« December 2008 | Main | February 2009 »

January 30, 2009

Research funding and economic stimulus

This is what I sent our Senator:

Please support the higher NSF research grant funding in the House stimulus bill. Scientific research has long-term benefits, but it may also have a higher multiplier effect than generic tax rebates for rapid stimulation of the US economy. Most grant money goes to pay poor graduate students, who will spend it all locally on food and rent. The remainder goes for scientific equipment and supplies which, unlike consumer products, are mostly made in the US.

Even if we get this burst of funding, it won't solve the long-term problem: the supply of well-qualified researchers with good ideas is increasing faster than research funding, and this trend seems likely to continue. I'm not sure this scientist surplus is a problem for society as a whole, though, unless it's causing people with great research potential to waste their lives as doctors (helping only a few people rather than the millions that benefit from each scientific advance) or Wall Street crooks.

Inferring details of past evolution from DNA is tricky

Last week I discussed one of many papers that use the ratio of protein-changing to "neutral" genetic changes, along the branches of an evolutionary tree, to infer past natural selection. This week's paper presents data calling that approach into question. This does not necessarily undermine the overall conclusions of last week's paper, which were based on a variety of methods, including testing the actual performance of mutant proteins.

"Hotspots of biased nuclear substitutions in human genes" was published in PLoS Biology by Jonas Berglund and colleagues. I am not a molecular biologist, so will just summarize their main points. The paper is open access.

Most of our DNA does not code for proteins. Some of the noncoding DNA is known to have important regulatory functions. But there is lots of DNA whose function, if any, is unknown, but which is nonetheless highly similar among species, as if any change was lethal. Except, when someone tried deleting this DNA, a bit at a time, most of the deletions were not lethal or even (as far as they could tell) harmful. I discussed this work earlier.

Anyway, much of this noncoding DNA that differs little among most species is different in humans. Could these differences be what makes us different from other apes? Quite possibly. But are all these human-vs.-chimp differences important? Maybe not. An unexpectedly high fraction of the changes from the ape ancestor we share with chimps involved a change from A bound to T (a weak bond) to G bound to C (a strong bond). Unless noncoding DNA with stronger bonds is consistently better somehow (and only in humans!), this suggests that these changes are caused by some DNA-specific process and not by natural selection. In other words, these changes occurred whether or not they were beneficial, just as mutations do. Could similar AT=>GC changes have changed protein-coding sections of DNA?

The researchers compared 10,238 genes in humans, chimps, and macaques...

They focussed on those where DNA changes led to protein changes, often thought to indicate natural selection. Like the apparently-important noncoding regions of DNA, these protein-coding regions had more AT => GC changes than expected. Apparently, this can be caused by high rates of recombination, where stretches of DNA come from both parents.

So maybe recombination led to AT => GC changes which resulted in more protein-changing DNA changes than expected, which gave a possibly false signature of past natural selection. If so, then identifying which genes are actually key to uniquely human traits will be more difficult. I'm sure this will not be the last word on this subject.

This study does not cast any doubt on the common ancestry of humans and other apes. In fact, their specific conclusions depend on this common ancestry.


January 23, 2009

Staying ahead in the evolutionary arms race with viruses

This week's paper uses molecular methods to reveal new details of the evolutionary arms race between primates, including humans, and viruses. "Protein kinase R reveals an evolutionary model for defeating viral mimicry" was published in Nature by Nels Elde and colleagues in Seattle.

Protein kinase R (PKR) is an important defense against viruses in many species, from humans to yeast. When it detects a virus inside a cell, it activates eIF2-alpha, which shuts down protein production in that cell. With protein production blocked, the virus can't replicate and spread to other cells. Viruses, however, have evolved counter-measures. These include molecules that resemble eIF2-alpha. These molecular mimics interact with PKR and prevent its normal defensive activity.

Viral epidemics can be a major cause of death, so we expect populations to evolve PKR resistant to the eIF2-alpha-mimics produced by viruses. Can we find evidence of such evolution in primates?

The authors compared the DNA sequences of PKR genes in humans, other apes, and monkeys. Using a family tree for these species, they could infer the DNA sequence in the common ancestor of chimps and bonobos, their common ancestor with humans, our common ancestor with gorillas, and our common ancestor with orangutans. They did the same thing with the monkey family trees and then a combined primate family tree. See the Tree of Life web page for details on the primate family tree.

Once they had DNA sequences for each species and various ancestors they could infer changes in the PKR gene along each branch of the tree. A few changes might occur at random, but they were interested in nonrandom changes: those due to natural selection, presumably imposed mainly by viruses. They detected selection using the ratio of nonsynonymous changes (those where a change in DNA actually changed the PKR protein) to synonymous changes (those where DNA changed without changing the protein). This widely used method is based on the fact that some amino acids, the building blocks of proteins, are coded for by two or more different triplets of three DNA bases. Random mutation is equally likely to change any of these bases. But natural selection tends to eliminate mutants with the "wrong" amino acids in their protein. If the current protein is working well (resisting the virus), then the "wrong" amino acid is anything different than the current one. But if the current PKR protein isn't working, perhaps because the virus evolved, then natural selection will eliminate those with the current version and replace them with mutants whose PKR works better.

Nonsynonymous (protein-changing) changes in PKR outnumbered synonymous ones along most branches of the primate family tree. The most extreme discrepancy (22:0 !), indicating strongest selection by viruses, was along one of the branches shared by monkeys but not humans. For the branch leading to our common ancestor with gorillas and chimps, the ratio was only 1.1, indicating only weak selection by those viruses that interact with PKR. Maybe those monkeys should start covering their mouths when they cough.

How did viruses respond to evolution of resistance in their hosts? The researchers did a similar analysis of evolutionary changes in viruses. Again, nonsynonymous changes were more common, showing that primates impose selection on virus populations to evolve new counter-measures against PKR.

Earlier, I mentioned that yeast cells also make PKR. This made it possible for the researchers to test different PKR versions against different viruses. (Apparently, nobody at the Yeast Liberation Front volunteered as test subjects.) When a yeast cell is infected by a virus, yeast PKR shuts down protein production, just as in primate cells, resulting in easily detectable slowing growth. In yeast, our PKR (and that of other apes) works much better against some viruses than monkey PKR does. The researchers went on to identify key amino acids responsible for resistance and susceptibility.

Interestingly, eIF2-alpha itself doesn't seem to have evolved much. This evolutionary arms race has mainly involved viruses evolving new mimics of eIF2-alpha and primates evolving new versions of PKR, resistant to the viral mimics, but still able to activate real eIF2-alpha.

How soon (and in what countries) will we see attempts to increase virus resistance in humans, using germ-line modification of PKR?

January 20, 2009

Making scientific careers family-friendly

Like most US scientists, I have been distressed by how the Bush administration twisted or ignored scientific evidence on global warming, abstinence-only education, etc. and I am optimistic that things will improve with President Obama. But some problems will be difficult to solve. Today's NYT has an article on under-representation of women in science. Relative to their fraction of the population, my impression is that African-Americans are even more under-represented, but that doesn't seem to get as much attention. Or maybe it's just that the perceived or actual reasons for under-representation are different, calling for different remedies.

One problem that may deter some women from pursuing research-university faculty positions is that these jobs are so demanding they make it difficult to also raise children. Many of my female colleagues are doing both well, but it's got to be hard. A proposed solution from the article:


Dr. Mason and other legal experts suggest that President Obama might be able to change things significantly for young women in science — and young men — by signing an executive order that would provide added family leave and parental benefits to the recipients of federal grants, a huge pool of people that includes many research scientists.

This is the sort of idea I would expect from a "legal expert" unfamiliar with the realities of scientific careers. I'm not sure it would do any harm, but I doubt that it would help much. The basic problem is that only about 10% of NSF grants are funded. This has implications that undermine the value of the proposed "solution":
1) The worst work-vs.-family struggle is usually during the assistant professor years, when someone is trying to earn the long-term job security of tenure. Once an assistant professor gets an NSF grant, tenure is usually assured -- they must have had a strong publication record to get the grant, plus (with only 10% success rates) the grant itself is evidence of scientific stature -- so they can afford to spend more time on family matters, with or without a presidential order.
2) But even then they can't relax that much, if they want to make enough research progress to get the grant renewed in 3 years. It would be great if a presidential decree could somehow put more hours in the day, but it can't. Relief from teaching would help, but who's going to teach those classes? Assistant professors who don't have grants yet and are already overworked?
3) With funding rates so low, how much time should one spend on research? More time than 90% of one's colleagues, who are making the same calculation, spend! This is not a recipe for work/family balance.

Somehow, we need to increase the percent of grant proposals that are funded. Putting more total money into research would help in the short-run, but it would encourage universities to create even more faculty positions, to get a shot at that money. This increase in the number of people competing for grants would bring success rates back down. We should increase research funding, because research benefits society, but a research-funding bubble is not a solution to professors' problems.

Spreading the wealth, making those who already have one grant ineligible (or less eligible) for another, might be a better approach. This would be less likely to encourage excessive growth of faculty numbers. If we assume that grants are now going to the best scientists, then spreading the wealth might mean a net transfer to slightly less able (but still world-class) scientists. But the top scientists could then focus on their best ideas, rather than dividing their time among many projects. The best ideas of second-tier scientists may be better than the second-best ideas of top scientists. Under the present system, the second-best ideas of top scientists may still win grants, because those scientists have published more papers, but have they published more per grant dollar? If not, then spreading the wealth would be a better way to invest in research, even apart from its benefits to work/life balance.

Another common problem is that of two scientists married (or otherwise attached) to each other and unable to get faculty positions in the same city. This was the situation my wife and I faced for many years, before I decided to retire early and move here. Even if each partner is highly qualified, what are the chances of two simultaneous openings at the same place, with one being first choice for one position and the other first choice for the other? If at least one of the partners is really outstanding (a future Obama cabinet member, say), some universities will offer a package deal, but that's far from certain. Any ideas?

January 19, 2009

Safe-crackers have vaults in their cells

This is the most amazing thing I've seen in awhile. Vaults are abundant in our cells and bigger than ribosomes and apparently I'm not the only biologist who had never heard of them. They seem to be important in defense against bacteria, but nobody understands them in detail yet, apparently.

Liberal education, basic research, and Neal Stephenson's "Anathem"

In today's NYT, Stanley Fish laments the demise of liberal education, which, he says

"is distinguished by the absence of a direct and designed relationship between its activities and measurable effects in the world..."
This could perhaps also be a criterion by which basic research is distinguished from applied research. Another characteristic that basic research shares with liberal education is that each offers fewer career opportunities than there are people who want to pursue it as a career. This led me to suggest, in a previous post, that the only people who should consider grad school in science are:


1) those who expect to enjoy grad school itself...
2) those who think they would be happy in some science-related job requiring a PhD, even if it's not a professorship
Even the second option is becoming competitive enough that I might delete "requiring a PhD" and suggest doing a Masters first. My rationale is that an MS will often make you more competitive for jobs that strictly require only a BS, whereas having a PhD may make you "overqualified" for positions that don't require an advanced degree. Potential employers may assume you will quit as soon as you can find a more-academic position. You may be in the first category if your undergrad research was published or if you frequently went to graduate seminars as an undergraduate.

But back to Stanley Fish's "Last Professor." I see "effects in the world" as the only reason to spend tax money (or allow tax exemption) to support liberal education and basic research. The physicist who discovered the electron once proposed a toast

"To the electron, may it never be of any use to anybody."
I disagree with the sentiment, but this example shows how research whose practical value is not immediately obvious can eventually have a big impact. Similarly, I am happy to pay taxes to support the teaching of literature and philosophy because I assume those courses will develop the thinking skills of future citizens and leaders. Barack Obama might be an example.

