Deputy Tracy Valois is quizzing the Home Affairs minister in the States on Tuesday to find out if Customs officers have the skills and resources to identify and detain certain breeds of dogs." This comes after the dog – named Mr Bronx -was impounded at the animal shelter after the family returned from a holiday France with the pet (who had already been in Jersey since last year). Customs said it was a “pit bull”, a dangerous breed. And yet they had allowed its importation in the first place!
One site has some interesting facts.
DNA tests of pit bull-looking dogs often come up with some surprising results. One dog, which looked to all intents and purposes like a pit bull, turned out to be 40 percent poodle! That's a funny thought, but for the dogs it's a real problem. Many cities and counties – even whole countries – have laws that ban pit bulls. Law enforcement officers can go into people's homes and take away any dog who has "the appearance of a pit bull." Even if they're 40 percent poodle. They can be taken to the pound and then killed. (1)
How can this be? Another site gives me details, and shows how hard it is, given experimental conditions for identifications of the type to be made on physical aspects of the dig themselves:
“Pit bull” is not a breed but a type that describes several breeds. The American Staffordshire terrier, Staffordshire bull terrier, and American pit bull terrier are all pit bulls
An experiment showed how poor even experts are at identification:
At each shelter, the researchers picked out 30 dogs of all different sizes, shapes, and colours, and noted how each dog had been identified. They brought shelter workers from cage to cage and asked them to name each dog’s breed based on its appearance. If the assessor felt strongly that the dog had a secondary breed, they could note that. “Mixed breed” was also an option when they had no idea.
A vet on the research team examined all of the dogs, noting their height, weight, age, colour, and other characteristics. The vet also drew a small amount of blood from the dogs and sent it to a lab that could test their DNA.
The researchers’ hypothesis was correct. “We found that different shelter staffers who evaluated the same dogs at the same time had only a moderate level of agreement among themselves,” Levy said in the press release. And they fared even worse against the DNA analysis.
Shelter workers were able to spot real pit bulls and pit bull mixes 33 to 75 percent of the time, depending on the worker. But they labelled non-pit-bull dogs as pit bulls up to 48 percent of the time. That’s almost a 1 in 2 chance that a dog with no pit bull DNA could be lumped in with the unfortunate pit bulls. (2)
This is very worrying, In the case of Mr Bronx, as there is no indicator that DNA testing has been used at all and it is clear that visual and physical assessments are highly suspect when it comes to false negatives – that is, incorrectly assessing that a dog is a pit bull when it is not.
In February Customs contacted the family again and told them Mr Bronx was being impounded and would have to stay at the Animals’ Shelter until he could be assessed. Although an independent expert, paid for by the family determined the dog was not a pit bull, a Customs’ expert disagreed, a decision that was recently upheld in court.
Meanwhile, in Canada, the dangers of misidentification mean that there is a tendency to label a dog as a pit bull purely on superficial characteristics and behaviour. A report notes:
A dog that bit a woman in Notre-Dame-de-Grâce and was labelled a pit bull is not actually a pit bull, a DNA test by the SPCA shows.
As more municipalities mull bans on pit bulls and other dogs considered dangerous, the animal welfare group wants to show that identifying breeds is trickier than it looks. "It is virtually impossible — every expert, every report you will read, every peer review study explains that you cannot visually identify a dog's breed simply by looking at them," says Alanna Devine, director of animal advocacy for the Montreal SPCA.
In June, a man in NDG was charged with assault for ordering his dog to attack his wife. At the time the dog was identified by police as a pit bull, but the DNA test showed it was a mix of Rottweiler, mastiff and golden retriever. The SPCA says less than one percent of a dog's genes determines its appearance, and that there's no link between specific breeds and aggressiveness. (3)
But in America, DNA testing was used in a similar case to that of Mr Bronx
“Mindy is a canine victim of profiling. She was labelled a pitbull and that made it hard to find someone to adopt her, so shelter volunteers turned to science.”
