« May 2007 | Main | July 2007 »

June 30, 2007

Tracing the spread of agriculture with stone-age human DNA

This week's paper is "Palaeogenetic evidence supports a dual model of Neolithic spreading into Europe" by M.L. Sampietro and others, published online in Proceedings of the Royal Society. The paper is interesting both for its findings and for its methods.

We know that agriculture spread from the Near East -- do people in Asia call this the Near West? -- to western Europe, starting around 10,000 years ago. But did this mostly involve farmers moving, or the spread of agriculture without major movement of people?

People have tried to figure out past population movements using genetic differences among modern populations, but it would help to have genetic information from people who lived thousands of years ago, as well. This is technically challenging, however...

It is possible to extract DNA from bones thousands of years old, but to get enough DNA to analyze (i.e., to determine its sequence), it needs to be amplified, using the polymerase chain reaction. PCR uses several enzymes that respond differently to temperature, and cycles through different temperatures, doubling the number of copies of DNA fragments containing the primer sequences, with each cycle. Most labs use a programmable "thermal cycler", but it can be done manually, using three different-temperature water baths.

Want to know if there are any bacteria that can oxidize ammonia to nitrate, in your soil sample? (You might, because nitrate is much more likely to leach into groundwater than ammonium is.) Use PCR with primers common to all ammonium-oxidizing bacteria. Want to know how many there are per gram of soil? Use quantitative PCR. Remember the old joke about counting cows by counting the number of legs and dividing by four? With quantitative PCR, you can count the number of copies of a gene, then divide by the number of copies of that gene per bacterial cell.

The problem, of course, is that PCR will amplify any DNA that contains a near-match to the target sequence. In particular, this includes DNA in skin cells of anyone who handled the 5000-year-old bones, as well as any contamination during sample preparation for PCR. (The problem of contamination is a concern for forensic uses as well. "If you aren't guilty, Detective, why was your DNA at the scene of the crime?")

It helps somewhat to use genes that were abundant in the original sample. This study, like other similar studies, analyzed DNA from mitochondria, distant descendants of bacteria, which are central to energy metabolism and present in multiple copies per cell. But even so, contamination was a serious concern.

An unusual strength of this study was that they had access to all six people who were known to have handled the bones and extracted the DNA samples, from the time the bones were dug up to when the PCR was run. So they could detect and correct for the 17% of DNA sequences in the samples that resulted from accidental contamination by these people. (In the future, maybe only Australian aborigines should be be allowed to work on archaeological digs in Europe, and vice versa. Or I guess they could wear gloves.)

There were still a number of complications, which experts in the field are probably discussing. But the data seem to show that the people who live in the Iberian Peninsula today are closely related to those who lived there 5000 years ago. This contrasts with a recent comparison showing that Stone Age people in Northern Europe were different from those who live there now. And the Stone Age people in Iberia were quite different from those in Northern Europe. At least the women were; mitochondria are inherited only from mothers.

Combining the genetic data with archaeological information on pottery etc. -- using multiple lines of evidence is always a good idea -- the authors concluded that migration of large numbers of people was probably important to the spread of agriculture through southern Europe, but it might have been introduced into northern Europe by only a few travelers.

I don't know enough to have an opinion about these conclusions, but I think it's cool that we can study the genetics of people who have been dead for thousands of years!

June 27, 2007

Individual and kin selection in legume-rhizobium mutualism

OK, I've been critiquing other people's work for a while. Your mission, should you choose to accept it, is to critique something I've written. It's the summary for a grant proposal I'm about to submit. It will be reviewed by ecologists and/or evolutionary biologists, but they're not likely to be specialists in legume-rhizobium symbiosis. So if something isn't clear to an intelligent but nonspecialist audience, you'll let me know, right? If you're not all too busy reading the many interesting evolution articles in today's New York Times, that is. By the way, the great Myxococcus xanthus photo in Carl Zimmer's article is from Supriya Kadam, who did her PhD with Greg Velicer and just finished a year as a postdoc in my lab.

