Thursday, March 17, 2011

An army of ants

We've posted many times over our views on what we believe are the excesses, and even disingenuous self-promotion, of the atOmic bombs that are becoming so predominant in the science landscapes:  too many projects promulgated on a prevailing, perhaps ephemeral, view that we must use high and costly technology to measure absolutely everything on absolutely everyone (GWAS, biobanks, proteomics, connectomics, exposomics,....).  You can search MT for those posts.

It isn't that science is bad, it's that it has become a system for capturing funds as much as for solving the nominal problems on which the largess is provided.  We raise various objections, but of course we  are fully aware that we and others who attempt to raise these issues are ants relative to the elephants, and a small army of ants at that.  It might be debatable how much merit there is in our views,  but there is no debating how much leverage those views have:  none!  The elephants, those with strong vested interests, certainly and inevitably trample the ants.  Only something like a true budget crunch (not the proposed 5% cut in NIH funding that has generated so much loud bleating by the research interests), might force a major change.  At least at first, such a budget cut will only intensify the competition among the elephants and the trampling of ants.  More likely, things will evolve as people get bored with lack of definitive results and move on to other things.  Whether there will be an atOmics meltdown is unpredictable, but in our environment, it won't  movie things in a more modest direction--in claims or costs.

As a general policy, there is a way to rectify the situation, up to a point at least.  It is to mandate that studies can only be of some maximal size, no bigger.  Small samples are weak samples.  They can't detect everything.  But what they can detect is what's clear and most important!  They leave the chaff behind.

If we weren't endlessly detonating atOmic bombs, scattering so much statistical debris across the landscape that we can't see if anything remains standing, what small-size studies do is tell you clearly and replicably what really matters.  Restraint focuses the mind.

And those are the topics--for example, major disease-related genes--that can be addressed by standard scientific methods and, hopefully, problems that can actually be solved (if science establishments actually want to solve problems rather than ensure that they remain unsolved and still fundable).  And once they are solved, what remains will then perhaps become more easily detectable and/or worth studying.

If we did this, we'd have more, if smaller, grants to go around, to more people, younger people not yet irradiated by atOmic bombs, who might have cleverer or more cogent ideas.  And where complexity matters, but its individual components can't be identified in the usual way, it will focus these younger and fresher minds on better ways to understand genetic causal complexity, and the evolution that brings it about.

Yes, we are but ants, and we know very well that it's not going to happen this way.

No comments: