hit counter joomla

The New England Journal of Medicine published a study yesterday showing that small changes in the DNA in the long arm of chromosome 16 are associated with autism.

Blogging on Peer-Reviewed Research

i-60b957fad1b9e1a3e306d0c520e02043-kids_singing.jpgI met a teenager with autism last summer, when I attended family night at the Seattle Park and Rec summer camp program for kids with special needs. It’s a fantastic program. The kids spend a week or more at summer camp and the parents get some much needed time off.

I sat down on a log and my daughter (one of the counselors) introduced to me a boy in her cabin and told me that he’s interested in trains. Since I rode a train for three days once when Northwest Airlines went on strike, I told him about it. For the rest of the night we talked about trains. Every few minutes, he would look at me and tell me something about trains and where they run.

Sandy, did you know that there’s a train station in Seattle?

Yes, I do.

They go to California and north all the way to East Coast.

hmmm, uh. cool.

15 minutes later.

Sandy, did you know there are trains that run through Seattle?

This went on all night until I left.

When you meet people with autism, it’s hard not to be interested in what makes them the way they are. We do know that many cases are inherited and that 10% of the cases have been linked to specific abnormalities in chromosome structure, but the cause (or causes) of the other 90% of the cases has remained a mystery.

The study by Weiss, et. al. looked at all 46 chromosomes in 751 families in search of genes related to autism.

They found that 1% of the children with autism from these families had changes in the DNA in the p11.2 region of chromosome 16. The changes were either duplications or deletions of a 147,000 base region of DNA. The children with autism had changes in this region, their parents (without autism) did not.

It’s not yet known which of the 25 genes in this 147,000 base region is related to the development of autism, but the trail is getting warmer.

Weiss, et. al. 2008. Association between Microdeletion and Microduplication at 16p11.2 and Autism. NEJM www.nejm.org January 9, 2008 (10.1056/NEJMoa075974)


  1. #1 Dave Briggs
    January 10, 2008

    It’s not yet known which of the 25 genes in this 147,000 base region is related to the development of autism, but the trail is getting warmer.

    It’s great to see the trail getting warmer! Anything that can be done to help these people and their families is a wonderful thing!
    Dave Briggs :~)

  2. #2 Emily
    January 11, 2008

    FYI, those of us in the world of autism get our hackles up when referred to as “these people.” 😉

    This paper has caused a great deal of interest in the autism blogosphere, a smallish but vocal group of people with autism, parents of children with autism, and autism advocates. There is a controversy among us people about tracking genes vs. focusing on symptomatology. As a scientist, I like to begin at the beginning and go on from there (kind of like the White Queen in Alice in Wonderland), so I like the search for genes. Some of us spend a great deal of time explaining the relevance of mechanism to the development of treatment, therapy, and cure. All part of a continuing effort to ensure scientific literacy continues to spread.

  3. #3 Germophobe
    January 12, 2008

    It’s not genes folks. You don’t have to believe me. Believe Charles Darwin. Natural selection would never select towards a heritable gene that trashes the fitness of nearly 1% of our children.

    I’m not implying that genes aren’t involved but they only offer increased susceptability to something else in the environment.

    If scientists found damaged genes, something in the environment hosed them. Radiation, pollution, pathogens, etc.

    Researchers are finally coming to terms with the fact that despite all of the genetic correlations Autism is environmental.

    “Until recently, about 90 percent of autism research has focused on genetics, and only perhaps 10 percent on environmental factors, said Dr. Gary Goldstein, chairman of the scientific board of Autism Speaks, a national research and advocacy group. In the coming years, he expects the ratio to be 1 to 1.”

    “any major article or proposal concerning the causes of autism is coming to be considered incomplete if it doesn’t talk about a potential role of environmental factors.”

    Autism and Schizophrenia strongly linked to Flu Virus exposure

  4. #4 Germophobe
    January 13, 2008

    Whoops, I forgot this one.

    Autistic kids have inflamed brains

  5. #5 Sandra Porter
    January 14, 2008

    Natural selection would never select towards a heritable gene that trashes the fitness of nearly 1% of our children.