Individual and kin selection in legume-rhizobium mutualism

Intellectual merit: The legume-rhizobium symbiosis is an ideal model system to study the evolution of cooperation. Cooperation is often undermined by short-term self-interest, despite shared long-term interests, as in tragedies of the commons.

By fixing nitrogen inside legume root nodules, rhizobium bacteria are, in effect, cooperating with other rhizobia infecting the same plant. Each rhizobium cell pays a high metabolic cost to fix far more nitrogen than it needs for its own use. The extra nitrogen enhances host plant photosynthesis, which may increase carbon supply to the nitrogen-fixing rhizobia. But what if unrelated rhizobium strains, perhaps in other root nodules on the same plant, benefit equally? These are likely competitors for limited nodulation opportunities in future years. Individual selection will not favor paying an individual cost to obtain benefits that are shared equally with competitors.

If fixing nitrogen is individually costly, it may still be favored by kin selection. Fixing nitrogen helps protect millions of clone-mates in the same nodule from host-imposed sanctions, which our earlier work showed reduce rhizobium numbers in nonfixing nodules. Still, many rhizobia risk sanctions by diverting resources from nitrogen fixation to their own reproduction or that of kin. Those that potentially increase their reproduction, by fixing less nitrogen, can be said to cheat those that fix more.

The role of kin selection in rhizobium cheating is hypothesized to differ qualitatively among host legume species. Some legumes suppress reproduction of bacteroids, the differentiated rhizobium cells that can fix nitrogen, inside their root nodules. These bacteroids are expected to cheat by transferring resources to their reproductive kin, rather than by hoarding resources themselves.

These hypotheses will be tested with experiments that 1) look for host sanctions against individual reproductive bacteroids, which would disprove the hypothesis that only kin selection favors nitrogen fixation; 2) measure resource hoarding by reproductive versus nonreproductive bacteroids, using various host species; and 3) measure fitness effects of carbon transfers to reproductive rhizobia inside nodules and in soil nearby.

Proposed methods have been successfully applied with rhizobia in our lab. These include flow cytometry, to measure resource hoarding by individual cells. Custom-made root- and nodule chambers allow exposing individual nodules, or part of a root system, to different treatments, such as rhizobium strains known to differ in transfer of resources to kin, or nitrogen-free air to prevent nitrogen fixation. Mathematical models, using data from the experiments, will analyze how the evolutionary stability of rhizobium cooperation depends on factors like the percent of root nodules with more than one rhizobium strain.

Broader impacts: This project will include curriculum development, as well as research experience opportunities for graduate, undergraduate, and high school students. The latter will be the target audience for a web site on statistics and hypothesis testing for science fair projects. Outreach via cross-disciplinary talks, reviews, and a weblog will show broader audiences how understanding evolution can help solve societal problems. These expected impacts are consistent with activities under our current NSF grant.

Soybeans and alfalfa, both of which depend on rhizobia for nitrogen, are among the top three US crops, in terms of land area. The proposed research could help in the development of varieties that selectively enrich soils with the best indigenous rhizobia, based on nitrogen actually fixed, rather than easily mimicked recognition signals. Improved crop management may also be possible. For example, the impact of nitrogen fertilizer on rhizobium evolution depends on the extent and nature of host sanctions. Improved biological nitrogen fixation by legume crops can decrease nitrogen fertilizer use on subsequent crops in a rotation, reducing nitrogen runoff and dependence on fossil fuels.

June 24, 2007

Trade-offs in defense against retroviruses

I have written about evolutionary trade-offs before, starting with early posts about trade-offs between seed size and seed number in plants, and trade-offs between the ability of insects to escape predators by flying away, versus the ability to hide from them by playing dead. I have also given some examples of the increasing use of sophisticated experimental (often molecular) methods in evolutionary biology. This week's paper combines both themes.

The paper is "Restriction of an extinct retrovirus by the human TRIM5-alpha antiviral protein" by Shari Kaiser, Harmit Malik, and Michael Emerman, published in Science (vol.316 p.1756).

Retroviruses are made of RNA, but make DNA copies of themselves that can insert into the DNA of host cells they infect. HIV, the cause of AIDS, is a well-known example, but there are many others. If DNA copies of the retrovirus are inserted into cells giving rise to sperm or eggs, they can be passed to the next generation, as endogenous retroviruses. If the DNA inserts somewhere where it turns an important gene on or off, it may kill the host. Or, once in a while, this change may turn out to be beneficial. The few beneficial changes are the ones that survive and spread, just as the few mutations that are beneficial are the ones that persist.

VWXYNot has an interesting discussion of how a creationist web site misused one of her papers as evidence of "intelligent design." She shows how shared endoviruses can be used to infer shared ancestry, providing yet more evidence that we share a recent ancestor with apes, less-recent ancestors with monkeys, etc. But that's not what this week's paper is about....

...except that the retrovirus it discusses is found in chimps and gorillas, but not in humans. Did our common ancestor somehow pass this endogenous retrovirus, PtERV, on to chimps and gorillas, but not to us? That seems unlikely, since each of them has multiple copies. How could we miss inheriting some of them, if they came from an ancestor shared by all three species?

In fact, if this difference in retrovirus infection were the only information we had about humans, chimps, and gorillas, it would make me wonder whether chimps might be more closely related to gorillas than they are to us. Then, they could both have inherited PtERV from an ancestor they share with each other, but not with us. But most modern family trees put chimps and humans together, on a separate branch from gorillas, so I'd look first for another explanation...

...and, as Deep Thought once said, "there is an answer; a simple answer." The chimps and gorillas didn't inherit the endogenous retrovirus from a common ancestor. By comparing the viral-origin DNA in chimps and gorillas, the authors estimate that they picked it up 3-4 million years ago. Although this is 1000 times longer ago than some religion-based estimates of the age of the earth, it is only one-thousandth the actual age of the earth. It was about the time when Lucy lived, and at least a million years after the last common ancestor of humans and chimps lived. So, their ancestors living at that time caught the virus; ours didn't. [See comments for additional evidence for this.]

But why didn't we? 3-4 million years ago, the (nonshared) ancestors of humans, chimps, and gorillas were all living in Africa, perhaps near each other. Could our ancestors have somehow been more resistant to this virus than theirs? That's the main question answered by this paper.

Ideally, the authors would have resurrected some 3-million-year-old virus to test. This probably would have involved crossing rope bridges, being chased by tigers (not ordinarily found in Africa, but maybe escaped from a zoo?), and stealing a bit of amber (containing mosquitoes that had bitten chimps) from a Mayan temple (don't ask me how that got to Africa). But, instead, they decided to resurrect the original PtERV virus from endogenous retroviruses in modern chimps.

Mutations over the last 3-4 million years have changed each individual copy in the chimp genome, but there are enough copies per chimp that they were able to figure out the original DNA sequence. They made a hybrid virus containing the original RNA sequence from PtERV, particularly for that part of the virus targeted by a defense protein, TRIM5-alpha, found in all apes, including humans.

They then tested whether this virus could infect cells with the human version of TRIM5-alpha, or those from various apes and monkeys. Most apes and monkeys were susceptible to the PtERV virus, but modern humans, modern chimps, and modern sooty mangabeys were resistant. Because chimp DNA is full of "fossil" endogenous retroviruses, we know that their ancestors 3-4 million years ago must have been susceptible, but they've apparently evolved resistance since then. We just beat them to it. Clever of us...

...except that all the species that are most resistant to PtERV are also highly susceptible to infection by HIV. They were able to identify a specific region in our TRIM5-alpha gene that determines specificity in virus defense. One sequence gives resistance to PtERV, another to HIV. Take your choice.

But, would it spoil some vast eternal plan if we had two different copies of TRIM5-alpha? Would that give resistance to both viruses? While we're at it, wouldn't it be nice if we had a different genetic code (for translating DNA into protein) from all other animals? We'd still be able to digest their proteins (same amino acids), but any virus that could reproduce in them wouldn't be able to reproduce in us, because it would make nonfunctional proteins (different amino acid order). Of course, if humans had a different genetic code, that would make it really obvious that we had been created separately from other life, which would undermine the need for faith... "and vanished in a puff of logic."

