# News this Week

Science  17 Aug 2007:
Vol. 317, Issue 5840, pp. 880
1. GRANTS MANAGEMENT

# NSF Survey of Applicants Finds a System Teetering on the Brink

1. Jeffrey Mervis

A new survey by the National Science Foundation (NSF) offers an inside look into the minds of those seeking funding from the $6 billion agency. It reveals a merit-review system that's bending but not breaking under increased strain. The survey, conducted for NSF by Booz Allen Hamilton, a consulting firm headquartered in McLean, Virginia, explores the effects of a stressful period (2000-2006) in which the agency boosted the size of grants but held the number of awards steady and had to reject an increasing number of applications. The result has been a lose-lose situation: Program managers and reviewers have had to work harder, and principal investigators have found it tougher to obtain a grant. The good news is that most of the 24,378 scientists who filled out the online survey (a 56% response rate) think the system is thorough and fair. Sent last fall to everyone who submitted a research proposal to NSF in the past 3 years (more than half were also reviewers), the survey also paints a picture of the typical applicant. He's someone (three-fourths are men) who underestimates his chances of success but would have a go regardless of the odds. He needs the money primarily to keep his lab intact and is prepared to try and try again if his initial application is rejected. He's reviewed up to a half-dozen proposals for NSF in the past 12 months, sometimes cutting corners, and thinks that few contain potentially transformative ideas. Yet he believes his own research, if funded, stands a good chance of shattering the existing paradigm in the field. NSF officials say the survey results and accompanying report by an internal committee (nsf0745) will help them address the growing burden on the staff and the community without compromising the merit-review process. NSF Director Arden Bement is particularly concerned about what he calls “churn”: excellent proposals that keep getting revised and reviewed because funding rates are so low. “They clog up the system,” he says, “and they're a burden on everyone: program officers, reviewers, and scientists, who should be doing research rather than rewriting their proposals.” But altering one part of the process has a ripple effect. Responding to complaints that grants were too small, for example, NSF has boosted the size of the average award by 34% since 2000. But those larger grants swallowed up nearly all the 44% growth in NSF's budget over the period. At the same time, the number of applications rose by 47%. With the number of awards holding steady, funding rates plummeted from 30% in 2000 to 21% last year. That decline, in turn, caused some program managers to limit the number of applications from scientists and/or their institutions. About one-quarter of NSF's 350 funding opportunities now impose some restrictions on applicants; an infrastructure or training initiative might allow only one application per school, for example, and some program solicitations might do likewise for principal investigators. Other programs have begun to require preproposals, with only those that make the cut submitting a full proposal. The survey found that about one-third of scientists have submitted preproposals, but that the jury is still out on whether the approach saves scientists' time, improves the quality of the final proposal, or curbs the number of applications. Bement doesn't expect the practice to spread but adds that it can be a useful tool. “I think the directorates have been judicious in selecting those limitations that are effective without restricting access,” he says. NSF officials don't know the reason for the increased demand, and the data don't support one pet theory. The budget of the National Institutes of Health (NIH) has basically been flat since a 5-year doubling ended in 2003, and there was speculation that applicants would turn to NSF for help. But that's not the case: The number of submissions to NSF's biology directorate has risen by roughly the same rate through thick and thin NIH budgets. In addition, only 11% of survey respondents said that their decision to seek NSF funding was greatly affected by funding cuts elsewhere. The survey also found that one in six scientists decline an invitation to review proposals, either by mail or in person at NSF's Arlington, Virginia, headquarters. Nearly two-thirds of those who say no cite a lack of time. Some 70% who said no reported being asked to do more reviews, and 36% say they are devoting less time to each review. But 15% also cite a conflict of interest—perhaps as a potential collaborator—despite the agency's attempt to spot such conflicts ahead of time. In addition, one in six admit that the quality of their reviews has suffered as the number increases. Bement says that bigger NSF budgets would solve many of these problems, and that he's encouraged by congressional support to date for President George W. Bush's proposal last year to double NSF's budget over 10 years. Otherwise, he says, the situation will only get worse. “My fear is getting into an irreversible downward spiral, with funding rates so low that people stop submitting and we can't find enough reviewers,” he says. “Right now we're on the precipice.” 2. SOUTH AFRICA # Firing of AIDS Policy Champion Seen as Setback 1. Robert Koenig PRETORIA, SOUTH AFRICA—Last week's dismissal of South Africa's deputy health minister has angered HIV/AIDS treatment activists and exacerbated concerns among researchers and clinicians that the government's commitment to a new 5-year public health strategy on AIDS may be slipping. The deputy minister, Nozizwe Madlala-Routledge, had been the most outspoken government advocate of a stronger HIV/AIDS campaign and a champion of a plan, adopted this spring, that calls for cutting in half the HIV infection rate (nearly 19% of the adult population) and quadrupling the number of infected persons receiving antiretroviral (ARV) therapy by 2011. South Africa has the world's largest number of HIV-infected persons, an estimated 5.5 million, of which about a quarter-million now receive ARV therapy. In firing her, President Thabo Mbeki claimed that Madlala-Routledge hadn't been a team player and had made an unauthorized trip. Pediatric AIDS researcher Hoosen Coovadia of the University of KwaZulu-Natal in Durban said he was concerned about Madlala-Routledge's removal because she had “worked hard at reestablishing trust and confidence with civil society, including its much-derided scientists.” AIDS researchers and the government have often been at odds over issues such as the Health Department's long-running support for nonstandard medical remedies. Coovadia and others worry that the deputy minister's departure could endanger the implementation of the HIV/AIDS plan, the result of lengthy negotiations that ended earlier this year (Science, 23 March, p. 1651). Francois Venter, who heads the Southern African HIV Clinicians Society, called the firing “awful” because Madlala-Routledge had convinced activist groups of “a new era of cooperation in tackling a difficult issue.” The AIDS and Rights Alliance for Southern Africa, a network of 20 AIDS and human-rights groups in southern Africa, saw the dismissal as “a political reprisal for the deputy minister's outspokenness and truthfulness.” Mbeki's office denied any change in HIV/AIDS policy and made public his dismissal letter to Madlala-Routledge, which expressed his concern about her “inability to work as part of a collective.” The president cited her failure to receive prior approval to travel to Madrid in June to attend an International AIDS Vaccine Initiative (IAVI) conference. When the deputy minister learned that her request had been rejected, she flew back to South Africa without attending, explaining later that Mbeki told her that “he believed that politicians have nothing to say in a conference of technocrats and researchers.” Even so, several government officials from around the world attended the session. In a statement, IAVI said planners “thought it was important to include a high-level representative of South Africa's health ministry,” given the extent of the nation's HIV/AIDS epidemic. At a fiery news conference in Cape Town last week, Madlala-Routledge expressed hope that her dismissal would not weaken the new strategy: “People are waiting to see if the spirit of unity we had achieved will remain intact. We really do need a united front.” Government spokesperson Themba Maseko told Science that the dismissal will not impact policy, insisting that the government “reaffirms its commitment to the strategic plan and is doing everything in its power to ensure that we meet its targets.” But many activists were skeptical. Several groups have complained, for example, that after 5 years, South Africa's mother-to-child HIV prevention program has not been expanded beyond 30% coverage. The Health Department counters that it is making progress. Activist groups are planning to review their strategies. Mark Heywood, director of the AIDS Law Project in Johannesburg, says more pressure needs to be exerted “to make sure the AIDS plan is implemented fully.” 