Friday, May 25, 2007

SCIENCE Questions: Asking the Right Question

In July 2005 Science magazine published a list of the top questions in science [Science, July 1, 2005]. I was reminded of this list when I attended a meeting last month because the publishers of Science were handing out a special isue devoted to those questions. There are two categories; the top 25 questions, and 100 other questions. (It was the 125th aniversary of the magazine, hence 125 questions.)

I'd like to spend some time discussing those questions because not only are they interesting from a scientific point of view but they also reveal a great deal about science journalism and the public perception of science.

The issue began with an essay titled "In Praise of Hard Questions." The author, science writer Tom Siegfried, notes that hard questions stimulate science. He says,
The pressures of the great, hard questions bend and even break well-established principles, which is what makes science forever self-renewing—and which is what demolishes the nonsensical notion that science's job will ever be done.
Everyone agrees with the sentiment behind this statement. We all know that asking the right questions is the essence of good science. We all know that hard questions challenge prevailing models. On the other hand, we also know that there is such a thing as a stupid question in spite of what your Professors might have told you. Stupid questions can mislead scientists and stiffle creativity.

The opening article quotes David Gross, the 2004 Nobel Laureate in Physics who says,
One of the most creative qualities a research scientist can have is the ability to ask the right questions.
So, what are the "right" questions to ask? In my experience, the "right" questions are not immediately obvious. As stated above, they often challenge the prevailing dogma and this means that in the beginning they are dismissed as silly questions. Over time, the idea that this is a good question becomes more and more acceptable until finally it starts to stimulate active research.

What this means is that at any given point in time the "right" questions are only known to a few scientists on the cutting edge. These are ones who have begun to understand that the old questions aren't working any more. The vast majority of scientists will be sticking with the paradigm that's about to be overthrown. If you were to take a vote they would overwhelmingly dismiss the very questions that need to be asked.

Now, don't get me wrong. This is the way science is supposed to work. We all know that 99.9% of all attacks on orthodoxy deserve to be dismissed. The wonderful thing about science is that the 0.1% of "right" questions will almost certainly bubble to the surface. The real tricky part is picking out that 0.1% in advance.

So, if you were the editors of Science magazine how would you identify the important questions in science without falling into the trap of reinforcing orthodoxy and missing those very questions that a small group of experts are beginning to pay attention to? One way would be to seek out those experts and ask their opinion. This seems to be what is being advocated in the lead article where Tom Siegfried says,
Science's greatest advances occur on the frontiers, at the interface between ignorance and knowledge, where the most profound questions are posed. There's no better way to assess the current condition of science than listing the questions that science cannot answer.
But, Science magazine did not ask the experts at the frontiers. The actual procedure is explained in the editorial accompanying the July 1, 2005 issue. According to editors Donald Kennedy and Colin Norman, here's what they did.
We began by asking Science’s Senior Editorial Board, our Board of Reviewing Editors, and our own editors and writers to suggest questions that point to critical knowledge gaps. The ground rules: Scientists should have a good shot at answering the questions over the next 25 years, or they should at least know how to go about answering them. We intended simply to choose 25 of these suggestions and turn them into a survey of the big questions facing science. But when a group of editors and writers sat down to select those big questions, we quickly realized that 25 simply wouldn’t convey the grand sweep of cutting-edge research that lies behind the responses we received. So we have ended up with 125 questions, a fitting number for Science’s 125th anniversary.
The "right" questions were selected by editors and science journalists. I'm going to examine some of these questions in the next few days, concentrating exclusively on biology questions. Let's see how well they did when asked to identify the top "hard" questions in science.


  1. An important question conspicuous in its absence is "how can complex genetic diseases evolve"?

    We "know" that traits can only become common in a species gene pool if they are positively selected for. There are multiple single gene disorders (aka Mendelian). Usually they are sporadic, or weakly selected against, or as in the case of sickle cell trait provide a benefit, explaining their persistence and expansion in the gene pool.

    So how can complex genetic diseases evolve? The simple answer is that they can't, and that what we perceive to be a "complex genetic disease" actually is a "complex genetic feature", that we do not fully understand. Sort of like anaphylaxis. Anaphylaxis exhibits complex genetics, and can kill you dead. Is it a "disease"? No, a robust immune response is a "feature" that is positively selected for. Anaphylaxis is what happens when that immune system is triggered to robustly. Evolution has selected organisms with robust immune systems to minimize non-reproduction from the SUM of an immune system that is not strong enough, and one that is too strong. It is not that anaphylaxis is a "feature", rather an immune system that can support anaphylaxis is superior to one that cannot. Anaphylaxis is an extreme immune response to an extreme immune system challenge.

    All "well-evolved" systems will minimize the non-reproduction caused by, and prevented by that "well-evolved" system. Stress responses are among the most conserved pathways. Presumably as ancient pathways, they are "well-evolved". Presumably under some circumstances they will prevent death, and under slightly more extreme conditions will cause death, just like anaphylaxis.

    I suggest that what are presently perceived to be "complex genetic diseases", are actually "complex genetic features", which are being operated in regions outside of where they prevent death.

  2. daedalus2u says,

    We "know" that traits can only become common in a species gene pool if they are positively selected for.

    I don't understand what you're trying to say but I know one thing. You are wrong about evolution. Most genetic traits become common in a species by random genetic drift and not by natural selection.

    It's quite possible for deleterious mutations (genetic diseases) to persist in a population, especially if they are recessive. Do you have a problem with that or are you making some other point that I don't understand?

  3. I agree that recessive deleterious traits can persist. But if they start out as a single mutation, how do they become common? I agree that they could become common by "chance", via genetic drift. But for many genes to become common (in the absence of a bottleneck) such that those genes are not "neutral", but cause a specific and characteristic disease, strains my credulity.

    The point I am trying to make (quite inarticulately), is that some genes that are associated with what we term "complex genetic diseases" may have been selected for under different circumstances where they were advantageous.

    One thing I am working on is regulation of ATP consumption. I will send you a poster I presented on it. A point I try to make in it, is that pathways that conserve ATP in the short term (as in running from a bear) might be deleterious in the long term.

    We know that ischemic preconditioning reduces ATP consumption and prolongs cell life during subsequent ischemia. But obviously there is something bad about being in the ischemic preconditioned state for too long, or cells would evolve to be in that state continuously.

    A pathway that conserves ATP (i.e. turn off the proteasome) under one condition (ischemic preconditioning) might be good. But turning off the proteasome is a reliable way of inducing accumulation of protein inclusions characteristic of Alzheimer's and Lewy body neuropathies.

    That is what I think is happening in many of these "complex genetic diseases", common pathways beneficial under some circumstances are being operated outside the range where they are beneficial.

  4. The scientist do ask the right questions