June 21, 2007

Opportunity cost of grad school, etc.

Rob Knop liked my previous post. The comments on his post are well worth reading. For example, someone pointed out that, even if you don't go into debt to finance grad school, there's still usually an economic opportunity cost. During the years you spent in grad school and as a postdoc, you might otherwise be paying down a home mortgage, saving for retirement, etc., not to mention nonfinancial opportunities, like starting a family.

Terence Tao also has some good career advice,. It's aimed mainly at mathematicians, but much of it is relevant to science in general.

Of course, not going to grad school could also have opportunity costs, especially nonfinancial ones. If you really want to do research, with reasonable freedom to choose your own research problems, a PhD is almost essential. It's not just a matter of getting paid to do research. A PhD-level position is also needed to get access to all the expensive equipment, and grants to pay for expensive supplies, needed in most of science today.

Because it's a much smaller commitment, I often encourage students to consider doing an MS (2 years) before deciding whether to tackle a PhD (4-8 years plus 2+ years postdoc). An MS is good background for various science-related jobs, including teaching high school science, most technician jobs, etc. Nobody ever takes my advice, though, maybe because it's usually not that hard to switch from a PhD program to an MS.

A technician, often with an MS or even BS, doesn't get to set the direction of a research program, but, in the right lab, is a full-fledged team member with plenty of opportunity for scientific creativity. There's a little less job security, relative to a tenured faculty position, but my prized technician was snapped up by another lab when I left UC Davis.

June 20, 2007

Choosing a major professor

Advice sent to an aspiring grad student, without identifying particulars.

1) Don't use a generic subject line when contacting a prospective major professor. I almost didn't open your email, thinking it was probably spam. Snail mail with a stamp and handwritten address really stands out, but email with "reprint request" will probably be opened. Ask for a PDF of a paper or two whose abstract looks interesting, but you don't have full-text access.
2) Read the papers, plus others you can find on-line or in your nearest university library. You might also consider going to a relevant scientific meeting in addition to, but not instead of, reading scientific papers. The nice thing about a meeting is that you can ask questions and talk to people from lots of different labs all in one place. The problem with talks is that if something isn't clear, it's gone, whereas with a paper you can read it twice, think about it, look up relevant definitions, etc.
3) Do the papers (including the part about weighing 5000 seeds or chasing lizards in the rain) make you think "I wish I'd done that?" If not, look for another lab, that does make you react that way.

4) Send your top few choices an email labeled "prospective grad student." Include enough detail from the papers so it's obvious you have read them with interest. (I trash emails saying "I'm really interested in your work on molecular biology.") Include information on any undergrad research, GRE scores, etc. Ask about opportunities to work in his or her lab.
5) Apply to those schools that seem most promising. Don't miss deadlines! Deadlines for fellowships may be earlier than those for admission. Having a professor who wants you really increases your chances of admission, though not necessarily of fellowship support.
6) Narrow down your choices. If someone has a research assistantship in hand, you may need to commit to them, if you want them to commit to you. Otherwise, it's common to apply 2-4 places and see who comes through with financial support. Don't make the mistake of picking the biggest assistantship over the most interesting project, but don't go into debt.
7) Try to visit each school before deciding. Perhaps because word about the shortage of faculty jobs has gotten around, the most promising grad students are actually in short supply, relative to demand. Unfortunately, demand often means professors wanting students, rather than professors having money to support students. If you're among the most promising applicants, they may invite you to a recruitment weekend and pay your travel costs. If you go, be sure you know what the several faculty nearest to your interests are doing at each school, not just the one you are most interested in. (If there's an unexpected personality conflict, will there be any other options there? Also, this makes you seem better prepared.) Talk to current students. Are they enjoying grad school, mostly? Are former students getting the kinds of job you would like?
7) Once admitted, see if you can start your research the summer before you start classes. It's a lot easier to keep experiments (or serious reading) going in the limited time between classes than it is to start them.
8) Don't get discouraged if your first experiments don't give clear results. This is normal. Figure out what went wrong and fix it, if possible, or try another approach, or try a slightly different question. If setbacks really upset you, consider another career.