3. NEUROSCIENCE # Enzyme Keeps Old Memories Alive 1. Greg Miller Many substances interfere with memory, as any hung-over partygoer can attest. But although booze and drugs can disrupt the making of new memories (such as the embarrassing antics at last night's party), they leave older memories intact. Neuroscientists think this is because, after a time, memories become wired into the brain in a way that makes them harder to wipe out: Long-term memories, in the generally accepted view, are maintained by structural changes to the synaptic connections between neurons. A study on page 951 adds to other recent evidence that may challenge, or at least complicate, this view. A team of neuroscientists reports that injecting a drug that blocks an enzyme called protein kinase Mζ (PKMζ) into the cerebral cortex of rats makes the animals forget a meal that made them sick weeks earlier. The findings suggest that the continuing activity of PKMζ is somehow necessary to maintain long-term memory, something that's not predicted by most current hypotheses on the mechanisms of memory. The work also hints at the possibility of future drugs that could tinker with memory—for therapeutic uses or for boosting brainpower. “This is a somewhat mind-blowing conclusion,” says David Glanzman, a neuroscientist at the University of California, Los Angeles. Enzymes similar to PKMζ are known to be important in early stages of memory formation, Glanzman says, but most researchers had thought that these compounds were not needed to sustain memory once synaptic changes—such as the growth of new synapses or the strengthening of existing ones—had occurred. In the new study, Todd Sacktor of State University of New York Downstate Medical Center in Brooklyn and Reut Shema and Yadin Dudai of the Weizmann Institute of Science in Rehovot, Israel, first gave rats an unsavory meal to remember. They added a novel taste, saccharin, to the rodents' drinking water, then 40 minutes later gave a nausea-inducing injection of lithium. Rats normally avoid saccharin-laced water for weeks after such an experience. But in the days that followed, when the researchers injected the PKMζ-blocking drug, called ZIP, into the insular cortex, where taste memories are thought to reside, rats lost their aversion to saccharin within 2 hours and did not recover it in the 25-day study period. Moreover, giving the injection up to 25 days after the nauseating meal erased the aversion. Yet when given prior to saccharin exposure, ZIP had no effect on the ability to form new memories—the rats still learned to avoid the flavored water. The findings extend a 2006 study by Sacktor and colleagues that showed that blocking PKMζ in the hippocampus of rats reverses long-term potentiation, a change in neural signaling thought to underlie learning and memory, and causes the animals to forget their fear of a place where they'd previously received a shock (Science, 25 August 2006, p. 1141). The new work suggests a broader role for PKMζ as a key component of different types of long-term memories stored in different parts of the brain, Sacktor says. Going forward, it will be important to figure out how specific ZIP's memory-erasing effects are, says Lynn Nadel, a neuroscientist at the University of Arizona in Tucson. “It's possible that ZIP erases all learning, no matter how old,” Nadel says. But if the drug works more selectively, it could one day have clinical applications, he says. For example, researchers and clinicians have been looking for compounds capable of eliminating the painful memories of trauma survivors (Science, 2 April 2004, p. 34). The flip side is cognitive enhancement, adds Richard Morris, a neuroscientist at the University of Edinburgh, U.K. “The next step might be to find out whether augmenting the action of PKMζ can help sustain memories for longer than occurs normally.” 4. ENDOCRINE DISRUPTERS # Controversy Continues After Panel Rules on Bisphenol A 1. Jocelyn Kaiser A federal advisory panel has poured itself the proverbial half-glass of water after digesting the latest studies on the human health risks of an estrogenlike chemical used to make plastics. The chemical industry has proclaimed that the panel's verdict last week confirms its contention that bisphenol A is safe. But environmentalists say the report has been tainted by industry and downplays the risks. Away from the fray, some scientists say the panel's comments about the chemical's effects on the developing brain represent heightened concern compared with previous formal reviews. Bisphenol A is found in everything from some beverage and baby bottles to the linings of food cans. Small amounts can leach out into food, and most people likely have detectable levels in their blood. These parts-per-billion levels are well below the safe dose set by the Environmental Protection Agency (EPA). In 1997, however, reproductive biologist Frederick vom Saal and others at the University of Missouri, Columbia, found that very low levels fed to pregnant mice could enlarge the prostates of their male offspring. Industry studies couldn't replicate the results, but a review concluded that the results were valid (Science, 27 October 2000, p. 695). Since then, other scientists have reported low-dose effects in rodents. Some findings have raised alarms, such as an increase in chromosomal abnormalities in the eggs of mice, discovered after bisphenol A leached from plastic mouse cages (Science, 4 April 2003, p. 31). Epidemiology studies have linked bisphenol A and human health problems, such as breast cancer and early puberty. In the first formal U.S. review of bisphenol A, the National Toxicology Program's (NTP's) Center for the Evaluation of Risks to Human Reproduction formed a 12-member expert panel to review more than 500 studies. Controversy accompanied the first meeting of academic, federal, and industry scientists in March: An environmental group pointed out that the contractor preparing a draft report had done work for chemical companies. NTP fired the contractor. Amid the scrutiny, the panelists last week stuck with a previous decision to set aside many academic low-dose animal studies that had injected the chemical, because the results failed to account for the detoxifying effects of ingesting bisphenol A. The panel also questioned the relevance of an enlarged prostate because it wasn't shown to be a precursor of cancer, NTP staffers say. The bottom line: On a five-point scale of concern ranging from “negligible” to “serious,” the panel had “minimal concern” for potential effects on the prostate during development or accelerated puberty and “negligible concern” that bisphenol A is harming adults. At the same time, the panel had “some concern” about risks to fetuses and children because of about a half-dozen reports of subtle neurological effects in rodents from minute doses of bisphenol A, including changes in brain structure and a reduced interest among males in exploring their environment. “They're very scattered. No single study has been replicated,” but “it raises some flags,” says panelist Jane Adams, a neurotoxicologist at the University of Massachusetts, Boston. This area topped a list of research needs from the panel. Some answers could come from the upcoming National Children's Study, said panel chair Robert Chapin of Pfizer. Vom Saal and others are upset that the panel's conclusions differ from a statement a week earlier by 38 scientists in the online Reproductive Toxicology expressing “great cause for concern” about its potential human health risks. In contrast, a European risk assessment in January found no risk to human health at current exposure levels. Reproductive biologist John Vandenbergh of North Carolina State University in Raleigh, a co-signer of the Reproductive Toxicology statement as well as a member of the NTP panel, believes that the latter's lower concern reflects added expertise in epidemiology and human health. “I think there is a human risk. What we're trying to do here is define what the risk is,” he says. After receiving public comment, the NTP itself will weigh in on the risks of bisphenol A. 5. MOLECULAR EVOLUTION # Resurrected Proteins Reveal Their Surprising History 1. Robert F. Service It's easy to see evolution's handiwork writ large—just compare a marigold and a musk ox. With a little work, scientists can also pick out subtle differences in the proteins of related organisms. Now researchers in Oregon and North Carolina have gone a step further: retracing the steps by which two proteins diverged from their ancient common ancestor. In a paper published online in Science this week (www.sciencemag.org/cgi/content/abstract/1142819), the researchers report that they resurrected a protein from ancient fish that swam the oceans some 450 million years ago and then worked out the protein's atomic structure. By comparing the protein to more modern versions and doing some deft detective work, they crafted something like a movie of the sequence of key mutations that enabled the ancestral protein to take on new modern functions. “It's a beautiful piece of work,” says David Haussler, a molecular evolution expert at the University of California, Santa Cruz. Haussler says the work underscores how chance mutations that seemingly add little value initially can help set the stage for major evolutionary leaps. “For me, what is so exciting is seeing the dynamics of evolution play out on the molecular level,” Haussler says. The study was led by Joseph Thornton, an evolutionary biologist at the University of Oregon, Eugene. Thornton and his postdoctoral assistant Jamie Bridgham turned to a pair of closely related protein receptors for two steroids, hormones that govern a wide range of metabolic functions in humans and other vertebrates. The first receptor, known as the glucocorticoid receptor (GR), binds to the hormone cortisol, which helps regulate the body's response to stress. The second, called mineralocorticoid (MR), binds to deoxycorticosterone (DOC) in fish and to aldosterone in humans and other tetrapods. The two receptors derived from a common ancestral protein, known as the ancestral corticoid receptor (AncCR), some 450 million years ago. Last year, Thornton's group reported they had resurrected that ancient protein by comparing the gene sequences of 60 steroid receptors from present-day organisms (Science, 7 April 2006, p. 97) and using evolutionary relationships to compute the most likely sequence of the ancestral protein. The team then synthesized a gene for that protein, cloned it into mammalian cells in culture, and regenerated it. Biochemical tests showed that, like the modern MR protein, it readily bound aldosterone and DOC, and it had a more modest sensitivity to cortisol. “So it's the GR's function that emerged with evolution,” Thornton says. That suggested that the gene for the ancient MR protein was duplicated in the ancestral organism—when fish with cartilage-based skeletons were the only vertebrates on Earth—and evolved to become GR, Thornton says. For its current study, the group set out to unravel those evolutionary steps. They started by teaming up with structural biologist Matthew Redinbo of the University of North Carolina, Chapel Hill, and his postdoctoral assistant Eric Ortlund, who crystallized the AncCR protein and used a powerful x-ray synchrotron to determine its three-dimensional structure, the first of an ancient protein. They then studied precisely how the structure differed from structures of MR and GR. The researchers worked their way back up the tree of life from AncCR, reconstructing other ancient proteins from ancestral organisms that likely lived around 440 million and 410 million years ago. Whereas the older of these retained the MR-like behavior of AncCR, the more recent protein was found to have a more GR-like function. Comparing computer models of the likely structures of these two proteins, Thornton's team found 37 amino acid changes. In all modern GRs, the amino acids at five of those sites are always the same—a clue that those mutations are vital to the ancient protein's GR-like function. When the researchers inserted the mutations into AncCR, no single one of them transformed it into a GR-like receptor. But one pair of mutations did turn the ancestral receptor into one that prefers cortisol to aldosterone and DOC. “It didn't get us all of the way there, but it was a huge shift,” Thornton says. Thornton's team suspected that adding the other three conserved mutations into the protein would create a more modern GR. But it didn't happen. Instead, the mutations completely killed the receptor. Reviewing the atomic structures of the ancient and modern proteins, the researchers found another pair of mutations that likely helped reinforce the portions of the protein destabilized by the trio of “killer” mutations. When they added these mutations before the other three highly conserved mutations, they got a protein that works like a modern GR. “That really blew our minds, that we were able to predict the functional effect of these mutations that occurred over 400 million years ago,” Thornton says. The re-creation sheds light on exactly how evolution works at the molecular level, Haussler says. For example, it shows that evolution first made big leaps in changing the AncCR receptor to its GR-like state and then made smaller tweaks in refining its binding abilities. It also shows that mutations having no effect on a protein's function can set the stage for key mutations. If the history of life could be replayed like a movie, Haussler says, such stage-setting mutations probably wouldn't always occur when needed. And that means the movie would likely have a different ending. 6. MARINE GEOLOGY # Support Is Drying Up for Noah's Flood Filling the Black Sea 1. Richard A. Kerr It was a hypothesis of biblical proportion: In 1997, two marine geologists proposed that a cataract with the power of 200 Niagaras filled the Black Sea 8400 years ago, driving Neolithic farmers into Western Europe and inspiring the story of Noah's flood (Science, 20 February 1998, p. 1132). Now, 10 years later, a torrent of research is still arriving, and almost all of it comes down hard on any Black Sea flood. The proffered geologic evidence for a catastrophic event was misinterpreted, researchers write in more than 1000 pages of papers, and the raft of data collected around the Black Sea the past 50 years all points to a gradual filling starting thousands of years earlier. Putting it mildly, “the majority wisdom would be against” a flood, says geologist Norm R. Catto of the Memorial University of Newfoundland in St. John's. He is editor-in-chief of Quaternary International, where a new collection of papers appears. But a small cadre of researchers maintains that the flood hypothesis is sound and hints that definitive evidence is in the offing. The latest surge of research comes in 15 papers in the June issue of Quaternary International and 35 papers in a 971-page book, The Black Sea Flood Question, published earlier this year by Springer. The new papers agree that the archaeological record shows no sign that people living around the Black Sea 8400 years ago fled from a rapidly advancing sea. “At this point, there just isn't any evidence for something big and catastrophic” in the archaeological record, says archaeologist Allan Gilbert of Fordham University in New York City, an editor of the new book. One apparent piece of supporting evidence—the discovery of the remains of a wood-and-mud house littered about with stone tools 91 meters beneath the Black Sea (Science, 22 September 2000, p. 2021)—has not panned out. “It looks peculiar,” says Gilbert, but there's no sign it's anything more than a random bunch of rocks and sticks. Then there's the geologic evidence used to gauge the depth and salinity of the Black Sea over the past 15,000 years. The tools include drowned beach dunes, seismic probing of bottom muds, oxygen isotopes, microscopic fossils, and pollen. Citing such data, the originators of the flood hypothesis—longtime marine geologists William Ryan and Walter Pitman of Lamont-Doherty Earth Observatory in Palisades, New York—have argued that 10,000 years ago, the Black Sea was a modest-sized lake lying perhaps 100 meters below its current level. It was cut off from the salty Mediterranean Sea, they say, because sea level was too low to spill through the Bosporus. When melting glacial ice raised sea level, the Black Sea basin filled up in a geologic instant about 8400 years ago. But most of the authors of the book consider that scenario “a myth,” e-mails another editor of the book, Valentina Yanko-Hombach of the Avalon Institute of Applied Science in Winnipeg, Canada. The Black Sea, she says, was never that low, and it rose gradually over millennia. Not so fast, says coastal geologist Liviu Giosan of Woods Hole Oceanographic Institution in Massachusetts. Giosan, a native Romanian who studied oceanography in that Black Sea coastal country, has reviewed the new book for Quaternary Science Reviews and found it wanting. In particular, he says, the “vast amounts of data” collected around the Black Sea by Soviet scientists and researchers from former communist countries around the Black Sea are suspect. He distrusts much of the carbon-14 dating of lake levels and is frustrated by the traditional lack of access to primary data. As a result, he writes, “many conclusions of studies presented in the book should be considered with a grain of salt” until researchers buttress them with more-direct measures of lake level. Lately, geologists are in fact looking at more direct sea-level gauges, as well as geologic indicators of past flow through the Bosporus. Richard Hiscott and Ali Aksu of Memorial University of Newfoundland and colleagues have reported several signs that the Black Sea filled slowly and gently. In the Quaternary International issue, they describe a core from the shallow Black Sea shelf that contains sediment laid down beneath tens of meters of water when Ryan would have that spot high and dry. On the Black Sea floor just north of the Bosporus, they have mapped old beach ridges and lagoons formed as the lake level slowly rose. And south of the Bosporus, they found a delta built by outflowing waters 10,000 years ago, when Ryan's scenario would have the Black Sea totally cut off. Ryan and a half-dozen colleagues disagree. “I've found myself following those who criticize the flood,” says Ryan, “getting my own data from their sites, and in every case finding a very different story.” Where Hiscott and Aksu find a delta built by Black Sea outflow, Ryan and colleagues find a delta formed by a nearby river, as they will soon report in Marine Geology. As for the putative beach ridges, Ryan says unpublished coring results show they are actually mud brought in by bottom waters still flowing through the Bosporus today. He also speaks of an as-yet-unpublished description of “an extraordinary debris fan” right where the flood would have dumped its gougings. Erosional features there bear a striking resemblance, he says, to those created by catastrophic outbursts from glacial lakes (Science, 20 July, p. 307). Despite the continuing debate, Giosan, who has worked with Ryan, is guardedly optimistic. “There is momentum toward solving this,” he says. “Maybe we'll solve it someday.” 7. ALAN KRENSKY INTERVIEW # Drawing a Map for the Twenty-Seven Divisions in NIH's Army 1. Jocelyn Kaiser BETHESDA, MARYLAND—Alan Krensky should have packed a flak jacket when he moved here. The director of the National Institutes of Health's (NIH's) new planning office says he's already been battered with criticism as he settles into his job, which is to bring more order to the sprawling$29 billion biomedical research agency. “One thing I've learned is, any flag you raise, someone will attack it from one side or the other, and sometimes both sides at the same time,” says Krensky, who came to NIH from Stanford University School of Medicine last January.

Yet Krensky seems undaunted. As the official director of the NIH Off ice of Portfolio Analysis and Strategic Initiatives (OPASI) since 8 July, he faces two main responsibilities. The first is to oversee implementation of the Roadmap, the $483 million set of crosscutting initiatives started 4 years ago by NIH Director Elias Zerhouni that has been criticized as taking money from investigator-initiated grants. Second, OPASI will look across NIH's research portfolio and find better ways to manage it while respecting the turf of its 27 institutes and centers. Observers say those are formidable challenges. “A lot of this is going to boil down to how well he gets along with institute directors,” says former NIH director Harold Varmus, now president of Memorial Sloan-Kettering Cancer Center in New York City, who first called for more central planning at NIH 6 years ago in Science. In an interview, the 55-year-old immunologist talked about his experiences at Stanford, where he headed a$500 million children's health initiative. He also outlined some ideas for OPASI, which will soon become part of a division under Krensky's direction that Congress created to encompass NIH offices such as those on women's health and AIDS.

Krensky is now overseeing midcourse reviews of the Roadmap projects and crafting more rigorous policies, such as time limits, for the next round, which will launch this fall with projects in epigenetics and the microbiome. His edited remarks follow.

Q: What is your response to the claim that the Roadmap takes funding away from investigator-initiated R01s?

I think when people attack it or have concerns, they really misunderstand what it's all about. I don't view the Roadmap as monolithic at all. What attracted me is that it was a new way of looking at things. One term that Elias [Zerhouni] used is a learning laboratory. It's a place to ask questions, a place to do things outside the box.

These are 5-year programs, potentially renewable for 10. And that's what I like. If people don't use it or don't like it, it should disappear. And the idea that it sundowns makes it different from anything else here.

The scale of many of the early projects was very large. I'm not against big projects. If people say, “This has been terrific for my research,” it's a success. That being said, I'm very interested in other areas that the Roadmap can do and test. I believe we want more pilots. I believe in small things.

Q: Is OPASI, with its $483 million research fund, essentially a 28th NIH institute? That is a real problem. Originally, the common fund was a tap from all the institutes and centers. So it was a co-ownership, a co-op essentially. Last year and this year, Congress made it a line item so it now has an income stream like any other [institute]. And so what's central to this is, we're not an [institute], it's about working together. This goes back to what I call cajoling, but we are really working with the entire community—and that's both internally and externally—to develop these programs, and we're much more like the concierge than the bosses. Q: What is the “fingerprinting” [of individual grants by disease topic] about? [NIH is] basically building a system that will scan the grants and determine the portfolio [for a particular disease]. Now there are all kinds of ramifications of that. Any number in any [disease] area will be different because it was obtained in a different way. I'm expecting the community and Congress to respond to the numbers, and the communications are going to be very important. The most important part is, if someone else came and ran the same thing, they'd get the same number. Q: Doesn't Congress want NIH to address redundancies? In my time here, of all the words that give me the greatest fear, the word “redundancy” comes to the top of the list. The first point made to me is that redundancy is important in science. If you do an experiment once and no one proves that it's correct, it's sort of worthless. So we have to have a certain amount of repetition. I think redundancy is in the eye of the beholder. For some people it resonates tremendously, for others it is a very worrisome topic. One investigator said to me, “Am I the redundant one?” So in our R01 community, people worry about that. But at the same time, if people find out they're doing the same thing, they should know about it. I can't prescribe how they respond to that or what they do, but it's of interest. 8. ORNITHOLOGY # Gambling on a Ghost Bird 1. Erik Stokstad Top-notch ornithologist wagered their reputations by declaring that the ivory-billed woodpecker still exists. Two years later, the odds of its survival appear long ON 28 April 2005, John Fitzpatrick told the world what he had been keeping secret for more than a year. At the headquarters of the Department of Interior in Washington, D.C., flanked by two Cabinet secretaries, Fitzpatrick announced a conservation miracle. The majestic ivory-billed woodpecker—an emblem of southern old-growth forests that was last seen during World War II—still persisted in the Big Woods of Arkansas. “In the world of birding,” he said, “nothing could have been more hoped-for than this Holy Grail.” It was an extraordinary claim and a rare piece of good news in conservation. It was also a crowning achievement for Fitzpatrick, director of Cornell University's prestigious Lab of Ornithology, who had fielded probably the most intense search for a bird ever. The 14-month stealth mission yielded several eyewitness sightings, sound recordings, and a video, published online in Science that day. “We have conclusive proof that the ivory-billed woodpecker has survived into the 21st century,” Fitzpatrick declared in a video released by Cornell. Private donors and federal agencies opened their wallets. The world celebrated a second chance to save the awe-inspiring bird. And yet after more than 2 years of herculean efforts and sometimes vituperative debate, indisputable evidence of the bird's existence has not emerged. Fitzpatrick still believes his team saw an ivorybill, although he never did himself, in the Big Woods in both 2004 and early 2005, and he speculates that it has either flown elsewhere or died. Skeptics think the mesmerizing ivorybill was never there to begin with and that the Cornell team mistook other woodpeckers and overinterpreted a blurry video. “Why would you announce … one of the biggest things ever in North American ornithology and not have concrete, irrefutable evidence?” asks Mark Robbins of the University of Kansas in Lawrence. To many critics, this is a story of good intentions gone awry and the power of belief, amplified by secrecy. A top-notch team of scientists was misled by hope, it seems to them, and buoyed by confidence that more searching would bring the definitive photo. Fitzpatrick and his colleagues reject those explanations, defend their objectivity, and say they have no doubts or regrets. Now, as the U.S. Fish and Wildlife Service (FWS) begins to assess the efficacy of the searches it funds, most birders and ornithologists seem resigned that even if an ivorybill was in Arkansas in 2004, the chance to save the species is past. “I want to hope against all odds,” says James Bednarz of Arkansas State University in Jonesboro. “But my scientific logic says it's deep in the vortex of extinction.” ## Fleeting glimpses The largest woodpecker in the United States, the ivorybill (Campephilus principalis) lost practically all its old-growth habitat when loggers cut down the bottomland forests of the southeastern United States. As the birds became scarce in the 1880s, ornithologists and birders raced to shoot the survivors for their collections. By the 1960s, most ornithologists were convinced the ivory-billed woodpecker was extinct. Yet every few years, a hunter or birder would announce a sighting. Experts assumed that they were misidentifying a pileated woodpecker (Dryocopus pileatus), a large species still abundant in the bottomland forests. In 1966, bird author John Dennis reported seeing an ivorybill in a swamp in east Texas. He swam naked through the water and managed to get a close look, yet no one believed him. Even a respected scientist caught the fever. George Lowery Jr. of Louisiana State University (LSU) in Baton Rouge, past president of the American Ornithologists' Union, brought two photographs of ivory-billed woodpeckers to AOU's annual meeting in 1971. Lowery believed that the photos, taken by an acquaintance, were real, but other ornithologists thought the birds looked like posed specimens. His reputation was tarnished. “I wish now that I had said nothing about these birds,” he later wrote. None of this boded well for David Kulivan, a forestry student at LSU. He spotted what he thought were two ivory-billed woodpeckers while turkey hunting near the Pearl River on 1 April 1999 (not an auspicious day of the year to report seeing ivorybills). He recounted the sightings to ornithologist James Van Remsen, curator of birds at LSU's Museum of Natural Science, who was persuaded enough by Kulivan's account to organize a search. Zeiss Sport Optics funded a well-publicized effort in 2002. Cornell also mounted a small expedition, led by Fitzpatrick. There may have been no one better placed to save the ivory-billed woodpecker than Fitzpatrick, who is shrewd, ambitious, and decisive. “Fitz never goes halfway on anything.” says Frank Gill, who retired as chief scientist of the National Audubon Society in New York City. “He can move mountains in a way that no other ornithologist can do.” A Harvard graduate who went on to a Ph.D. at Princeton, Fitzpatrick bushwhacked through the Amazon in the 1970s and '80s, discovering seven new species of birds. He made an even bigger mark studying endangered Florida scrub jays and helping to create a national wildlife refuge to save scant remaining habitat. In 1983, as curator of birds at the Field Museum in Chicago, Illinois, he was awarded AOU's highest prize for research. After a month in the Pearl River, neither group had found anything. Late-night TV comedian Jay Leno mocked the search by reading a newspaper headline: “Researchers fail to find extinct bird.” Eventually, the Louisiana Ornithological Society dismissed the Kulivan sighting. Still, the Cornell team won kudos from other researchers for its cautious analysis of their sound recordings, which turned out to have captured gunshots, not the distinctive double-knocks made by ivorybills. Despite heading home empty-handed, the experience fired up Fitzpatrick. “The chance to be there was a dream come true,” he says. ## Secret mission Another opportunity arose just 2 years later. Fitzpatrick was in his office at 8:30 a.m. on 1 March 2004 when Tim Gallagher came in, wild-eyed. Gallagher, an avid birder who edits Cornell's Living Bird magazine, had just returned from the Cache River National Wildlife Refuge in Arkansas, where he and a friend had seen an ivorybill. Fitzpatrick grilled him for details and finally asked: “What are the chances that the bird you saw was not an ivory-billed woodpecker?” Gallagher replied, “I'm absolutely positive that this bird was an ivory-billed woodpecker.” Fitzpatrick immediately sent Gallagher back to the swamp with a top graduate student. Then in mid-March, he convened a meeting of the Sapsuckers, a crack team of birders from Cornell that competes in the World Series of Birding. Several days later, they were tromping and paddling through the Arkansas swamp. But during that week, the only woodpeckers they saw were pileateds. The team was frustrated, and most of them had to return to their day jobs at the lab. But Fitzpatrick decided to press ahead, having great confidence in Gallagher's sighting. “I have to put my faith in those people able to separate fact from fiction,” he says. He was also convinced that if he didn't act, the bird would truly go extinct. There had been no previous exhaustive searches, he points out. Cornell had the tip, the resources, and the gumption. “Nobody else had the balls to do it,” Fitzpatrick says. He insisted on secrecy—a decision that would later bring the team criticism for being insular and insufficiently skeptical. Fitzpatrick feared that if word of the search got out, “the place would become Coney Island with birders piling in all over the place.” Ultimately, some two dozen police officers were ready to protect the habitat after the announcement, but there was no onslaught. The Nature Conservancy, which was involved in the search, had its own concerns. It had been buying land to conserve bottomland hardwood forest and feared that news of the search would drive up prices. More volunteers arrived, all signing legal confidentiality documents. The cover story for curious locals was that they were doing a biological inventory for The Nature Conservancy. The bird was code-named Elvis. Between 5 and 11 April, there was a flurry of sightings, all by lone, amateur observers. Concerned about the lack of corroboration, Jeffrey Wells of Cornell, the logistical manager, decided to double up the observers. After that, there was just one more sighting. On 25 April, David Luneau—an electrical engineer at the University of Arkansas, Little Rock, who participated in the Pearl River search—and his brother-in-law filmed a 4-second glimpse of a bird fleeing a tree. It has become without doubt the most analyzed bird video in history. Like the others, Fitzpatrick was initially disappointed by the video's quality. Although the team was convinced from the sightings that the bird was there, and they had intriguing recordings of double knocks and “kent” calls, they wanted solid evidence—clear photos or video or a nest hole that would convince skeptics. As Luneau has said, “If you have something like a picture or video or sound recording, … then others are able to make up their minds based on science rather than on their feelings on how much they believe somebody.” With time running out on the 2004-2005 season—the leaves would soon be emerging on the trees and it would be impossible to see anything—Fitzpatrick and the others began planning for the next field season. Fitzpatrick raised about$4 million in cash and pledges for what would become the largest ornithological search in modern times, coupled with a concerted effort to conserve the ivorybill's habitat. As a board member of The Nature Conservancy, Fitzpatrick had rubbed elbows with the likes of Henry Paulson, the former CEO of Goldman Sachs and now U.S. Treasury Secretary. Paulson is an avid birder and, with his wife, donated money to support the search.