    Your argument about evolution doesn’t work or apply here. Evolution isn’t that simple. In fact, evolution can select for genes that can have harmful consequences. Some genes are harmful when we have two copies (i.e. the sickle cell allele of the hemoglobin gene) but beneficial when we only have one copy (people with one copy are more resistant to malaria).

    Now, why doesn’t your argument apply here?
    Because the parents didn’t have the mutated form of the gene, so natural selection wouldn’t have a chance to act.

    That brings us to the question of where did the mutated genes come from? And, also to your suggestion that this form of autism could have an environmental cause.

    You could be right, but this paper doesn’t provide any evidence either way. Environmental factors might play a role in causing the mutation.

    One of the interesting things about this paper was that the parents didn’t have the mutated region on chromosome 16. This means that the mutations would mostly likely have occurred, in the parents, during development of either the sperm or eggs (I don’t think the paper said which one). Environmental toxins can cause mutations in developing sperm, maybe eggs, too, I don’t know. This would be an interesting topic to research.

  6. #6 Germophobe
    January 14, 2008

    Hi Sandra,

    There is a 99.99% chance that the mutated genes are the result of negative, environmental input. 250,000 years of natural selection virtually ensures this to be the case. I wish the article would have mentioned what percent of children with the mutated gene DID NOT get Autism. I believe identical twins are 50% concordent for Autism. Meaning if one has the disorder the other doesn’t about half the time. With this gene mutation, I dunno.

    You are correct about heterozygote advantage. Sickle Cell is the result of a good gene that can also have very bad consequences. However, in the case of Sickle Cell it is region specific. Those types of genetic disorders are always region or group specific because they respond to a specific threat in the environment. The plants, animals and even germs located just 100 miles from where you live are probably very different than those just outside of your town.

    Autism doesn’t work like that. It is a global phenomenon, occurs in children and is very common. That is a giant, red flag that the environment is the root of the problem.

    If Autism happened in middle-aged people after they had children or was group specific a gene would make a lot more sense. But kids? Perhaps a rare varient of Autism is caused by a strange combination of genes. But nearly 1% of children and still climbing? Not a chance.

    Up until very recently 90% of all Autism research dollars have gone towards genetic research. Of course they are going to find something. I would hope so!

  7. #7 Germophobe
    January 14, 2008

    Hi Sandra,

    One last thing that is maddening to me. I’m no genius but I posess enough basic knowledge about genetics and illness to be able to spot B.S. when I read it.

    Here is a timely article from a good site, Science Daily.
    The article explains that scientists have found a gene (CNTNAP2) that causes Autism. No ifs ands or buts, it causes Autism.

    But I know B.S. when I read it. No such gene exists or could exist in any common form. So I google and find that CNTNAP2 doesn’t cause Autism but it does raise susceptability. That’s a huge difference. There are genes that raise and lower our chances of catching the common cold, but they have nothing to do with the disease. The cold is triggered by a Rhino virus.

    When are scientists going to embrace Natural Selection again? I guess when the funding runs out for gene research. 😎

  8. #8 Sandra Porter
    January 14, 2008

    Hi Germophobe:

    I can answer some of your points.

    1. First – natural selection is not relevant to this paper. Natural selection can only act on forms of genes that exist, not those that are newly created during egg or sperm formation. These mutations were new – (that’s what de novo means). They occurred in the parents but only in a few of their cells – those that become sperm and eggs. The mutations were not present in other cells – skin, teeth, eyes, heart, etc. .

    The mutations that happened in these cases of autism are more like the mutations that cause Down’s syndrome. They’re not targets for natural selection because they occur during development of the sperm or the egg cells.

    2. For this one:

    I wish the article would have mentioned what percent of children with the mutated gene DID NOT get Autism.

    They did.


    By contrast, in a control sample of 18,834 subjects who did not undergo screening for a psychiatric or language disorder, only two deletions were observed ….


    3. As far as the Science Daily article, those articles are written by journalists not scientists. I looked at the article and I can see what you mean about the confusion between a gene that increases susceptibility and a gene that causes a disorder. The language in the article is sloppy.