June 19, 2007

Who should consider grad school in science?

This entry is inspired by "Why I got out of research" at http://vwxynot.blogspot.com/ and Rob Knop's blog entry Get out; you're not good enough , and is addressed to readers considering grad school in science.

There are more people qualified for faculty positions at research universities than there are openings. By "qualified" I mean having earned a PhD, done a postdoc, and published at least one senior-authored peer-reviewed journal article from each. By this definition, one can be qualified without necessarily being competitive in today's academic job market.

Those of us lucky enough to get such a research university position find that (as vwxynot put it):

"Even if you do make it big and get your own lab, you're suddenly responsible for your whole team's job security as well as your own. Grants depend on the quality of the researcher and their work, yes, but also on trends, fads, luck, nepotism, reputation, political interference and geography."

The importance of nepotism, politics, and geography probably varies among countries, but there's no doubt that only a fraction of good proposals get funded. And yet, getting grants is often an expectation for tenure.

So, if most PhD's won't get a research university faculty position (RUFP), then who should consider going to grad school in science?

1) those who expect to enjoy grad school itself, at least most of the time.
2) those who think they would be happy in some science-related job requiring a PhD, even if it's not an RUFP

There could be a third category: those who are confident of being among the lucky few that get those scarce faculty positions. But I suspect that these are a subset of category 1. Not everyone who enjoys grad school will get a RUFP, but those who do get one probably enjoyed grad school, mostly. They were smart and creative (and lucky) enough to pick important and interesting questions, and hard-working (and lucky) enough to answer them. They published two or more papers each from PhD and postdoctoral work, including one or more in prestigious journals. And they enjoyed this enough to make up for working long hours (reading, writing and thinking about science as well as lab or field work) for little pay.

I wrote "little pay", not "no pay." I strongly discourage anyone from running up major debts to go to grad school. It's too much risk for too little certainty of reward. If you are aiming for a research position, try to get a research assistantship or fellowship that will pay you (a little) while you work on your thesis research. If you are more interested in teaching at a 4-year college, try to get teaching assistantships. If you can't get either, at least for most years, that could be a sign of trouble. Your professor may not be good enough at getting grants or your university may be poorly funded; either way, they may not have the reputation that will help you get good postdocs and jobs. Or, if other students are getting assistantships and you aren't, that may be some indication of who's going to be most competitive for jobs later on.

Maybe someday the long-predicted scientist shortage will arrive and there will be great jobs for all qualified candidates, but don't count on it. Actually, the scientist shortage is already here in terms of important problems needing research, just not in terms of jobs and grants!

I don't think I would have regretted the years spent in grad school and two postdocs, even if I hadn't been lucky enough to get good research jobs afterwards, first with USDA and then at two great universities (UC Davis and University of Minnesota). I made an interesting discovery using some fun toys, interacted with great people, swam and played music sometimes, and didn't go into debt. But I can think of several fellow grad students and postdocs who didn't get as rewarding jobs, despite being at least as smart and hardworking as I was. I hope they enjoyed grad school, too.

If you've been thinking about grad school in evolutionary biology and this hasn't discouraged you, the grant proposal I'm submitting in July includes funding for another grad student. Overall funding success is <15%, but grant panels have liked my last three proposals....

June 18, 2007

Can plants recognize kin?

This week's paper is "Kin recognition in an annual plant", by Susan Dudley and Amanda File of McMaster University, just published online in Biology Letters.

Researchers in several countries have recently shown that roots respond differently to another root from the same plant than they do to a root from a different plant. Typically, they grow more aggressively towards a neighbor's root than towards one of their own. But what if the neighbor is a close relative?

Dudley and File compared root growth when four related plants (kin) were grown together in a pot, versus four unrelated plants. As the left side of the figure shows, roots grew less when surrounded by kin. In the "solitary" treatment, a single plant was grown in a pot one-fourth as big. Root growth in these plants was not affected by above-ground interactions with kin vs. nonkin.