At the same time, Fitzpatrick was communicating with the Department of the Interior, where he also had connections. James Tate, science adviser to Secretary Gale Norton, was a former assistant director of the Lab of Ornithology. “We wanted to get as much buy-in from the government to put money into the conservation of this area as we could,” Fitzpatrick says. He also decided they would not announce the finding until they had tangible evidence and a paper accepted for publication, probably at the end of the 2005 field season.

To the team's disappointment, nothing better than the video turned up—despite efforts including hoisting observers 25 meters above in a cherry picker. “It became clear that, in all probability, we were not going to obtain any more video evidence anytime soon,” says Martjan Lammertink, a woodpecker expert who joined the team that season. By February 2005, Fitzpatrick recalls, he realized that “we need to begin to act as though the Luneau video plus sightings plus sound is going to be enough.”

The team went back to the Luneau video. The more they looked, the more convinced they became that it could not be a pileated woodpecker. The wings had a white trailing edge. The wing beats seemed very fast. And the size of the bird, measured as it perched on the tree, was much too big. To bolster their argument, the group took crude models and reenacted the escape flight of the bird, albeit with stiffly flapping wings. They filmed at the exact spot Luneau had taken the video, using the same camera. “The most parsimonious and logical conclusion is that it is probably an ivory-billed woodpecker,” Lammertink says. Fitzpatrick e-mailed Don Kennedy, editor-in-chief of Science, about submitting a paper. In an editorial published with the paper, Kennedy recalls that he “responded in a New York second!”

The manuscript went out for reviews in early April and was scheduled to be published in mid-May. But on Monday, 25 April, the story leaked. In preparation for an announcement by the Department of the Interior, Tate quickly flew to Florida and drove in a raging rainstorm to meet Fitzpatrick at the Archbold Biological Station to evaluate the evidence. “I'd seen George Lowery and John Dennis have their reputations ruined by naysayers,” Tate says. “I did not want that to happen to the secretary, or to me.” Tate, who had studied a kind of woodpecker called the yellow-bellied sapsucker, came away convinced.

Editors at Science rushed the final production of the paper so that it could be published online, along with the video and the recordings, before the news broke in the media. “Science wanted to do this with an embargo and make a splash,” Fitzpatrick says. It worked: Stories ran in 459 U.S. newspapers, 174 television shows, and 43 radio shows. At the press conference, Interior and the U.S. Department of Agriculture announced joint funding of \$10.2 million for the conservation of the ivory-billed woodpecker and its habitat. Fitzpatrick and a few others were whisked back to Cornell on a private jet.

