Wednesday, June 9, 2010

The Real Numbers are Numbing Enough ....



I don't think anyone has ever accused me of being a defender of inbreeding.

I'm against it, and I am not shy about saying so. In fact, I have been pretty vocal on the subject for a pretty long time.

Having said that, it's important to note that sometimes folks wade into debates a little too quickly. While their heart may be in the right place, their enthusiasm may exceed their attention span.

Such is the case in an article from In These Times, which was forwarded to me this morning. The author (a summer journalism intern) writes on the perils of canine inbreeding, noting that:

In some breeds, dogs have a 75 percent inbreeding coefficient (the indication of how related they are), according to Susan Thorpe-Vargas, a breeder of Samoyeds who holds a Ph.D. in genetics. (A brother and sister have a coefficient of 25 percent.) John Armstrong, a veterinarian with the Canine Diversity Project, reports that certain lines of standard poodles have an inbreeding coefficient of 70 percent.

According to data from the University of Pennsylvania school of veterinary medicine, three-fourths of puppies with a 67 percent or higher coefficient will die within 10 days.


I have to say I paused here. You see I saw a couple of obvious red flags.

The first is that there is NO breed of dogs that has a coefficient of inbreeding of 75%.

None.

The average coefficient of inbreeding (COI) of pedigree dogs is not 70 percent or 60 percent or 50 percent or 40 percent. The average COI is not 30 percent of even 20 percent. It is less than 15 percent. [see Golden Retriever example with sample of over 275,000 dogs].

The second red flag is the notion that a COI of X percent will result in a specific neo-natal mortality of Y percent.

Not true. Things are not that simple.

If a couple of dogs with no negative recessive genes at all (a very rare thing) are bred together, a very high COI can be achieved without a jaw dropping level of neo-natal mortality.

YES, as a general rule, the higher the COI the more likely there are to be both generalized and breed-specific health problems, but there is no specific mathematical health outcome like neo-natal mortality.

Things are not that simple.

The third point is that John Armstrong is dead and has been for a number of years.

This is a small point, I know, but it jumped out at me as "reports" is an active verb and John Armstrong is not doing much of it these days.

Armstrong, of course, did write about the health outcomes that befell a particularly inbred strain of standard poodles, but that example was chosen because it was an extreme example of inbreeding, not because it was an average.

So where did this young reporter get her information?

It appears she gleaned it from a not-too-well written online "genetics tutorial" put out by the University of Pennsylvania School of Veterinary Medicine.

I went to the link supplied, and found no footnotes and no author. By backlinking, I discerned it was an online tutorial to provoke thought (and perhaps discussion) among young veterinary students.

A quick Google, and I found that the beagle colony example given (it was not all dogs as suggested in the In These Times article) applied to only a single heavily inbred beagle colony at the Argonne National Laboratory.

As for the reference to Scott and Fuller, I assume it is to their book, entitled "Genetics and the Social Behavior of Dogs." You can read that book yourself, but it does not quite say what the In These Times author or the University of Pennsylvia veterinary school web site suggests. In fact, the evidence is a bit contrary.

For example Scott and Fuller note that Basenjis are among the most-inbred dogs in the Kennel Club (at the time they had a COI of .25) , but they also note that they also had the lowest neonatal mortality rate of the five breeds listed -- half that of beagles, and even less than half that of shelties, cocker spaniels and fox terriers.

Whoops!

Did Scott and Fuller say hybrid dogs (cross- and mixed-breeds) had lower rates of neonatal mortality than pure bred dogs? Of course!

No one is arguing that hybrids are not generally healthier than pure breeds, or that higher coefficients of inbreeding are not related to defect and disease.

But numbers and data exist for a reason, and the case for Kennel Club change can be made without misusing data or ignoring it.

Of course, error of fact exists from one end of the Internet to the other.

So why have I not rolled over and just let this one go?

Well for one thing, the article is sufficiently shocking in message and sufficiently "science sounding" that it has already popped up several places.

And yet it is wrong.

And it is not just a little wrong. It is very wrong.

And it is not just wrong in fact, it is wrong in message.

You see, if you read this article and believe it, you might think there is no hope for pedigree dogs.

After all, if pedigree dogs have a 75 percent inbreeding coefficient, as this article suggests, then everything about the Kennel Club has to change, and yet we all know that is not going to happen.

But suppose you know that the average coefficient of inbreeding in pedigree dogs is around 10 percent, and that only a "sick minority" of breeders are producing dogs with a COI of more than 20 percent?

What then?

Well then, there is hope!

Then there is no excuse for the Kennel Club not taking action.

After all, if they ban the registration of new puppies that have a coefficient of inbreeding (COI) of more than 10 percent, it will not flip the boat, but only change its course.

They can embrace a common sense policy, and do right by the dogs without engaging in economic suicide.

Perfect!

So let's ask them to do that.

In fact, let's demand it!
.

No comments:

Post a Comment