News this Week

Science  08 Jul 2005:
Vol. 309, Issue 5732, pp. 226

You are currently viewing the .

View Full Text

Log in to view the full text

Log in through your institution

Log in through your institution


    Deep Impact Makes a Lasting Impression on Comet Tempel 1

    1. Richard A. Kerr

    Splat! Mission accomplished. The two-part Deep Impact spacecraft—a bulletlike hypervelocity impactor and its watchful mothership—performed flawlessly on 4 July, punching a hole in the icy dirtball of comet Tempel 1 in full view of all the world.

    In the first hours, at least, the collision revealed none of the hoped-for secrets of the solar system's formation; real science doesn't always make for instant science. But mission scientists have no doubt that Deep Impact returned much of the raw data they need. "We do impact cratering simulations [in the lab] in pieces," says team member Peter Schultz of Brown University in Providence, Rhode Island. At Tempel 1, "we saw all the pieces come together in one giant event." Eerily, the real thing bore a fair resemblance to computer animations based on lab experiments and numerical simulations.

    Deep Impact wasn't always unalloyed fun for team members. The cost-constrained, PI-led Discovery mission had a checkered history of cost overruns, near-fatal reviews by NASA headquarters, and technical problems, including an onboard computer that had to be rebuilt. "We were very close to being canceled," says PI Michael A'Hearn of the University of Maryland, College Park. All the scrutiny may have paid off, however. Despite bumpy trials early on, the computer and its comet-targeting software deftly homed the impactor in on Tempel 1's nucleus, snapping pictures down to the last 3 seconds before impact.

    On target.

    A fireball (brightest splotch) expands above comet Tempel 1 as a vertical column of debris (shadow cast toward top) rises from the collision with Deep Impact.


    The death-plunge pictures were revealing. The nucleus of Tempel 1 "looks very different from Wilt 2's or Borrelly's," says A'Hearn. Those are the other two comet nuclei closely imaged by spacecraft. Unlike on those nuclei, "a lot of things on Tempel 1 look like [impact] craters," he says. A band of smooth terrain of unknown origin wraps around the waist of the elongate, 14-kilometer-long body. Other features include topography that formed when the sun ate away at primordial ice in layered strata, said A'Hearn.

    The encounter had a rather conventional outcome, considering that this spring scientists "didn't have a clue" what was going to happen, as A'Hearn put it (Science, 27 May, p. 1247). Tempel 1 didn't just swallow up the impactor, the way something as accommodating as a marshmallow might. Nor did it form a small, bowl-shaped crater, the way a strong material would. In the images returned by the afternoon of the first day, the first sign of contact was a very small, faint dot of a flash, says Schultz. That was the impactor, a clothes-washer-size, copper-laden bullet, penetrating the surface. After 150 milli-seconds, a "really bright flash" saturated the flyby spacecraft's camera. The impactor had penetrated the nucleus and vaporized, and now a ball of incandescent comet vapor was expanding above the surface.

    At the same time, the shadow of a growing vertical column fell across the nucleus, apparently cast by material shooting out of the penetration hole like a roman candle, says Schultz. A curtain of ejecta zoomed upward as the curtain expanded outward across the nucleus, "just like the movies" based on the experiments, says Schultz: "It looks so similar to the experiments." That implies to Schultz that Tempel 1 is not armored by a thick hard crust, as some had imagined, but wrapped in a soft, dusty layer.

    Still, team members had yet to identify the much-anticipated crater hidden beneath suspended impact dust. Further image processing should reveal it, A'Hearn said. Schultz thinks it will be big. "Now we have to go back and do more complicated experiments and compare them with numerical simulations," he says. So far, he and his colleagues have hardly mentioned the spectroscopic data that in coming months should reveal the composition of freshly exposed primordial material—presumably the same stuff that made up the planets. That analysis will take much longer than an instant.


    Britain's Research Agencies Endorse Public Access

    1. Eliot Marshall

    Starting in October, all investigators funded by the big eight research agencies in Britain may be required to put their papers and meeting talks in a free public archive “at the earliest opportunity, wherever possible at or around the time of publication.” An oversight group, Research Councils UK (RCUK), handed down this formula last week as its final proposal after months of consultation with interested groups. By one estimate, it would cover half of all U.K.-funded research.

    Despite the mandatory tone, journals will find some wiggle room that may allow them to keep their usual embargoes. RCUK says its mandate is “subject to copyright and licensing arrangements” that can restrict what authors do ( RCUK spokesperson Heather Weaver said this phrase recognizes that “publishers vary” in how they handle rights, and the government is setting no fixed time frame for free data release—other than “as soon as possible.”

    Advocates for the open-access movement praised the RCUK announcement. Some think it comes closer to their goals than a policy announced earlier this year by the U.S. National Institutes of Health (NIH), which merely encourages authors to put papers in the U.S. PubMed Central database within 12 months of publication (Science, 29 April, p. 623, and 11 February, p. 825). Peter Suber—a professor of philosophy at Earlham College in Richmond, Indiana, and leader of the Public Knowledge advocacy group in Washington, D.C.—described it as “an excellent policy” because it is mandatory, unlike NIH's. But he says the copyright “loophole … will allow publishers to impose embargoes.”

    Publishers, whose revenues are threatened by the open-access movement, found fault with the RCUK approach. A group representing 320 nonprofit, academic, and scientific society journals—the Association of Learned and Professional Society Publishers in Clapham, U.K.—released a critique on 30 June by Executive Director Sally Morris ( Among other concerns, it warns that the open-access trend may “siphon off” subscriptions to society publications.

