down to culture or genetics? Some groups of chimps use twigs to fish for termites, others don't. Some groups communicate by knuckle-knocking, others by slapping. While the accepted wisdom is that these are purely cultural differences, now it seems that genes may be involved too."

Anonymous said...

"You're assuming people with sub 100 IQs are even going to understand the argument."

What's hard about "Whitey didn't do this to you, you were born that way, get over it"?

corvinus said...

In my experience, most of the dogs called pit bulls are APBTs, and they are fantastic family pets. The thing you have to watch for is dog-dog aggression, not dog-human. They make crappy guard dogs because they've had the dog-human aggression bred out.

A lot of people are scared of pit bulls because of their appearance. As they look like canine Orcs, they automatically seem fiercer. But I've met several pit bulls who didn't know me and they seem to act as friendly as a lab. Lovable galoots, all of them. (If they were abused, that would be another matter entirely.)

daver said...

Guess what people?

The present 'dumbest prevalent idea' on this subject is that genetic codes have to differ a lot to make presentation of individuals and behavior differ a lot.

It's. Not Fucking. True.

The idea is very analogous, actually, to that other prevalent idea: that we use 'only 2% of our brain' (or 10%, or what evar.) (Yeah, right. We evolved large brains to keep them mostly empty forever for no good reason at all.)

Any true computer nerd can tell you that programs that differ in only _one bit_ can manifest _completely_ different (and arbitrary) outcomes. An artificial example, sure (as theory-of-computation examples often are.) If you think that makes it irrelevant, you're simply

wrong

wrong

wrong.


peace, y'all..

2 said...

Of course genetic differences in dogs are out of fashion. It is more fashionable to pick up a shelter mutt. There are many reasons for the shelter mutt being more fashionable than the purebred, but I suspect that one of them is a knee-jerk reaction to eugenics, even as practiced in dogs. The left could embrace canine eugenics and use it for their own purposes. They could breed brave new breeds of dogs that would be designed to solve leftist or shared concerns: they could be bred to pick up litter and metabolize waste foods or they could be bred as better companion animals for the sick and disabled for example. But despite advances in genetic knowledge like this, we haven't had any breeding advances, only new uses/training for existing breeds. The move seems to be against productive use of breeding and toward the popularity of mutts. For that matter, what was the last species we aggressively bred for some function? I'm not talking about advances within already domesticated livestock (making cows better milk producers, more docile, etc). Seems like we could breed dolphins for life guard duty or ocean clean up or mice for pest control. But the prevailing thought is against messing with the natural order of things, that is, the the prevailing thought is against putting order into the random mother nature's more random processes. Could our culture's race obsession spill over into our treatment of animals?

Svigor said...

In my experience, the dogs that seem most behaviorally suited to violence are all the "varmint" dogs, little dogs bred to go into holes and kill small, vicious varmints like badgers.

Unknown said...

Pro - mutts are cheaper
Con - mutts are not as predictable in temperament as purebred dogs are.

The former wins out for me. If you want a rescue dog, guide dog, herding dog, police dog, et cetera, then you want specific breeds. But if all you want is a pet, then there's nothing necessarily wrong with getting a mutt.

RafeK said...

The various bull and terrier and bulldog breed are for the most part poor guard dogs and unlikely to attack humans because human aggression resulted in their ancestors being culled(a human agressive dog could not be rescued safely from a dog fight, bull bat, ratting contest etc).

There has been a recent trend to breed bull breeds for human aggression usually by crossing them to rottweilers or similar dogs. Which is unfortunate whoever pit bull types had the most consistent temperament testing of any breed at the local human societies I volunteered at even despite this trend. Bull breeds are have more capacity to inflict violence and are more attractive to the type of people who abuse dogs this probably explains the rather recent pit bull biting trend. Only 60 years ago pits were considered the safest breed to leave with a child.

Anonymous said...

Only 60 years ago pits were considered the safest breed to leave with a child.

The pet dog in the ancient "Our Gang" children's comedies was a Pit Bull named "Petey".

Truth said...

"The pet dog in the ancient "Our Gang" children's comedies was a Pit Bull named "Petey""

Actually, "Petey" was a Bull terrier. A Pit Bull Terrier is a different breed of "Bull Terrier" specifically bred to fight in a "Pit." And they are more aggressive and look very different (shorter muzzle) because they've been killing the docile ones for hundreds of years.

Svigor said...

The various bull and terrier and bulldog breed are for the most part [...]

That's pretty much my understanding as well. If you're breeding a fighting dog, you have to cull the ones who turn on master because the consequences are much more dire than with Cocker Spaniels. And yeah, my understanding is also that thugs began ruining the breed decades ago, not so much by crossing with human-aggressive breeds, but by failing to keep up standards (letting poor stock propagate).

Even temperament is a great way to put it. I grew up with an APBT and he was about as game as a dog can get, but he had the temperament of a saint with us. The provocations he put up with, other dogs wouldn't have I think. Not a single bite.

Anonymous said...

Everybody thinks his pit bull is wonderful. It's always the other fellow's that's the problem. When your pit bites some kid's face off, it's because he was "provoked" or was "just having a bad day." When the other fellow's pit does the same, it's because the man is an irresponsible abuser letting the breed go to seed.

Around and around it goes...

Meanwhile, steer clear of all pit bulldogs. (Warning: graphic images.)

RafeK said...

David I have never owned a pit, I did volunteer in human societies for 12 years, and saw more pit bulls then I care to remember, I can't remember one that was brought in due to having biten which was not uncommon with other breeds. They had the highest temperament tests, yes better then labs and in my experience were able to maintain their temperments better under the stress of the kennel then any other breed.

I remember one pit who was left at the kennel for over a year without developing any signs of kennel crazyness thats incredible. By contrast my husky/mal mix spent three months in a kennel and was going to be be put down because he had become uncontrollably agressive. It has taken me over a year of hard training to get him to the point he is safe and you can take food and toys from him without being bitten. Its just my experience but I don't think there is any valid science behind the opposite anti pit position either just some very poorly reported statistics. Go to any humane society and ask them about pit bull temperaments those people see that breed more then any other and the will tell you their wonderful dogs.

RafeK said...

Truth Petey was pit bull http://en.wikipedia.org/wiki/Pete_the_Pup
This is what a bull terrier looks like
http://www.breederretriever.com/photopost/data/718/medium/mini_bull_terrier.jpg
Rather different

CuChulainn said...

Many serious genetic disorders are the result of a single nucleotide substitution in a single gene, so even a minute change can have a profound effect.

Kylie said...

RafeK said..."Truth Petey was pit bull"

Not exactly. "Pit bull" is a catchall term for several breeds. But yes, the various Peteys were members of the bull breeds. According to Wikipedia, there were various Peteys throughout the years, starting with an APBT (American Pit Bull Terrier) and going on to an AmStaff (American Staffordshire Terrier). No Bull Terrier played the part of Petey, as far as I can tell.

The Bull Terrier AKA the English Bull Terrier is quite different in appearance from the other two.

All three breeds, though, are Molossers.

And for the record, no, I don't own a "pit bull". I do have two of the bull breeds, a bulldog and and an OEB.

Anonymous said...

What's hard about "Whitey didn't do this to you, you were born that way, get over it"?

Why do I get the feeling that is a quote from Omar Thornton's exit interview?