Fig. 1A from Dudley and File (2007) "Kin recognition in an annual plant"

Kin selection theory predicts that plants (or animals) should compete less severely with relatives than with nonrelatives. For example, Hamilton's rule states that a gene for altruism (helping another at some cost to oneself) will spread if the cost of helping is less than the benefit to the one helped, times that individual's genetic relatedness to the helper.

Strictly speaking, relatedness should be measured relative to one's usual competitors. If a plant drops 1000 seeds (half-sibs, since they all have the same mother), and there's only enough light or water for a few to live, don't expect a lot of altruism. But a plant with a sibling on one side and a stranger on the other might be expected to behave more aggressively (in terms of root growth, etc.) towards the stranger. This assumes that plants can distinguish kin from nonkin. I would have thought two roots bumping into each other would have a hard time measuring relatedness, but this week's paper suggests that they do, somehow.

I have suggested previously (see Darwinian Agriculture) that reducing aggressive root interactions could be a key to increasing efficient use of water and other soil resources by crops. Some experimental papers that appeared to support this result may have resulted from a tendency of plants to put more roots in larger pots, even apart from nutrient levels or the presence of another plant. See Hess and de Kroon (2007) "Effects of rooting volume and nutrient availability as an alternative explanation for root self/nonself discrimination" (Journal of Ecology 95:241).

This week's paper doesn't have that problem, as the number of plants per pot was the same in the kin vs. stranger treatments. On the other hand, the plants were sampled early enough that it's not clear whether there was a benefit to plants sharing pots with kin rather than with strangers. I hope other people will repeat this work with different species, so we can see how common an effect of kinship on rooting pattern is, and whether it often affects overall plant growth.

Strange searches

Fellow science bloggers are discussing the strangest search terms that have led readers to their blogs. I haven't seen anything nearly as weird as what some report, just a few moderately strange ones:

"grants to support carpenter ant research"
"examples of week rocks" (maybe "weak" was meant?)
"opportunity cost essay" (there may be an annoyed economics student out there somewhere)
"disadvantages of shark-fin soup"
"scientists who contributed to the evolution of ecology" (only qualifies as strange because similar search came from two different IP addresses in the Philippines the same day; maybe a school assignment?)
"how plants and animals r useful to humans"
"weird scientific papers" (I think I'm offended by this one)
"this week in" (after TWI Amateur Radio, but before TWI Tech)
"Why should farmers know about evolution??" (not really a strange search, but made me wonder whether the searcher had any pre-existing bias on the question)

June 14, 2007

A junkyard for natural selection?

A major paper was just published in Nature. "Identification and analysis of functional elements in 1% of the human genome by the ENCODE pilot project" was written by a consortium involving contributions from many scientists. I will discuss a few of their more interesting findings, related to questions like "how much of our DNA doing something useful?"

They found that "the majority of its [DNA] bases are associated with at least one primary [RNA] transcript." We know that only a small fraction of our DNA gets translated into protein (by ribosomes using information from messenger RNA), but this result shows that most of it gets turned into something like messenger RNA anyway. Much of the DNA that doesn't code for protein is repetitive sequences. These could be considered "extra selfish" genes, as their main activity seems to be making copies of themselves and inserting the copies randomly. (By "randomly", I mean random with respect to the effects on the health of the individual affected. They may not be random in some biochemical sense.)

Self-replicating DNA sequences may sometimes have beneficial side effects for the species, if not for individuals. For example, a repetitive sequence may insert itself somewhere in a way that changes gene regulation, such as which proteins are made when. Usually, a random change will have a negative effect on survival and reproduction - bad for the individual -- but the occasional positive changes will increase by natural selection - good for the species. So repetitive DNA can be useful in the same sense that mutations caused by radiation or chemicals are useful. See this discussion at vwxynot.

What do RNA transcripts that don't get translated into protein, but aren't limited to self-replication, do? This is an active area of research. Some serve important functions, including regulation of other genes. In some cases, RNA that doesn't code for protein but serves other functions may be a remnant from the hypothetical "RNA world", where RNA once served as both genetic material (a role now played by DNA) and as enzymes speeding chemical reactions (a role now played mainly by proteins). For example, much of a ribosome is RNA.