After being abandoned, Mindy spent 6 months at the Trumbull, Connecticut Shelter. Because she looked like a pitbull, no one wanted to adopt her, so shelter workers looking for a way to help the sweet-natured dog find a home decided to solicit donations to test Mindy’s DNA to find out what she really was.
“Mindy is about 70-percent boxer and also bull terrier. She has some bulldog further down the line and a little bit of English cocker. So much for pit bull. What Mindy also has is a great personality and a bouncing, prancing way of getting around.”
Chalk up another victory for DNA in Kansas City where a man recently won an eight month legal battle with the city to keep his dog after a DNA test showed the dog wasn’t a pit bull.
Niko spent eight months at KCK Animal Control Kennels during his owners fight with the city. Animal Control Staff said the dog was a pitbull (a breed banned in the city), despite his owner’s assertion that Niko is a boxer mix. (4)
The American pit bull terrier is a term which can apply to any of the American Staffordshire terrier, the Staffordshire bull terrier, and a newer breed called the American bully. But there are no exact defining characteristics, because they share a lot of common characteristics with at least 25 other breeds of dogs, such as smooth coats or blocky heads.
In fact, a 2010 research article entitled “A Simple Genetic Architecture Underlies Morphological Variation in Dogs” demonstrated that “the dog, in contrast to some other species studied to date, appears to have a simple genetic basis dominated by genes of major effect”
Where there are genes of modest or small effect, most of the phenotypes – or traits - including body size, body mass index (BMI), etc appear to be under the control of hundreds of genes, each contributing a very modest amount to the overall heritability of the trait.
“The alternative model is that mutations of large phenotypic effect underlie most of these traits in dogs and that the same variants have been transferred to a wide diversity of dog breeds leading to phenotypic diversity from a narrow genetic base” (5)
This means that domestic dogs exhibit tremendous phenotypic diversity, including a greater variation in body size than any other terrestrial mammal, and moreover, much of the range comes from human intervention in the breeding process.
But this also means that...
“visual dog breed identification is accurate less than 25% of the time—even by professionals. According to Dr. Angela Hughes, a Canine Geneticist for Mars Veterinary, there’s a good explanation for that. “There are about 20,000 genes that go into making up a dog,” she explains. “For example, yellow colour is one gene; short legs is one gene. Of those 20,000 genes, only a couple hundred of them have anything to do with what your dog looks like.”
Dr. Hughes stresses that this is why a dog’s behaviour cannot be predicted by how he looks. “The genes that create a dog’s appearance are not the same genes that are influencing his behaviour,” she says. “That’s why it is important that we don’t pigeon-hole a dog based on how he looks.”
This can be particularly important in cases of breed-specific legislation (BSL), such as when any dogs that appear to be “pit bull type dogs” are banned from cities or automatically euthanized at shelters. Says Dr. Hughes, “It is incredibly difficult to say with any certainty that ‘this is a pit bull’ based on the fact that a dog has a blocky head shape, wide jaw and muscular build. Those same physical characteristics can be achieved from a variety of breeds, such as Boxers, Mastiffs, Bulldogs and many others. What’s more, those physical traits do not influence how that individual dog will behave, as his behaviour may be coming from genes of breeds that he looks nothing like.”(6)
Let us hope that more robust means of identification like DNA are used. Estimating breed on characteristics seems to be more akin to determining human personalities by phrenology, the study of the shape of the skull; it is not scientific, and a Magistrate's Court should not be allowed to determine a breed on that when science is available to provide far more certainty.
References
(1) http://www.pickthepit.com/
(2) http://mentalfloss.com/article/75759/dna-tests-show-many-shelter-dogs-are-mislabeled-pit-bulls
(3) http://www.cbc.ca/news/canada/montreal/pitbull-attack-ndg-1.3710598
(4) https://smartdogs.wordpress.com/2008/04/15/dogs-saved-by-dna-testing/
(5) http://journals.plos.org/plosbiology/article?id=10.1371/journal.pbio.1000451
(6) http://www.omagdigital.com/article/doggie+DNA/2010262/0/article.html
 |
| Mr Bronx |
One site has some interesting facts.
