Genetics 102: "Autism ain't a genetic disease!" - Not!
After posting my commentary on the rather appalling state of genetics education among senior editors of major news organizations (see here), I received an e-mail in which the author roundly castigated me for implying that autism might be a genetic disease. Well, for starters, I am not implying that autism is a genetic disease, I'm coming right out and asserting that it is.
Proponents of the myriad autism causation hypotheses - mercury, vaccines, gluten, casein, plasticizers, etc. - all insist that exposure to these substances cause autism in a subset of children. None of them has said (or even implied) that any of these substances causes autism in every child who is exposed to them. The reason they don't (and can't) say that is that it would be painfully easy to refute such a claim. It is obvious that the majority of children exposed to these substances do not develop autism - since even the highest estimates of autism prevalence can only claim that a third of a percent of children are autistic. This leaves 99.7% of children - most of whom have the same exposures as the autistic children - who are not autistic.
So, how do we explain how exposure to a substance (be it a chemical, a virus or a protein) can cause no illness in most people and yet cause illness in a few? Well, the answer to that problem is genetic variation. How is it that a small percentage of people exposed to Coxsackie virus B4 develop diabetes and most people do not? Genetic variation. How is it that, in the pre-antibiotic days, Plague was not 100% fatal? Genetic variation. How is it that some groups of people are prone to breast cancer and others are prone to gastric cancer? Genetic variation.
As unpalatable as it may be to some of the autism advocates, any claim that exposure to a substance causes autism has to explain how that exposure doesn't cause autism in 100% of children exposed to it - and the answer is... genetics.
Personally, I'm fine with the idea that there are a group of children who are genetically susceptible to develop autism after exposure to some substance or another. This would be completely in line with many other disorders. In fact, it could probably be argued (and successfully) that all disease is the result of some aspect of the genome. You could even make an argument that certain accidental and traumatic injuries are the result of risk-taking behavior that might also be encoded in the genes.
So, as painful and uncharitable as it may seem, autism is - in all probability - a genetic disease. Like so many others. To deny this is to take yet another step away from reason and toward...the Dark Side.
Prometheus
Proponents of the myriad autism causation hypotheses - mercury, vaccines, gluten, casein, plasticizers, etc. - all insist that exposure to these substances cause autism in a subset of children. None of them has said (or even implied) that any of these substances causes autism in every child who is exposed to them. The reason they don't (and can't) say that is that it would be painfully easy to refute such a claim. It is obvious that the majority of children exposed to these substances do not develop autism - since even the highest estimates of autism prevalence can only claim that a third of a percent of children are autistic. This leaves 99.7% of children - most of whom have the same exposures as the autistic children - who are not autistic.
So, how do we explain how exposure to a substance (be it a chemical, a virus or a protein) can cause no illness in most people and yet cause illness in a few? Well, the answer to that problem is genetic variation. How is it that a small percentage of people exposed to Coxsackie virus B4 develop diabetes and most people do not? Genetic variation. How is it that, in the pre-antibiotic days, Plague was not 100% fatal? Genetic variation. How is it that some groups of people are prone to breast cancer and others are prone to gastric cancer? Genetic variation.
As unpalatable as it may be to some of the autism advocates, any claim that exposure to a substance causes autism has to explain how that exposure doesn't cause autism in 100% of children exposed to it - and the answer is... genetics.
Personally, I'm fine with the idea that there are a group of children who are genetically susceptible to develop autism after exposure to some substance or another. This would be completely in line with many other disorders. In fact, it could probably be argued (and successfully) that all disease is the result of some aspect of the genome. You could even make an argument that certain accidental and traumatic injuries are the result of risk-taking behavior that might also be encoded in the genes.
So, as painful and uncharitable as it may seem, autism is - in all probability - a genetic disease. Like so many others. To deny this is to take yet another step away from reason and toward...the Dark Side.
Prometheus
posted by Prometheus at 2:29 PM
6 Comments:
That was very interesting that Mr. Olmstead (I assume that is who you meant) responded to your article.
Did he look at any of the several studies of the Amish and their genetics. A search of "genetics and Amish" brings up over 200 entries (not all are studies), including all the ones you referenced (did he check those?).
Or when "genetics autism" is put into the www.pubmed.gov search page there are over a 1000 entries (again, not all represent research papers, several are reviews, letters, etc). I know that there are researchers at my local university that have combined with other researchers around the country for twin studies to check into the genetics angle.
I really don't remember that Mr. Olmstead actually had a bibliography in his article... double check, click on the links to his articles... uh, no he didn't. So if he didn't do the book-work, how can he categorically say that autism is NOT genetic? How is he supposed to make us think he did research when he didn't document it?
Just to clarify - when I used the term "author", I was referring to the author of the e-mail. Mr. Olmsted has not noticed this humble blog as yet.
In fact, Mr. Olmsted has continued writing about autism at a prolific pace, with a fairly consistent slant - e.g. that autism is caused by mercury and anyone who says otherwise is, at best, mistaken.
Prometheus
Ah, I see... I get confused every so often --- especially when starting a summer schedule (I double-booked one kid for Wednesday).
Thanks (though it still does not excuse the Wash Times journalist for his total lack of research skills).
Today the Associated Press had an article about the Clinic for Special Children and the Amish Research Clinic...
http://www.usatoday.com/tech/science/genetics/2005-06-27-amish-genetics_x.htm
Prometheus,
Is it overreaching then to say that there is a possibility that there is a subsection of the population who are genetically suscpetible to heavy metal toxicity. If so, how would you go about testing this hypothesis, and would you treat these found to be affected, using FDA approved methods for heavy metal toxicity?
Fortunately/unfortunatley there isn't a whole lot of actual data on the variation of humans to heavy metal toxicity. Naturally, no ones been keen to inject heavy metals into people and study the effects, even at very low levels, in terms of how they process and excrete these metals.
To Ian,
It is absolutely probable that people vary - perhaps widely - in their susceptibility to mercury and all other "toxic" exposures. This has been shown time and again in industrial and accidental exposures - some people react badly to smaller doses that cause other people no effect.
However, this still leaves it to the autism-mercury promoters to demonstrate that mercury can cause autism.
I must reiterate that showing that mercury is toxic is not the same as showing that it can cause autism. Cyanide can cause neurological damage, as can carbon monoxide, organophosphates and the herpes virus. However, none of them (or mercury) have been shown to cause AUTISM. Until that connection is shown, the autism-mercury supporters have not made their point.
Prometheus
<< Home