Of course, if some DNA that doesn't code for protein turns out to play an important function, that doesn't disprove the hypothesis that most of it is "junk." By junk, I mean that you can change most of its sequence without adverse effects. If, on the other hand, it turns out that very little of our DNA is junk, then one of my favorite examples of "stupid design" would fall, although there are many others. (We don't expect perfection from natural selection, in contrast to a hypothetical omniscient designer. It might be harder to distinguish the products of natural selection from those of a busy committee with a lot of other projects on their agenda, and varying in competence!)

This paper doesn't show that junk DNA is rare, however. In fact, they say that only "5% of the bases in the genome can be confidently identified as being under evolutionary constraint in mammals." In other words, there is "a large pool of neutral elements that are biochemically active but provide no specific benefit to the organism." They go on to say that these "may serve as a 'warehouse' for natural selection." Any DNA sequence can serve as raw material for natural selection, I guess, just as a junkyard can yield useful materials for a tinkerer. But duplication and modification of existing functional genes (either protein-coding or regulatory) is more likely to yield something useful than a bunch of self-replicating junk DNA is. See this related discussion from Sandwalk.

"Regulatory sequences that surround transcription start sites are symmetrically distributed, with no bias towards upstream regions." I would have thought that the DNA sequence near where transcription into RNA starts would be more likely to have a role in regulation, but no. I should have known better, because my wife (who is always right, supposedly) is working on an interesting example of downstream regulation.

June 10, 2007

Rock-paper-scissors for high stakes

Chapter 3 of The Origin of Species is titled "Struggle for Existence", which Darwin uses "in a large and metaphorical sense including dependence of one being on another, and including (which is more important) not only the life of the individual, but success in leaving progeny." Differences among plants and animals in their success in leaving progeny depends on their adaptation to the physical environment, but also their interactions with each other. For example, "A plant which annually produces a thousand seeds, of which only one of an average comes to maturity [this must be true, if population size is constant], may be more truly said to struggle with the plants of the same and other kinds which already clothe the ground."

If the traits that maximized survival and reproduction were always the same, those with those best traits would quickly displace those with alternative traits. But changes in the physical and biological environment mean that no one genotype is consistently best. This week's paper is about frequency-dependent selection, where the fitness of each genotype depends on how common it is. If less-common genotypes tend to increase in frequency, no single genotype will take over.

The paper is "An experimental test of frequency-dependent selection on male mating strategy in the field" by C. Bleay, T. Comendant, and B. Sinervo, of the Universities of Bristol and California (Santa Cruz), published on-line in Proceedings of the Royal Society.

Male side-blotched lizards have orange, blue, or yellow throats. Orange males bully blue males, taking over their territories. Yellow males resemble females enough that they can sneak into orange territories and mate with females. But blue males guard their females so closely that they aren't fooled, so they outcompete yellow males, but they can't defend their territory against orange males. (There may be some lessons here for border security of large vs. small countries, but never mind.) Field observations show successive replacements consistent with these observations: blue => orange => yellow => blue and so on. It's like the rock-paper-scissors game, except that each individual male is stuck with his throat color and therefore (in practice) with the corresponding strategy.

Field observations were consistent with frequency-dependent selection. For example, yellow males can often fool orange males (who are trying to maintain large territories) but not blue males (guarding one mate), so yellow males should leave more progeny when orange males are more common. This seems to be the case, but could this correlation be due to some other factor, such as interactions between weather and temperature tolerance differences among the different types of males?

To find out, Bleay and colleagues experimentally manipulated the frequency of the different genotypes, seeding lab-raised lizards at field sites on the California coast. Some sites got mostly orange males, some mostly blue, etc. Then, as females at each site became pregnant, they were brought into the lab to lay eggs, so that DNA analyses could be used to identify who had fathered each egg. After any resulting lawsuits were settled, mothers and babies were returned to the field.

As predicted, orange males sired more babies when their competitors were mostly blue (because they could expand into their territories) than when they had to worry about sneaky yellow males, or compete with other orange males. Yellow males did best against orange and blue best against yellow, also as predicted, but these differences were much less clear. Also, up to 93% of variation in male reproductive success was unexplained, with differences in quality of territories thought to be important.