DNA tests of pit bull-looking dogs often come up with some surprising results. One dog, which looked to all intents and purposes like a pit bull, turned out to be 40 percent poodle! That's a funny thought, but for the dogs it's a real problem. Many cities and counties – even whole countries – have laws that ban pit bulls. Law enforcement officers can go into people's homes and take away any dog who has "the appearance of a pit bull." Even if they're 40 percent poodle. They can be taken to the pound and then killed. (1)
How can this be? Another site gives me details, and shows how hard it is, given experimental conditions for identifications of the type to be made on physical aspects of the dig themselves:
“Pit bull” is not a breed but a type that describes several breeds. The American Staffordshire terrier, Staffordshire bull terrier, and American pit bull terrier are all pit bulls
An experiment showed how poor even experts are at identification:
At each shelter, the researchers picked out 30 dogs of all different sizes, shapes, and colours, and noted how each dog had been identified. They brought shelter workers from cage to cage and asked them to name each dog’s breed based on its appearance. If the assessor felt strongly that the dog had a secondary breed, they could note that. “Mixed breed” was also an option when they had no idea.
A vet on the research team examined all of the dogs, noting their height, weight, age, colour, and other characteristics. The vet also drew a small amount of blood from the dogs and sent it to a lab that could test their DNA.
The researchers’ hypothesis was correct. “We found that different shelter staffers who evaluated the same dogs at the same time had only a moderate level of agreement among themselves,” Levy said in the press release. And they fared even worse against the DNA analysis.
Shelter workers were able to spot real pit bulls and pit bull mixes 33 to 75 percent of the time, depending on the worker. But they labelled non-pit-bull dogs as pit bulls up to 48 percent of the time. That’s almost a 1 in 2 chance that a dog with no pit bull DNA could be lumped in with the unfortunate pit bulls. (2)
This is very worrying, In the case of Mr Bronx, as there is no indicator that DNA testing has been used at all and it is clear that visual and physical assessments are highly suspect when it comes to false negatives – that is, incorrectly assessing that a dog is a pit bull when it is not.
In February Customs contacted the family again and told them Mr Bronx was being impounded and would have to stay at the Animals’ Shelter until he could be assessed. Although an independent expert, paid for by the family determined the dog was not a pit bull, a Customs’ expert disagreed, a decision that was recently upheld in court.
Meanwhile, in Canada, the dangers of misidentification mean that there is a tendency to label a dog as a pit bull purely on superficial characteristics and behaviour. A report notes:
A dog that bit a woman in Notre-Dame-de-Grâce and was labelled a pit bull is not actually a pit bull, a DNA test by the SPCA shows.
As more municipalities mull bans on pit bulls and other dogs considered dangerous, the animal welfare group wants to show that identifying breeds is trickier than it looks. "It is virtually impossible — every expert, every report you will read, every peer review study explains that you cannot visually identify a dog's breed simply by looking at them," says Alanna Devine, director of animal advocacy for the Montreal SPCA.
In June, a man in NDG was charged with assault for ordering his dog to attack his wife. At the time the dog was identified by police as a pit bull, but the DNA test showed it was a mix of Rottweiler, mastiff and golden retriever. The SPCA says less than one percent of a dog's genes determines its appearance, and that there's no link between specific breeds and aggressiveness. (3)
But in America, DNA testing was used in a similar case to that of Mr Bronx
“Mindy is a canine victim of profiling. She was labelled a pitbull and that made it hard to find someone to adopt her, so shelter volunteers turned to science.”
After being abandoned, Mindy spent 6 months at the Trumbull, Connecticut Shelter. Because she looked like a pitbull, no one wanted to adopt her, so shelter workers looking for a way to help the sweet-natured dog find a home decided to solicit donations to test Mindy’s DNA to find out what she really was.