## The powder keg explodes

The announcement, Gill recalls, provided a spark that “hit the powder keg of hope and expectations in a way that was just unprecedented. Once it got started, it really got out of control.” The town of Brinkley, Arkansas, nearest to the sightings, went wild with promotion. Some 70 experts and officials, including a brigadier general, joined the federal recovery team—a record number. Many scientists were also swayed. At first, “I was completely accepting,” recalls Geoffrey Hill of Auburn University in Alabama, who became more skeptical after taking a close look at the video. “It was Science, it was the Lab of Ornithology, and it was Fitz.”

But others say they looked at the video with dismay. “I was worried right from the start,” says Noel Snyder, a retired FWS biologist. He and a few others privately expressed concerns to Fitzpatrick about the strength of the evidence. But they kept quiet, not wanting to rain on a joyful and highly publicized parade.

Jerome Jackson was among the early skeptics. An ornithologist at Florida Gulf Coast University in Fort Myers, Jackson is no stranger to ivorybills, having seen more than 300 museum specimens and written a detailed history called In Search of the Ivory-Billed Woodpecker. And in 1986, when FWS convened a meeting to discuss declaring the ivory-billed woodpecker extinct, Jackson argued against it and conducted a small search.

Jackson and three other scientists prepared a paper for PLoS Biology, arguing that the Luneau video showed a pileated woodpecker. “All we wanted to do was have everyone go, 'Wait a minute!' before any more money got spent,” says co-author Robbins. “We didn't want to see precious conservation dollars wasted on something that might not be there.”

This made the Cornell team and its sponsors nervous. Not long after The New York Times reported the existence of the skeptical but not-yet-published paper, Jackson says, Tate called Jackson on a Saturday night and told him to “back off.” Tate denies that and says he just wanted to discuss Jackson's criticisms. “My concern was that the skeptics would destroy our opportunity, destroy that second chance to get the biological information of what the birds needed,” Tate says.

Days before publication, and after writing a rebuttal, the Cornell team offered to play the critics additional, unpublished recordings that hadn't been fully analyzed before the submission of the Science paper. The recordings convinced co-authors Richard Prum of Yale University and Robbins that at least two ivorybills were living in the Big Woods. They withdrew the paper on 1 August, saying they didn't want to undermine conservation efforts. (In retrospect, now that it's clear the recordings are not solid evidence, they regret the move. “I blinked,” Prum says.)

But Jackson, who had been out of town and unreachable, still thought that the doubts needed to be aired. In a long, invited article published in The Auk in January 2006, he accused Fitzpatrick's team of “delving into 'faith-based'ornithology and doing a disservice to science.” In a March 2006 response in The Auk, Fitzpatrick's group charged that the Jackson article was “a series of factual errors and poorly substantiated opinions.” Jackson, they implied, was “compromising science with sound bites.”

After another round of rebuttals commenced, Fitzpatrick confronted Jackson during an August 2006 meeting in South Carolina and asked him not to publish. Jackson recalls Fitzpatrick heatedly telling him, “You are going to be independently responsible for the extinction of the ivory-billed woodpecker because you are preventing me from raising money for conservation.” Shortly thereafter, Fitzpatrick contacted Jackson again and offered co-authorship on a future paper if Jackson would pull his letter. “That's not how I operate,” Jackson told him. Fitzpatrick says he wanted to focus on the bird and avoid another unproductive exchange: “It was not my desire to prolong and underscore resentments and personal disagreements.”

The tone was much more restrained in a Technical Comment and response published in Science on 17 March 2006 (p. 1555). Like the authors of the stillborn PLoS paper, David Sibley, who wrote and illustrated The Sibley Guide to Birds, thought the Luneau video showed a pileated woodpecker. In the Comment, Sibley and three co-authors argued that the white on the wings is the underside of a pileated's wings, not the trailing edge of an ivorybill's. Moreover, several frames show a black trailing edge, like a pileated's. The white on the back of the torso, which Fitzpatrick had called “clearly evident,” was actually “a vague pale blur” of just a few pixels. In addition, they asserted that the size estimate wasn't valid, because what Fitzpatrick identified as a perched bird was instead already in flight.

The Cornell team has stuck to its guns.

Since then, other papers, one published in March in BMC Biology and another in The Wilson Journal of Ornithology in June, also found the video and acoustic evidence unconvincing. “It's all sort of evaporating,” Snyder says. He and others aren't interested in rehashing the Luneau video; they would rather see new evidence. It hasn't arrived. The second massive search, during the 2005-2006 season, also came up dry.

## Stalemate

Fitzpatrick and Lammertink say they will remain convinced that the Luneau video shows an ivory-billed woodpecker until they see evidence that a pileated could look and fly like that. “Have we boxed ourselves in? Maybe so, but I don't think it's so unusual in science,” Lammertink says.

Skeptics, on the other hand, won't believe in ivory-billed woodpeckers until they see clear proof, such as a roost tree where birds can be repeatedly observed. In the absence of more evidence, the American Birding Association in Colorado Springs, Colorado, continues to list the bird as “probably or actually extinct or extirpated.” The majority of birders appear to be agnostic. In an online poll by Birding magazine, published in April, 75% responded that the ivorybill might or might not exist.

So what made Cornell so sure? Hill thinks it is the weight they attached to the video. “In retrospect, the Luneau video may loom as one of the most unfortunate things to ever happen to the Laboratory of Ornithology,” he writes in his book, Ivorybill Hunters. Without it, he speculates, the Cornell team probably would have interpreted the sightings more cautiously. Instead, they threw themselves into a highly involved analysis of murky data. “It was cast as a scientific analysis of these pixels,” says Frank Gill. “It had all this pizzazz of technology. That was brilliant on Fitz's part, but it was weird to go to this length.”

Jeffrey Walters of Virginia Polytechnic Institute and State University in Blacksburg, who says he was one of the reviewers, says he was swayed by the entire case, including the multiple sightings. He argues that it's unlikely that all the observers were mistaken. But Sibley counters that the odds are fairly high—if observers are hoping to see the birds. All the best sightings were from at least 20 meters away and lasted no more than 10 seconds. “It's just a perfect recipe for your brain to fill in the gaps,” Sibley says. “You get a brief glimpse and an impression, … and your brain turns it into an ivory-billed woodpecker.”

Conducting the analysis in secret compounded the problem, Prum says. “That process of self-convincing took place in isolation from fresh air, from people who didn't report to the boss,” says Prum. “Frankly, I think it's antithetical to good science.” One solution, Prum suggests, would have been to send the Luneau video to woodpecker experts and ask them to identify the bird without knowing the team's conclusion. Fitzpatrick rejects the charge of groupthink, insisting that the team was as objective as any scientists could be. Both Fitzpatrick and Science's Kennedy defend the decision to publish, noting that the paper was vetted by peer reviewers. “We got more than satisfactorily positive reviews,” says Kennedy, who adds that he wasn't fazed by the lack of a clear video. “I thought that it was very important, even if there was some possibility that this might be wrong.”

Meanwhile, the search for the bird continues, although it has been scaled back. In the third field season, which concluded in April, Cornell conducted a smaller, mobile search. Rather than focus on a single area, Lammertink and three colleagues spent 5 months, 7 days a week, searching 16 regions by foot and canoe. In addition, FWS also supported searches by other agencies and groups in Texas, Tennessee, Florida, and South Carolina. Again, nothing conclusive turned up. Hill is convinced that he and his team saw ivorybills in 2005 and 2006 along the Choctawhatchee River in Florida, but he admits he can't deliver enough evidence yet.

Lammertink, too, remains optimistic. “There are big areas of unexplored habitat, where on rational grounds you can see that small populations might persist.” Fitzpatrick anticipates another year or two of searching at most. “It's just too expensive,” he says, noting that it's become harder to raise money. Even if the team quits emptyhanded, Lammertink says, it will be difficult to prove the bird is not there. “It may always remain a question mark.”