    4. So why then would you see multiple cases of autism occurring in one family?

    I suspect that if a mutation happens during the formation of sperm or eggs, it could contribute to multiple eggs or multiple sperm that carry that mutation.

    You’re raising interesting questions that deserve longer explanations. When I get time, I’ll write about this topic more in a longer post.

  9. #9 Germophobe
    January 14, 2008

    Thanks for responding Sandy,

    I know the paper didn’t address this point but…
    Doesn’t the fact that the mutations only occured in certain cells indicate genetic disease? Viruses do this all the time and then get incorporated into the genome. Retroviruses that harm us get konked out pretty quickly. Those rare few that are beneficial spread throughout the population as if they were our own genes.

    Ancient retrovirus required for sheep to reproduce

    Maybe my brain is a little slow but I still don’t totally understand the percent of people WITH the gene who don’t end up with Autism.

    By contrast, in a control sample of 18,834 subjects who did not undergo screening for a psychiatric or language disorder, only two deletions were observed ….

    So they are saying that without this deletion there is essentially zero mental illness in human populations?

    What the heck? Am I reading that correct?

  10. #10 Germophobe
    January 14, 2008

    Just one more piece of data about my earlier point.

    Here is a more or less complete list of human genetic disorders.

    If you click any of those links you’ll see that heritable genetic disorders are all rare or group specific. Very few are more common than about 1 in 10,000 people. Some occur in 1 out of millions. Lots require strange combinations of many genes. Hereditary deafness is one example of that, some types require 70+ genes.

    I guess what I’m sayin is that the term “genetic disorder” is thrown around so much (particularly in the press) that people forget to read the fine print. In virutally every single case scientists are refering to genes that increase or decrease the likelihood of a disorder. Few actually cause the disorder. Genes or gene combinations that actually cause disorders, particularly in young people don’t last long in nature. The number of exceptions is exceedingly tiny.

  11. #11 Sandra Porter
    January 14, 2008

    Germophobe: You’re absolutely right about terms getting thrown around in an inaccurate way.

    Genes don’t cause disease.

    It’s the mutations in genes that can disease, and even then, only a fraction of the genes cause a problem.

    As this question:

    Doesn’t the fact that the mutations only occured in certain cells indicate genetic disease?

    This is a long answer. I’ll try to get to this question in a blog post later this week.

    And now, a little bit about the screening process and percentages of people with or without. The authors of the paper reported the information on frequency in a few different groups, I just picked out the group of 18,834 people as an example.

    The group of 18,834 people were not autistic (as far they knew), so these would be “normal” people, I guess, as far as the study was concerned. They found 2 people out of the 18,834 who had a mutation in the same region of chromosome 16. I calculate the percent of normal people, then, by dividing 2 by 18,834 and multiplying the result by 100. That gives me 0.01 percent.

    So 0.01 percent of “normal” (non-autistic) people had the mutation.

    It could be that these 2 people really do have autism and haven’t been diagnosed. I don’t know. The conclusion that the researchers drew was that this mutation occurs very rarely and that it’s found at much higher rate in autistic people than it would be in a group of people who do not have autism.

  12. #12 Germophobe
    January 14, 2008

    Hi Sandy,

    I look forward to your thoughts.

    In this particular disorder it appears that a small section of the gene is consistently altered/deleted.

    I’m not an expert and I’m sure there are alternate explanations but that sounds like a virus. The reason is the consistency of the mutation. viruses are the only environmental force that I am aware of that can reprogram our DNA with consistency. If the gene was generally breaking or unraveling it could be anything. If nearby genes tended to be damaged as well that would be a clue it was radiation, or pollution, etc. This sounds more like a lasar guided smart-bomb. The same little spot missing in tens of thousands of children.

    I guess we’ll know someday.

  13. #13 Germophobe
    January 22, 2008

    This is an old thread but…

    The role of viruses in the etiology of chromosomal abnormalities

    Even if scientists don’t know how it’s happening a virus, or some other enviromental agent is the only thing that makes sense. Mother nature doesn’t select towards common disfunctions.

  14. #14 Melissa
    April 29, 2010

    I feel that this could also be linked to transposons moving around during development!

New comments have been temporarily disabled. Please check back soon.