More failures to find human behavior genes

July 8, 2009 • 7:13 am

Over at the New York Times, science writer Nicholas Wade reports on the difficulties researchers have had in pinpointing genes that promote the appearance of schizophrenia:

The journal Nature held a big press conference in London Wednesday, at the World Conference of Science Journalists, to unveil three large studies of the genetics of schizophrenia. Press releases from five American and European institutions celebrated the findings, one using epithets like “landmark,” “major step forward,” and “real scientific breakthrough.” It was the kind of hoopla you’d expect for an actual scientific advance.

It seems to me the reports represent more of a historic defeat, a Pearl Harbor of schizophrenia research.

The defeat points solely to the daunting nature of the adversary, not to any failing on the part of the researchers, who were using the most advanced tools available. Still, who is helped by dressing up a severely disappointing setback as a “major step forward”?

Well, if you’ve been around a while, you’ll find this depressingly familiar.  Reports of finding gene after gene for things like alcoholism, “novelty seeking,” depression, and homosexuality have been withdrawn or not verified after further research.  There may be many genes affecting the propensity to behave a certain way, or to contract a certain disease, and variants of each gene may contribute only a tiny bit to the trait, making locating these genes a very difficult — or even impossible — task.  And once you find one of them, what can you do?  A gene that, say, increases your chance of getting hypertension by 2% is hardly of much interest medically.

Carl Zimmer discusses the problem on his blog, with a link to his nice essay on this topic in Newsweek.  It’s a hard problem, but even if we can’t do a lot about it now, at least the scientists can refrain from flogging their research with a lot of hype and false promises.

11 thoughts on “More failures to find human behavior genes

  1. Thanks for the link. I would just point out that there’s a big and important difference between a failed association between a gene and a phenotype versus lots of genes with tiny effects on a phenotype. The former is obviously a dead end. The latter may not lead to a quick medical fix, but it can uncover important pathways that are affected by diseases like schizophrenia. And scientists can then study that revealed pathway to understand the disease better. The trouble with the latter case is not so much the science as the promises made about it in the media (and press releases and grants). We need to re-learn how to tolerate slow progress.

    1. I agree with Zimmer here. I know people working on dyslexia genes, one of who recently published a functional paper in Plos Genetics on a common variant associated with dyslexia.

      “And once you find one of them, what can you do?”

      I suppose we could just stop looking. Or we could develop screens, or put it into a mouse model and find the mechanism and pathways. You see, dylsexia genes have a huge overlap with schizophrenia. We’re not just looking for cures, we’re looking for a better reckoning.

  2. Still, even rare and therapeutically unhelpful genes linked with autism or schizophrenia may (probably will) be helpful for understanding the etiology of those diseases. Knowing what missing genes, or repeated segments, predispose one to a disorder should help to tease out the complex set of causes of such problems.

    I realize that this is quite a different value than “the gene for ____” nonsense that was flogged there for a while, and the hype has been, sometimes still is, annoying. The genetic research is still valuable, though, just not in the simplistic manner that has too often been portrayed.

    Glen Davidson

  3. I think there’s often a false assumption that, since we can take some phenomenon and rigidly designate it with some name (“depression”, “homosexuality”), it must have a single underlying cause. It may just be an artifact of how we think about things (our tendency towards essences). Phenomenon X exists, therefore Gene-for-X must exist to cause X. The fact that X’s ontological status might be suspect is ignored.

    I don’t doubt that human behavior is determined largely (but certainly not entirely) by genes. But I seriously doubt there’s a one-to-one correspondence between genes and behaviors. The notion of an “alcoholism gene” is absurd. The notion that a range of genes, if developing in a certain way in a certain kind of environment in conjunction with other related factors, might result in a phenotype more likely to become dependent on alcohol if that individual happens to be exposed to it, seems plausible to me. But it’s not very sensational or sexy, and wouldn’t make for good headlines.

    1. …we can take some phenomenon and rigidly designate it with some name (”depression”, “homosexuality”)…

      That is because the purpose of language and human communication is to take complex phenomena, find the patterns and name it to give it a higher concept for further usage. Humans do this so automatically that the complexity gets lost.

  4. A new research effort needs to be initiated to search for the genes responsible for compulsive religious accomodationism.

    May I suggest Chris Mooney as an ideal initial research subject…

  5. Even during my relatively short stint in a major research institution I’d regularly get into *huge* arguments about dolling up the science. I’d say there’s absolutely no reason to heap on the bullshit and what is important is that the objectives of a project are clearly stated; better still have a plan for when to snuff out a project because it is getting nowhere. A negative result is still a result.

    I only wish all scientists would play that way – unfortunately I get the impression that more and more decide to play politics and the bullshit game. I have seen projects with millions poured into them each year and the results are – well, zilch – but the folks working on the project just keep their lips flapping and the money keeps pouring in. I’ve got a very bad reputation with one such group since I reviewed a number of their projects and pointed out that the chances of success were exactly 0% because they were proposing things which would violate what we currently understood of the physics of bulk materials. This was not finding and plugging a gap in an established model (such as the case of Einstein and the Newtonian laws of gravity), this was outright negation of fundamental and well-tested models.

  6. The entire approach of “pick some behavioral phenomenon, then screen the entire genome and look for correlations” seems to me to be based on a logical fallacy. Years ago authoritative textbooks on neurology had pictures of the brain with areas labeled with terms like “acquisitiveness” and “approbation”. There’s every reason to believe that the current criteria for schizophrenia (say, the DSM-IV ones) are equally naive with respect to the underlying causes of the different aspects that get labeled “schizophrenia” these days. Genetic screening for acquisitiveness, or in modern terms “business acumen” would probably yield data just as frustratingly suggestive.

    From an evolutionary perspective, these studies simply ignore the fact that genes don’t map 1-to-1 to phenotypes. You have an entire history of cellular differentiation and organismal development intervening between the events of genetic duplication and selective reproductive success to account for. Picking the phenotypical character first and trying to find “the gene” for it seems entirely backwards.

  7. A gene that, say, increases your chance of getting hypertension by 2% is hardly of much interest medically.

    I hope you don’t take this the wrong way, but I think you’ve misunderstood the reason that most researchs do Genome-Wide Association Studies these days. We know that we aren’t going to find genes that have more than an 8% increase in risk. But if we look at something like Crohn’s disease, a number of small-risk genes in different pathways, like the autophagy and Th17 pathways, have given us new understanding of how the disease starts up.

    The annoying thing about the schizophrenia paper was not so much that it failed to account for the variation in disease risk, but that it failed to turn up much in the way of genes or pathways that are involved in the disease.

Leave a Reply