Whether that uncertainty will haunt Cornell remains to be seen. “In some people's minds, the failure to find better evidence in the last couple of years has not been good for the reputation of the Lab of Ornithology,” says Russell Charif of Cornell. That specter doesn't worry Fitzpatrick. “I move with the actions that I deem appropriate for the possibility that the birds are there,” he says. “And I don't look back.”

9. BIODIVERSITY

# Predicting Oblivion: Are Existing Models Up to the Task?

1. Carl Zimmer*
1. Carl Zimmer's latest book, on E. coli and the meaning of life, will be published next May.

Huge numbers of species may be at risk of extinction from climate change, but coming up with precise estimates is proving tough

The most authoritative guide to today's extinction crisis is a database known as the Red List. Later this month, a group of scientists will gather in England to consider whether the Red List should be opened up to species that, for the moment, show no signs of trouble. Many scientists suspect that the next few decades of global warming could push some species toward oblivion. “The concern,” says the meeting's organizer, H. Resit Akçakaya, an ecologist at ecological software company Applied Biomathematics in Setauket, New York, “is that maybe some species that are threatened by climate are not reflected on the Red List.” But Akçakaya and others caution that the meeting is unlikely to come up with firm predictions of how many species will become extinct, let alone which ones will be particularly at risk.

The science of predicting extinctions from global warming is only a few years old, and the best models are rife with uncertainties. Experts generally agree that the models may be useful for giving a rough idea of the potential impact of global warming and may also offer guidance for planning preserves. But some scientists are concerned that policymakers will be expecting them to provide more precise estimates than they can deliver. “It's worrying, says Miguel Araújo, an ecologist at the Spanish National Research Council in Madrid.

Much of the current debate over climate-triggered extinctions focuses on what are known as climate-envelope models. Scientists analyze all the places where a species has been recorded and look for features of the climate that those places share. The key factors may be rainfall, for example, or the temperature during the winter.

In the early 2000s, scientists began to look at what happened to these climate envelopes in the scenarios climate scientists have projected for the coming century. “A number of us were noticing that these envelopes seemed to be winking out entirely,” says Lee Hannah, chief climate change biologist at the Center for Applied Biodiversity Science at Conservation International, a nonprofit in Arlington, Virginia.

Concerned about the prospect of mass extinctions, an international team of scientists, including Hannah, combined their data into a global analysis. They estimated the size of future climate envelopes, assuming shrinking climate envelopes meant an increased risk of extinction. Their sobering conclusion, published in Nature in 2004: Based on a midrange climate-warming scenario for 2050, “15-37% of species in our sample of regions and taxa will be 'committed to extinction.'”

The paper was enormously influential and figures prominently in the Intergovernmental Panel on Climate Change's (IPCC's) upcoming report on the impact of global warming. In a summary for policymakers, the IPCC authors warn that “approximately 20-30% of plant and animal species assessed so far are likely to be at increased risk of extinction if increases in global average temperature exceed 1.5-2.5°C.”

Some experts have criticized the IPCC for implying that climate-envelope models are more precise than they actually are. “Simply presenting those numbers as factual, saying this is how many species will go extinct, is misleading,” says Richard Pearson, a postdoctoral researcher at the American Museum of Natural History in New York City.

Pearson and other researchers have been testing climate-envelope models for their accuracy and consistency, and they've found some serious causes for concern. Araújo and his colleagues studied the ranges of 116 species of birds in England in the 1970s and 1990s. The red-backed shrike's range shrank dramatically to southeast England in the 1990s, for example, but climate-envelope models based on the 1970s data predicted that the bird's range would stretch all the way to the northern tip of Scotland. “We found that there were lots of uncertainties,” he says.

Araújo, Pearson, and other researchers published a study last year in which they compared the projections for a group of plant species in South Africa from several frequently used envelope models. “We found a huge difference between the models,” says Pearson. Their projections ranged from a 92% range reduction to a 322% expansion.

Scientists are currently debating how to make better predictions. Climate-envelope models are “simply mapping programs,” complains Daniel Botkin, a professor emeritus at the University of California, Santa Barbara. “There's no biology in that.”

To improve the performance of these models, Botkin urges researchers to include biological details about species, such as how quickly they disperse and how they interact with other species. Pearson and other modelers have already had some success at doing so, Pearson says. In a paper in press at Global Ecology and Biogeography, Pearson and his colleagues report that they can do a much better job of predicting the ranges of owls in Finland if they also factor in where woodpeckers live, as owls make their nests in woodpecker cavities. “In my opinion,” says Pearson, the role of biological interactions “is the biggest question out there at the moment, but we're just nibbling on the edges of that.”

Other researchers believe that a better strategy is to analyze existing climate-envelope models more effectively. “You have to find automated ways to extract information in intelligent ways from the data you have,” says Araújo. He and his colleagues have found that averaging the results of many climate-envelope models provides a more accurate prediction of where species can be found than any one model. “I think that's a much more useful way to go,” says Araújo. “This is likely to be a closer match to the truth than anything else we can produce so far.”

One problem with these so-called ensemble forecasts, however, is that they are a huge undertaking. Running thousands of models of thousands of species across an entire continent is far beyond the capacity of any existing software. “In the next 2 to 3 years, we won't be able to do it,” says Araújo, who is now developing a program he hopes will be up to the task.

Scientists are debating not just how to make these models better but also the best way they can be used to make conservation decisions. Some researchers are trying to estimate the percentage of species that global warming will put at risk of extinction across entire continents or even the entire planet. Walter Jetz, a biologist at the University of California, San Diego, and his colleagues recently published a study of the combined impact of climate change and land-use change on birds. They found that several hundred species may lose over half of their range by 2050. Land use will have the biggest impact on birds in the tropics, whereas climate change will be stronger at higher latitudes.

Araújo and his colleagues have a more modest goal: trying to predict patterns of change in different regions. They've been forecasting which parts of Europe will be particularly vulnerable to losing species through climate change, for instance. They've found that for plants, the mountainous regions in southern Europe will be hit hardest. For amphibians, the arid parts of southwestern Europe are most vulnerable. For now, he suggests, such estimates may be more useful for conservation than a misleadingly precise estimate of a rate of extinction for a particular species.

Identifying these sensitive regions may reveal how existing preserves may change and offer hints about how to design new ones. As the temperature warms, some preserves will no longer have a climate suitable to the species they are supposed to protect.

A number of researchers are using climate-envelope models to study how preserves may have to be altered as species shift their ranges. For instance, existing preserves could be linked by corridors to enable animals and plants to disperse from one habitat to another.

Hannah believes that scientists must move forward with this sort of planning now, even if the models have plenty of room for improvement. “The scary thing for me is that the stuff our models is showing happening decades from now, we're already seeing,” he says. He points to the extinction of frogs in the Andes, where researchers suspect that a changing climate may have fostered the spread of a lethal fungus. “These models are the best we've got at the moment, and when we see how the complexity of the world operates, it seems that it may be worse than these models are indicating.”

10. EVOLUTION

# Jumping Genes Hop Into the Evolutionary Limelight

1. Elizabeth Pennisi

With genomes from ancient fish to modern humans in hand, researchers are gaining new respect for the role transposable elements play in evolution

Call it a molecular gold rush. Researchers sifting through the supposed junk DNA between genes—a whopping 98% of the human genome—have in the past few years hit a mother lode of functional sequence full of clues about how genomes operate and change through time. And, as junk DNA has gained respect, so have mobile bits of DNA called transposons that are often the source of this genomic clutter.

Most researchers have taken a dim view of transposons, considering them molecular parasites that clog chromosomes with seemingly useless sequence, sometimes disrupting genes. Now, comparative surveys, along with experimental studies of gene regulation, are showing that transposons can influence when, where, and how genes are expressed. These so-called parasites “might be better viewed as symbiotic,” says Eric Lander, director of the Broad Institute of Massachusetts Institute of Technology and Harvard University in Cambridge, Massachusetts. David Haussler of the University of California, Santa Cruz, adds that “people are underestimating the impact of transposons” in genome evolution.