    RCUK specifies only that papers should be put in “an appropriate e-print repository (either institutional or subject-based), wherever such a repository is available.” More than 50 qualify in Britain alone. RCUK officials say this and other fine points will be worked out in consultations through 31 August, before the policy takes effect this fall.


    A New Skirmish in the Yanomamö Wars

    1. Charles C. Mann

    Sensational accusations that anthropologists mistreated Venezuela's Yanomamö Indians while studying them continue to roil the American Anthropological Association (AAA). Last week, AAA members voted 846-338 to rescind the association's report on the charges, which were leveled almost 5 years ago in journalist Patrick Tierney's book Darkness in El Dorado. Although opposition to the referendum was “very vocal,” says Thomas Headland, an anthropological consultant to SIL International in Dallas, Texas, who supported it, “I guess there's a silent majority among the 11 or 12 thousand members of the AAA.”

    Tierney's book set off a firestorm with its charges that researchers had “devastated” the Yanomamö, who live near the headwaters of the Orinoco River. The most explosive allegation—that prominent anthropologist Napoleon Chagnon of the University of California, Santa Barbara, and the late geneticist James V. Neel exacerbated and possibly caused a lethal 1968 measles epidemic—was quickly shown to be implausible (Science, 29 September 2000, p. 2251; 19 January 2001, p. 416). But researchers continued to battle over a host of other claims, including that Chagnon's widely known depictions of the Yanomamö as “fierce” and violent had provided intellectual cover to people trying to take over their land.

    In February 2001, AAA appointed a task force to “conduct an inquiry” into the growing storm. If the measure was intended to quell the dispute, it failed. Released in July 2002, the task force's 325-page final report exonerated Chagnon of the most serious charges (Science, 19 July 2002, p. 333) but argued that his association with a group of wealthy, allegedly corrupt Venezuelans was “unacceptable on both ethical and professional grounds” because visitors made many illicit trips to Yanomamö villages “without any quarantine procedures or other protections for the indigenous peoples.” More importantly, the task force concluded that Chagnon's “representations [of the Yanomamö as 'fierce'] have been damaging” to them.


    Napoleon Chagnon's depiction of the Yanomanö as “fierce” and violent continues to divide anthropologists.


    Almost immediately, anthropologists Thomas Gregor of Vanderbilt University in Nashville, Tennessee, and Daniel Gross of the World Bank in Washington, D.C., attacked the task force. In two critiques in the Chronicle of Higher Education and American Anthropology (the flagship AAA journal), Gregor and Gross scoffed that the investigation was “a model of ineptitude.” The five-member committee, Gregor says, based its conclusions “on biased interviews of selected, unrepresentative Indians.” Not consulted, Chagnon's defenders note, were indigenous leaders such as Jaime Turon, elected head of the Upper Orinoco district, who wrote in a 2003 open letter that Chagnon and his associates, far from hurting the Yanomamö, “were the only ones that helped us … in the 1960s and 1970s.”

    Gross and Gregor obtained the 50 signatures AAA bylaws require to hold a referendum on the report. “We were not attempting to mount a defense of Mr. Chagnon,” Gross wrote in an e-mail to Science—indeed, they have attacked each other's ideas in print since the 1970s. “However, Chagnon [and Neel] were subjected to a process that was highly loaded ideologically and in which they had no way of defending themselves.”

    Despite the strong rejection of the task force report, few expect a cease-fire in the Yanomamö wars. Robert Borofsky of Hawaii Pacific University in Honolulu has said AAA should keep “evaluating the charges”; Brazilian director José Padilha is filming a BBC documentary on the affair for broadcast in early 2006. Raymond Hames of the University of Nebraska, Lincoln, says Chagnon is a “lightning rod” for the conflicts now rending anthropology. The field is bitterly split, he says, between “people who try to do science and people who believe that science is impossible or—with a postmodern ring—is actually an unethical thing to do, a hegemonic tool of Western imperialism.” Chagnon's high-profile support of a data-driven view of anthropology, Hames—a Chagnon collaborator—and other anthropologists say, has made him a special object of opprobrium to the field's postmodern flank.

    Chagnon further “infuriated people,” Gross says, when he argued (Science, 26 February 1988, p. 985) that Yanomamö men “who had killed had higher reproductive success”—an evolutionary explanation for the high levels of violence Chagnon said he observed. The claim, Gross says, simultaneously drew the ire of researchers suspicious of what they saw as “crude biological determinism” and activists who believed that the depiction of the Yanomamö as warlike, which they believed inaccurate, “directly harmed” them.

    Inflamed by Chagnon's sometimes hot-tempered personal style, these conflicts have led to divisions that are unlikely to be resolved quickly. Indeed, Leslie Sponsel of the University of Hawaii, Manoa, one of Chagnon's most outspoken detractors, calls the vote “simply another smoke screen to distract attention from the multitude of diverse allegations made by Tierney, some of which were confirmed by various investigations.”

    Although Chagnon calls himself “pleased” by the vote, he believes that “activist anthropologists” will continue to use ethical charges “as a social and political weapon.” Meanwhile, he believes the AAA task force may actually have “worsened the plight of the Yanomamö because the [Venezuelan government] has, as a consequence of AAA actions, been shut off to researchers who might be more genuine and effective in their efforts to help them.”


    Madrid Heart Center to Be Rescued

    1. Xavier Bosch*
    1. Xavier Bosch is a science writer based in Barcelona.

    BARCELONA, SPAIN—Hoping to recover from a calamitous start, Spain's Ministry of Health is using private money to rescue a troubled heart research facility in Madrid. When the Spanish Cardiovascular Research Center (CNIC) got tangled in management problems last year, the director was let go. