A Genetic Link to Autism Found? Sceptical.

Okay, I’m playing the sceptic very strongly here.

Vathara told me about an article on Science Daily on Friday that claimed that they’ve found a definite genetic link to autism that doesn’t involve Fragile-X (which is more linked to intellectual delay/disability than autism directly). According to the Science Daily article, and the abstract of the article it’s taken from, it involves a mutation in the CHD8 gene that results in GI tract issues, insomnia, macrocephaly (large heads and wide-set eyes), and being on the autism spectrum. Okay.

How did they discover this? They looked at over 6,000 kids on the spectrum, and found 15 displaying those characteristics. They interviewed those 15 kids. A genetic analysis determined that they all had the CHD8 mutation. They then disrupted the CHD8 gene in zebrafish and, lo and behold, the fish developed macrocephaly and GI tract issues.

Oh, but wait. There’s one thing that the Science Daily article and the abstract (which is all I can access) don’t say. Namely:

Where are the neurotypical control groups showing: A) macrocephaly; B) the same GI tract issues; C) the same sleep disturbance issues; D) the same GI tract issues and macrocephaly; E) the same GI tract and sleep disturbance issues; F) the same sleep disturbance issues and macrocephaly; and G) the same GI tract and sleep disturbance issues and macrocephaly?

Now, it’s quite possible that the researchers did look at those control groups (or at least control groups D, E, and G), and they just aren’t listed in the article or the abstract for some reason. If this is so, and any of my readers can access the full text of the article, please let me know.

If the researchers didn’t have those control groups, on the other hand… how can they possibly claim that the gene mutation is linked to autism? It’s quite possible that it’s linked to GI tract issues and macrocephaly, but has nothing to do with autism. Or at least has very little to do with autism. Certainly the zebrafish can’t tell if they’re autistic.

So… potential problem there. Does anyone have any answers?

(wry look) tagAught

5 thoughts on “A Genetic Link to Autism Found? Sceptical.

  1. Even if they’ve found a genetic link to one facet of one type of autism, they haven’t found “the genetic link to autism,” because, as all genetic researchers currently agree, autism is a polygenetic phenomenon.

    It would be like “the gene for tallness.” There are scores of alleles involved in determining height. It’s pointless to try to find one because there isn’t one; it’s polygenetic. I recently took a behavioral genetics class and the professor said that the most studied and genetically understood human condition is schizophrenia and even after millions of dollars and scientist-hours, we’ve only identified around 10% of the genes involved.

    Autism is the same way: the scientists have to *try* to find the genes involved, but even with the cost of research going down and speed of research going up at a rate that is exponentially exponential (no exaggeration), they will not have located 50% of the autism-related alleles in our lifetime.

    • Oh, yes, definitely. Maybe the title should be “A Genetic Link…”? The article is looking at that one specific (and very rare – seriously, 15 out of 6,000?) genetic link.

      😉 tagAught

      • 15 out of 6000 who have a phenotype found in roughly 1% of the population, meaning (when considering humans in general), 15 out of 600,000, which makes it quite rare, indeed.

        It is an interesting thing to discover, but so tiny in the scheme of things . . .

        However, it doesn’t surprise me that they found one of the alleles that links to sleep issues (endemic among us) and GI issues (also something we experience in a much greater proportion than the general population.) I just hope that researchers don’t fall into the thought-trap I’ve seen many succumb to, of thinking that autism *is* sleep disturbance and GI issues. It’s something many of us cope with, but it doesn’t even begin to touch the core of what makes us who we are, as Autistic humans.

        • Again, yes, exactly! That’s part of what this post is about; aside from the potential lack of neurotypical controls (which of course brings up the question, is it really 15 out of 600,000?), we want to make sure we don’t conflate “autism” with co-morbids that are simply found more frequently in our population. i.e. Epilepsy. I think epilepsy shows up… either 3 times more frequently in the autistic population, or in about 33% of the autistic population… but autism is not epilepsy.

          Enjoying this conversation!

          🙂 tagAught

  2. This is one reason I like tossing articles at other people I know to read; it’s reassuring to know I’m not the only one with the “wait, what?” reaction.

    One critical thing that often gets shoved under the scientific rug is, it’s very hard to get a “We did X and had no results” study published. Meaning journal articles are often biased toward “We found X in this case”, even though it’s possible that 9 times out of 10 they would have found nothing. So… yeah. Control groups, anyone? Anyone?

Leave a Reply