Transposons, small packages of DNA that can splice into other sequences, seem to appear suddenly in a genome, copying, cutting, and pasting themselves throughout its chromosomes. Eventually, the genes for making them mobile are disabled by mutations, and the copied sequences themselves mutate until they become indistinguishable from the rest of the genome's junk DNA. But recent studies show that some transposons don't decay. A few have lasted hundreds of millions of years relatively unchanged and are found in the same place in the genomes of many species. To be so highly conserved, they must play a role so important to survival that evolution keeps them intact, weeding out deleterious mutations. One family of transposons, for example, first made its appearance during the evolution of tetrapods; and its descendents are still recognizable, suggesting that they may have helped shape the evolution of that particular group of animals.

## Beyond gene boundaries

A few researchers suggested early on that transposons may play important evolutionary roles. Nobel laureate Barbara McClintock called them control elements when she discovered them about 50 years ago, and 2 decades later, Roy Britten and Eric Davidson suggested that they provide fortuitous opportunities for evolutionary innovation. In 1969, they proposed that new branches on the tree of life and ever-more-complex organisms arose at least in part from changes in how genes were regulated. They also argued that these changes were often caused by repetitive elements, many later identified as transposons.

Sets of genes turn on at particular times during development in specific places to specify lungs instead of gills, brain instead of kidney, and so on. Certain regulatory DNA and proteins control these activities and, Britten and Davidson suggested, repetitive elements that copied themselves into different parts of the genome occasionally reconfigured these controls. In the new positions, this mobile genetic material could, for example, corral two independent gene networks and bring them under one regulatory roof, generating new cell types and consequently new structures. “You could have genes that are totally unrelated suddenly getting turned on in the same tissue at the same time,” explains Gill Bejerano, a molecular geneticist at Stanford University in Palo Alto, California.

Over the years, this idea lost momentum. “It was untestable,” recalls Davidson, a developmental biologist at the California Institute of Technology in Pasadena. But as researchers began comparing genomes rolling off the sequencing machines, it became clear that not all junk DNA was junk, and Davidson and Britten's idea began to look more plausible.

In 2004, Bejerano, then at the University of California, Santa Cruz, and his colleagues described more than 400 stretches of at least 200 bases that were the same in human, rat, mouse, chicken, dog, and, to a lesser extent, fish. Three-quarters of these so-called ultra-conserved regions resided outside genes (Science, 28 May 2004, p. 1321). Last year, Byrappa Venkatesh of the Institute of Molecular and Cell Biology in Singapore and colleagues compared DNA of elephant sharks and humans, which diverged 530 million years ago. Working with a very sketchy draft of the shark genome, they found 4800 conserved sequences. Like others, they found that these conserved sequences tended to cluster near genes for proteins that regulate transcription and DNA binding.

Greg Elgar of the University of London has found that many conserved elements burst onto the scene between the emergence of lampreys and sharks. The timing suggests that “the evolution of these sequences [was] key to the establishment of the vertebrate gene-regulatory network for development,” Elgar says.

Gradually, researchers began to realize that some of these conserved elements were transposons. The coelacanth, a “living fossil” species that dates back more than 400 million years, led Norihiro Okada of the Tokyo Institute of Technology to a whole conserved superfamily of transposons called short interspersed repetitive elements (SINEs). Okada and his colleagues first found two SINEs in the coelacanth whose sequences looked similar enough to each other to have a common origin; they then searched for the same sequences in genome databases. They found them in salmon, trout, hagfish, dogfish sharks, lancelets, catfish, zebrafish, and the sea urchin, they reported online in Genome Research on 22 May 2006. Yet another analysis unearthed 1000 copies of a subset of these SINEs, called AmnSINEs, in both humans and chickens. “It suggests that some of these [transposable elements] had acquired function very early on in evolution, and those functions have been retained,” says John Moran, a molecular geneticist at the University of Michigan, Ann Arbor.

When Broad Institute researchers Xiaohui Xie and Michael Kamal trekked through the human genome looking for stretches of DNA that occurred multiple times, they found one that looked quite a bit like the core of a zebrafish transposable element, also part of the SINE family. Eventually, they turned up 123 more copies, some more complete than others. They found this SINE's 180-base core in the same places in other genomes and a few copies in the coelacanth. This work appeared in the 1 August 2006 issue of the Proceedings of the National Academy of Sciences. Their SINEs proved to be the same ones that Okada discovered.

When Bejerano took a close look at his ultraconserved sequences, he too discovered the remains of a family of transposable elements. He saw this first in coelacanth DNA, calling it LF-SINE, “LF” for “lobe-finned” fish. He estimates that 10,000 LF-SINEs exist in this fossil fish and, from their sequences, knows they are still active. Through genome comparisons, he and his colleagues found LF-SINE variations in human, chicken, dog, and all the other tetrapod sequences in the public databases.

The conservation of the sequence in similar places in the genomes of all these species suggested they play a key role in genome function. Bejerano tested one that was located 500,000 bases from a gene coding for a transcription factor active in motor neuron development. That's not close enough to be the primary controller of the gene's activity, but Bejerano thought perhaps the transposon could exert a geneactivating effect from afar as a so-called enhancer. He linked the transposon's sequence to a gene that would cause cells expressing that gene to turn blue when stained and put that DNA into fertilized mouse eggs. The marker gene turned developing nervous system tissue blue precisely where Bejerano expected it, he and his colleagues reported 14 May 2006 in Nature.

## Significant force

No one was sure, however, how frequently transposable elements were co-opted in this way. In theory, the odds are in the transposons' favor. “A transposon has no problem making 50,000 copies and splattering them throughout the genome,” Haussler explains. Even if just one or two happen to land where they can be useful, they could still add up to be a powerful force in evolution. Yet “the impression was that there was a case here, a case there, that it was a really interesting fluke,” Bejerano says. Now work by Bejerano and others has shown that functional transposable elements are more than a fluke.

Jerzy Jurka of the Genetic Information Research Institute in Mountain View, California, working with Kerstin Lindblad-Toh and Tarjei Mikkelsen, both from the Broad Institute, has found a surprisingly large role for transposons in the evolution of placental mammals. They compared the newly deciphered opossum genome to those of humans and other placental mammals to identify regions that were conserved in placental mammals but not found in this marsupial. These genetic innovations likely underlie the developmental and other differences between the two mammal groups. More than 95% of these innovations in the placental mammals were outside protein-coding genes, and 16% matched up to one of more than a dozen transposon families, the researchers reported in the 10 May issue of Nature. They conclude that transposable elements were likely instrumental for regulatory changes underlying features characteristic of placental mammals. “It's such a significant fraction that it can't be dismissed,” Lander says.

Overall, Bejerano and his colleagues have just found more than 10,000 conserved transposons in the human genome, many dating back well before the split between placental and marsupial mammals. He and his colleagues first identified these sequences by looking for conserved stretches across a range of vertebrate genomes, including human, and subtracting out any that represented genes. They pinpointed transposon-derived DNA by matching up the shared sequences with known mobile elements in a database. The matches represent more than 5.5% of all the conserved noncoding sequence, he and his colleagues reported online 23 April in the Proceedings of the National Academy of Sciences.

At this point, the evidence for a role for most of these partially preserved transposons is circumstantial. Their conservation suggests they have a function; otherwise, they should slowly disappear. But new, more powerful tools for analyzing genomes in silico or for pinpointing where transcription factors bind DNA promise rapid progress toward understanding what these conserved transposons do (Science, 25 May, p. 1120). Says Bejerano: “We should have pretty spectacular answers pretty soon.”