I've mentioned before that evolutionary biology is increasingly an experimental field. Most of the experiments I've discussed have been done under laboratory conditions, where we can hold everything (almost) constant except for the experimental variables being manipulated. This paper is a good example of experimental evolution under field conditions.

Figure 2 from the paper, copyright by the Royal Society and reproduced under US "Fair Use" guidelines.

June 3, 2007

Scientific controversy: dinosaur-tail soup ?

This week I want to talk about scientific controversies. In politics or religion, any difference of opinion may qualify as a controversy, which some may try to "settle" by killing those with opposing views. Most scientists would agree that unsupported opinion isn't enough to make a scientific controversy. A scientific question is controversial only if people are actually publishing data that seem to lead to different conclusions.

Two papers in press in Proceedings of the Royal Society illustrate current scientific controversies. The first is "A new Chinese specimen indicates that 'protofeathers' in the Early Cretaceous theropod dinosaur Sinosauropteryx are degraded collagen fibers" by Theagarten Lingham-Soliar (linghamst@ukzn.ac.za) and colleagues at the Universities of KwaZulu-Natal and North Carolina and the Chinese Academy of Sciences.

The title pretty much says it all. (Collagen is what gives shark-fin soup its distinctive texture, hence the title of this entry.) If the conclusions in this paper become generally accepted, how would that change our overall understanding of evolution?

The two elements of evolutionary theory that upset creationists most wouldn't be affected at all, of course. Our confidence that the universe is a million times older than Bible-based estimates, and that humans and chimps share a recent common ancestor, is based on multiple lines of evidence for each, none of it dependent on which dinosaurs had feathers, if any.

But what about the claim that birds are descended from dinosaurs? Let's see what a leading textbook, "Evolutionary Analysis", says. Page 44: Sinosauropteryx had what "some paleontologists believe are primitive feathers." Page 45: they cite several papers, one questioning this conclusion. "More convincing are [true feathers on] the dromesaur fossils." Page 553: "Luis Chiappe (1995) used skeletal characters to infer the phylogeny [family tree] of early bird lineages." The tree shown has protofeathers near the base, followed by true feathers. If the P. Roy. Soc. paper is correct, that would only require revising the earliest branches of his tree.

So this is a real controversy, but it's only a controversy about where feathers appeared in the family tree of dinosaurs and birds. As the paper says, "the wider question of whether or not birds originate from dinosaurs does not concern the present study." The main fossil evidence that they did comes from analysis of skeletons, not feathers. We don't have DNA from dinosaurs, but genetic comparisons among living species suggest that birds are more closely related to crocodiles than to mammals (Science 283:998). So birds-from-dinosaurs still seems likely.

The second paper is "Context dependence in the coevolution of plant and rhizobial mutualists" by Katy Heath (heat0059@umn.edu) and Peter TIffin, whose lab is next to mine. Among other things, this paper shows that plants infected by two different strains of rhizobium bacteria often grew less than those infected only with the worst of the two strains. This result may become controversial soon, when Toby Kiers and I publish data apparently showing that plants infected by two different strains can grow more than those infected with only the best strain. Our experiments were done with soybean, whereas theirs used a wild relative of alfalfa, which houses rhizobia in a different type of root nodule (see photos). Also, our two strains were much more different than theirs. So maybe this doesn't really qualify as a controversy, at least not yet.



Nodule photos taken in our lab (c) Inga Spence... licensing from www.alamy.com.

When there is a controversy, should it be taught? We certainly shouldn't teach a conclusion as certain when it is still (genuinely) controversial. And students should learn about some past scientific controversies, to understand how they were resolved. The triumph of evolution would be a good example. Exposure to some current controversies would be good, too, assuming teachers have time to keep up with the literature, well enough to know what has been settled (at least until convincing new data to the contrary are published) and what is still controversial. I remember Professor Spanswick, at Cornell. telling us "I found the evidence for the chemiosmotic hypothesis convincing, but always presented it as a controversy... until Mitchell won the Nobel Prize for it."