Monday, December 14, 2009

Personalized Genomics--for the investigator!

GWAS, or genomewide association studies, that claim that major common diseases are due in major part to identifiable genetic variation, have largely been discredited relative to the heavily over-hyped promises that led to its massive global funding levels. Many variants associated with studied diseases have been found, but they typically account for only a small amount of risk, even though many large studies have been done. The promise of near immortality that was made by those who would benefit from the grant largesse for GWAS has not exactly been met.

Of course, being discredited is only in the sense of its being used by the same perps as justification for even bigger and more of the same, only under a different name (e.g., biobanks and personalized genomic medicine). Those in the know know, and knew, that this kind of research was hyped from the beginning. We've discussed why in many earlier posts so won't do that now.

A recent commentary in The Economist points out these disappointing results as a dirty little secret that, the author says, will be revealed in 2010 when analysis of all the GWAS in the pipeline is completed, and it turns out they don't explain much. What then for disease genetics? The commentary doesn't really say, but it does end with a sales pitch for a different point: we'll be getting whole genome sequences on large numbers of individuals, and this will lead us to understand group (read 'race') differences. He acknowledges that this might have some negative consequences, but is basically sanguine about it.

That approval might seem curious, until you see that the author is an evolutionary psychologist. That's a field built on an insistent belief in genetic determinism, and the commentary ends with absolutely standard determinism and essentially the justification for continuing more and more DNA-based approaches to everything and anything, in this case racial differences.

The Economist is a business publication so the logic and layout of the commentary is just what you might expect: raise a specter of investment risk, but then explain why it's OK to do it in a way that the author approves of and could benefit from.

The author is correct to raise the potential for a new era of Darwinian racist determinism, a revisit to the bad days of eugenics based on new claims of group traits dressed in modern genetic technological rhetoric. But it is totally self-serving to discredit excessive genetic determinism in relation to current GWAS approaches for disease, and then credit the very same kinds of approaches to every other kind of trait.

The problem is that nobody wants to take a serious look at the issues. There can be no denying that genes affect every kind of trait, normal and otherwise. There can be no doubt that different groups of people, no matter how you define them, will have at least some differences in almost any trait: how could two groups be exactly identical in anything, after all? There will always be some traits for which any two groups you might choose will differ so substantially that not only is the average different but there is relatively little overlap (for example, skin color between Swedes and Somalis). We can't be in denial about that.

On the other hand, to suggest that whole genome sequences will have widespread use in identifying variants with important deterministic effects is simply to ignore all the GWAS evidence that we currently have, and there's a ton of it, that even though most traits have a substantial genetic contribution to their variation, most of the individual contributing variants are too weak to identify or enumerate.

One problem with evolutionary psychology is that the traits that are chosen for study are those of clear cultural interest. The problem with that is that such traits are exactly the ones on which societal discrimination has been based. To say that we'll have to learn to deal with uncomfortable truths is to say that, whether meant that way by the author or not, some races will have to face the reality that they are rather deficient in certain areas (as a group!).

That is because difference is quickly equated with importance. There is little hesitation, even though defining behavioral traits is notoriously difficult and obviously culture-loaded. Why else would race and IQ (high class behavior of certain groups that just by chance include the investigators) or sexual behavior and innate sports ability (i.e., an animalistic kind of 'talent' of some others) be the objects of study? Why not shape of lung lobes or tarsal bones, or more abstract but less socially loaded traits?

These are not new points, but humans have a history and it's a mistake to ignore it. It would be nice if we could address the real issues more seriously without political correctness or the hyped vested-interest and funding-driven predominance, or glee of the elite who dream of and would be able to afford to design their Einstein children. And it would be good also if we could face the potential for human harm if we continue to accept the idea that you are not what you eat but what you inherit. We've seen how that can loose the forces that eat at people about other people they may not like or may have to compete with, and we know where that leads.

The actual scientific issues are subtle, but are buried beneath all sorts of biases, vested interests, and self-fulfilling preconceptions.

No comments: