Autism in the AAAS

« previous post | next post »

No, this is not me complaining again about the frustrating unwillingness of the AAAS to communicate with the public by making virtual versions of its marvelous symposia available on line. Instead, I'm going to tell you about the next symposium I'm about to sit in on: "Autism: Genetic, Epigenetic, and Environmental Factors Influencing Neural Networks", organized by Isaac Pessah and Cindy Lawler. The abstract:

Autism is a heterogeneous set of developmental disorders with complex etiologies. The goal of the symposium is to present a multidisciplinary perspective of how genetic, epigenetic, and environmental factors can interact to promote autism risk. The speakers will critically evaluate the evidence from human and animal studies that gene x environment interactions influence autism susceptibility, severity, and treatment outcomes. Genetic risk factors for autism will be reviewed. New evidence that autism may be associated with an increased copy number burden especially in regions of genomic instability, will be presented and discussed in relationship to environmental causes. How epigenetic mechanisms alter expression of genes relevant to autism will be reviewed in light of environmental chemicals that alter global and gene-specific DNA methylation patterns. Recent progress in understanding how impairments in neural connectivity contribute to autism will be reviewed. The role of methionine (MET) polymorphisms in autism risk and how polyaromatic hydrocarbons found in air pollution differentially influence individuals with the cMET autism risk allele will be presented. Evidence that low-level chemical exposures influence molecular and cellular processes that alter the balance of excitation and inhibition and neuronal connectivity relevant to the development of autism will be evaluated.


  1. bks said,

    February 18, 2012 @ 7:17 pm

    Pretty vague stuff. I would translate it as: "We got nothin'".


    [(myl) Not so. They have some promising and pretty specific ideas about genetic/epigenetic/environment interaction, with both human epidemiological evidence and evidence from mouse models, which suggest that a significant part of the problem may be due to effects on dendritic growth of environmental PCBs and also PDEs used as flame retardants, in children who have copy number variations in certain areas of the genome (affecting up to several hundred genes). If this is the true story, it would explain both why there has been some real increase in the prevalence of autism (beyond increased diagnosis and diagnostic substitution), and why it has been so difficult to figure out what the causes really are. See Pessah's abstract (in the linked pdf) and his recent and forthcoming papers for the details.

    It's a complicated story — they don't have "the gene for autism" and indeed deny that such a thing could exist; and the explanation involves the idea that the environmental toxins might cause hypomethylation of the extra copies of the regions involved, which in turn causes some changes in synapse physiology, which in turn affects dendritic growth in utero and in infancy. And it's probably even more complicated than that. At least, that's what I recall from the session — I wish that I had access to the presentation slides and/or a video, since some of the key results are apparently still in press.]

  2. Chad Nilep said,

    February 18, 2012 @ 9:53 pm

    More likely, "We got more than we can say in 200 words."

    It's hard to convey the complexity of one presentation in a conference abstract, let alone summarize and synthesize five of them.

  3. Ray Girvan said,

    February 18, 2012 @ 10:24 pm

    Hrrm. I'm sure – whether true or not – that an environmental explanation will go down well. Parents of autistic children (who are almost invariably on the spectrum themselves and so a PITA for their ability to fixate implacably on their preferred barking theory) can't abide any explanation that involves faults in their precious genes.

  4. Ellen K. said,

    February 19, 2012 @ 9:29 am

    Ray, my experience says people on the autism spectrum generally prefer to thinking of autism as genetic, ignoring that the data clearly shows that it's not purely genetic.

  5. bks said,

    February 19, 2012 @ 9:40 am

    I'm sure they're dead serious but "genetic/epigenetic/environment interaction" is just another way of saying "everything." Which is a lot like saying "nothing."

    [(myl) But the presentations described several large-scale studies which zero in on certain specific genetic (CNV) and epigenetic (hypomethylation) and environment (low levels of PCBs and PDEs) interactions, causing specific developmental problems (changes in synaptic physiology leading to increased dendritic growth) associated with ASD. I don't know whether this is the right story or not — we're talking about my memories of a three-hour symposium on preliminary work that's not in my field — but it's surely not "nothing".]

    BTW, for all non-trivial phenotypesP there is never a gene for P.

    [(myl) Indeed. But that doesn't stop many scientists and most science journalists from talking and writing about "the X gene", "the gene(s) for X", etc., for innumerable values of X.]

    I am not saying that the researchers are charlatans, nor that the research will not yield results, just that the abstract reads like they've got a lot of associations, but not, as we used to say, "the weenie."

    [(myl) But isn't this what good science in search of the explanations for complex phenomena ought to look like?]

  6. Theo Vosse said,

    February 19, 2012 @ 9:51 am

    My problem is that there is no causal model: we can't explain someone's behaviour from our knowledge of his (!) brain, and we can't explain the structure of someone's brain from our knowledge about his DNA and life history.

    [(myl) If you mean "we can't ever explain …", then you're clearly wrong on both counts. To take a trivial example, the brain/behavior links in the case of epilepsy are relatively clear; and in the case under discussion, there's a plausible sketch of a causal story about the links from CPV and hypomethylation to changes in synaptic physiology to changes in dendritic growth and attachment. If you mean "we can't always explain…" then of course you're right; and indeed the dissociation between what we know about the brain and what we know about the mind is remarkable; but so what?]

    So the jump from genes to autism is pretty big, and the results are correlational at best. I also find it hard to believe that such diverse causes would lead to identical changes in behaviour, and to only those.

    [(myl) Here's another case where it would be nice to have the symposium available on line. The large-scale epidemiological studies that were discussed in the session found CPV correlations with lots of things: from heart disease, immunological and gastrointestinal problems, schizophrenia… So if anything, the complaint ought to be that so far, there's no story about why this broad class of genomic variations is associated in different cases with such a wide range of differing anatomical, physiological and behavioral changes. Presumably it's because there are different sets of genes and gene-regulation circuits, engaged in a wide variety of ways, and interacting with a bunch of other factors. But that, as bks observed, is (almost) everything and therefore (almost) nothing.]

  7. Bob Ladd said,

    February 19, 2012 @ 4:28 pm

    It's easy to be skeptical about neuroscientific findings, but in the case of autism there really has been some scientific progress in the past 50 years. A couple of weeks ago I was looking up an alphabetically nearby word in my 1961 unabridged Webster's Third and happened to notice the definition of autism: "absorption in need-satisfying or wish-fulsilling fantasy as a mechanism of escape from reality". (And autistic is defined as "of, relating to, or marked by autism.") Those are the entire entries for those two words. In 2012 they sound like the definition of an entirely different condition, but they were presumably reasonably up-to-date and scientifically sound when the dictionary was published.

  8. AntC said,

    February 19, 2012 @ 11:24 pm

    Apologies in advance for a report-of-a-report, but I've seen a claim that diagnosis of autism has a much higher incidence on the U.S. East coast, peters out in the mid-West and is largely unknown on the West Coast. (It was mentioned in a lecture mostly about the educational system, and how 'concerned' parents are looking for some pretext for their kids under-performing. The lecture observed that with all the electronic gizmos kids have these days, no wonder they can't concentrate.)

    I think this was being taken much more as an indicator of medical attitudes than of the incidence of autism itself.

    Have Pessah/Lawler adequately controlled for sociological and attitudinal factors?

  9. Joe1959 said,

    February 22, 2012 @ 2:55 pm

    @Ray Girvan

    I am the father of two autistic children: as to where I lie on the spectrum, well you had better ask my wife that question! I am also member of the school board at a school for autistic children and am involved with a number of support organisations for autistic people and their families, so I know quite a few autistic people and quite a few families where one or members are on the autistic spectrum.

    From my experience I think most parents of autistic children, even those who my be on the spectrum themselves, would be grateful for any (well researched, scientifically valid) explanation for their children’s condition, in the hope that it might eventually offer hope for treatment or prevention.

    I would have no problem with an explanation that revolved around my “faulty” genes, but there are those on the autistic spectrum who would baulk at the term “faulty” (or indeed any implication that they [or their genotype] should be valued any less than [that of] the “neuro-typical”). Those of us who find it (more) easy to see things from the viewpoint of others, should try to understand that reaction from the point of view of people on the spectrum, who, in common with (some) other disabled people, worry that if their kind are less valued than the rest of humanity, that it might make it easier to justify discriminating against them, aborting foetuses with that condition / those genes, etc.

    [Apologies for turning this into Autism Log; no wonder Geoff Pullum doesn’t allow comments…]

  10. Lucy Kemnitzer said,

    February 26, 2012 @ 4:20 pm

    Coming in late, but autism diagnosis is not "largely unknown" at least in the part of the West Coast I inhabit. It's a common, and increasingly common, diagnosis for schoolchildren here.

    [(myl) My memory of the AAAS presentation is that statistics about large increases in recent years of diagnoses in California were featured. I believe that one of the papers cited was Hertz-Picciotto and Delwiche, "The Rise in Autism and the Role of Age at Diagnosis", Epidemiology 2009:

    Autism incidence in California shows no sign yet of plateauing. Younger ages at diagnosis, differential migration, changes in diagnostic criteria, and inclusion of milder cases do not fully explain the observed increases. Other artifacts have yet to be quantified, and as a result, the extent to which the continued rise represents a true increase in the occurrence of autism remains unclear.

    An earlier study shows a rise in California diagnoses during the period 1987-1994: Croen et al., "The Changing Prevalence of Autism in California", Journal of Autism and Developmental Disorders, 2002.

    A more recent review is Matson and Kozlowski, "The increasing prevalence of autism spectrum disorders", Research in Autism Spectrum Disorders 5(1) 2011.]

RSS feed for comments on this post