Maybe Fish and I would argue about the "direct and designed" part. For example, I would consider actual evidence as to whether liberal education really does lead to better thinking skills to be relevant to the discussion, just as it is relevant to the discussion of how we should allocate tax money between basic and applied research. As one minor example, if you compare people with similar income, which is a better predictor of buying a more expensive house than one can afford: not studying philosophy or not studying economics? If some philosophy courses protect against future stupidity better than others, can we figure out what the most-effective courses have in common?

We need to ensure that practice in writing about literature improves students' ability to write clearly about other topics (from trade policy to disease prevention) rather than making their writing more obscure. If I may use an example from computer software, which should be understood by anyone who can claim cultural literacy today, students need intellectual tools analogous to spreadsheets, which can handle a wide range of problems, not single-purpose tools like tax-preparation software.

One problem is that the effects of liberal education on an individual's decision-making skills may take decades to become fully evident. This can be equally true of practical results from basic research. If we take the long view, we might end up with something like the monastery-like "concents" in Neal Stephenson's "Anathem", where "avouts" pursue philosophy and math in moderate to extreme isolation from the outside world. Isolation from outside influence is seen as so important that, when one of these thinkers leaves the concent to help solve some problem in the outside world, he or she is never allowed to return.

I really enjoyed the book, as I have Stephenson's previous ones. It made me want a bigger chalkboard for my office. It made me think about the difference between "valuable new information" and "distractions", revisiting questions like "should I go to this 3-day scientific meeting, or spend 3 days in the library?" or "would adding more species to our soil microcosms make them more or less useful in understanding plant-microbe interactions in the field?" And it's an entertaining story.

January 10, 2009

Grooming and groveling reduces meerkat stress

This week's paper is "Manipulating grooming by decreasing ectoparasite load causes unpredicted changes in antagonism" by Joah Madden and Tim Clutton-Brock, published on-line in Proceedings of the Royal Society.

Social animals often groom each other, removing parasites (lice, ticks, fleas, etc.) from places that may be hard for the beneficiary to reach, while also relaxing them, reducing heart rate and lowering concentrations of stress hormones. This stress-reducing function is sometimes considered more important than the hygenic one, but that made me ask, in an earlier post, why natural selection hasn't made lower (healthier) levels of stress the default. If stress is harmful and grooming isn't always available, why depend on grooming to reduce stress?

Madden and Clutton-Brock tested the interaction among parasites, grooming, and social interactions by chemically treating meerkats to eliminate parasites. Half of the meerkats got treated with an antiparasite chemical while the other half got water as a control, but parasite levels dropped in both. Maybe the chemical rubbed off on nontreated individuals.

With lower parasite loads, they only groomed each other half as much. Self-grooming also decreased. Because the social function of grooming was thought to be important, they expected increased fighting, bullying, etc., but that didn't happen overall. The biggest overall change in social interactions was a near-doubling in "unprompted submissions", basically groveling. Looking more carefully at pairwise interactions, they found some increases in dominance interactions, which were often followed by grooming of the dominant individual by the subordinate. So lower parasite levels led to less grooming, which led to stress, leading to cranky dominants, who were then placated by grooming they didn't really need (at least not for parasites) or general groveling. They concluded that

"for meerkats, grooming primarily serves a hygienic function, and secondarily provides a facultative response to antagonism, functioning to halt persistence or escalation of antagonism. This contrasts with its proposed function suggested for primates as a pre-emptive strategy that inhibits the initiation of antagonism by establishing or reinforcing a social structure."

January 2, 2009

Ford Denison, amateur scientist

My NSF grant will run out soon, so I get to spend the year in which we celebrate the 200th anniversary of Darwin's birth and the 150th anniversary of The Origin of Species as an amateur scientist, like Darwin himself. I'm not as smart or as rich as he was, but I do have imaginative and hard-working students and much better equipment.

I'm working on two grant proposals and several papers while dreaming of getting back to writing my book, so no detailed paper analysis this week. But Nature is highlighting 15 major papers on evolution they have published in the last few years.