“Mindy is about 70-percent boxer and also bull terrier. She has some bulldog further down the line and a little bit of English cocker. So much for pit bull. What Mindy also has is a great personality and a bouncing, prancing way of getting around.”
Chalk up another victory for DNA in Kansas City where a man recently won an eight month legal battle with the city to keep his dog after a DNA test showed the dog wasn’t a pit bull.
Niko spent eight months at KCK Animal Control Kennels during his owners fight with the city. Animal Control Staff said the dog was a pitbull (a breed banned in the city), despite his owner’s assertion that Niko is a boxer mix. (4)
The American pit bull terrier is a term which can apply to any of the American Staffordshire terrier, the Staffordshire bull terrier, and a newer breed called the American bully. But there are no exact defining characteristics, because they share a lot of common characteristics with at least 25 other breeds of dogs, such as smooth coats or blocky heads.
In fact, a 2010 research article entitled “A Simple Genetic Architecture Underlies Morphological Variation in Dogs” demonstrated that “the dog, in contrast to some other species studied to date, appears to have a simple genetic basis dominated by genes of major effect”
Where there are genes of modest or small effect, most of the phenotypes – or traits - including body size, body mass index (BMI), etc appear to be under the control of hundreds of genes, each contributing a very modest amount to the overall heritability of the trait.
“The alternative model is that mutations of large phenotypic effect underlie most of these traits in dogs and that the same variants have been transferred to a wide diversity of dog breeds leading to phenotypic diversity from a narrow genetic base” (5)
This means that domestic dogs exhibit tremendous phenotypic diversity, including a greater variation in body size than any other terrestrial mammal, and moreover, much of the range comes from human intervention in the breeding process.
But this also means that...
“visual dog breed identification is accurate less than 25% of the time—even by professionals. According to Dr. Angela Hughes, a Canine Geneticist for Mars Veterinary, there’s a good explanation for that. “There are about 20,000 genes that go into making up a dog,” she explains. “For example, yellow colour is one gene; short legs is one gene. Of those 20,000 genes, only a couple hundred of them have anything to do with what your dog looks like.”
Dr. Hughes stresses that this is why a dog’s behaviour cannot be predicted by how he looks. “The genes that create a dog’s appearance are not the same genes that are influencing his behaviour,” she says. “That’s why it is important that we don’t pigeon-hole a dog based on how he looks.”
This can be particularly important in cases of breed-specific legislation (BSL), such as when any dogs that appear to be “pit bull type dogs” are banned from cities or automatically euthanized at shelters. Says Dr. Hughes, “It is incredibly difficult to say with any certainty that ‘this is a pit bull’ based on the fact that a dog has a blocky head shape, wide jaw and muscular build. Those same physical characteristics can be achieved from a variety of breeds, such as Boxers, Mastiffs, Bulldogs and many others. What’s more, those physical traits do not influence how that individual dog will behave, as his behaviour may be coming from genes of breeds that he looks nothing like.”(6)
Let us hope that more robust means of identification like DNA are used. Estimating breed on characteristics seems to be more akin to determining human personalities by phrenology, the study of the shape of the skull; it is not scientific, and a Magistrate's Court should not be allowed to determine a breed on that when science is available to provide far more certainty.
References
(1) http://www.pickthepit.com/
(2) http://mentalfloss.com/article/75759/dna-tests-show-many-shelter-dogs-are-mislabeled-pit-bulls
(3) http://www.cbc.ca/news/canada/montreal/pitbull-attack-ndg-1.3710598
(4) https://smartdogs.wordpress.com/2008/04/15/dogs-saved-by-dna-testing/
(5) http://journals.plos.org/plosbiology/article?id=10.1371/journal.pbio.1000451
(6) http://www.omagdigital.com/article/doggie+DNA/2010262/0/article.html
Posts
No comments:
Post a Comment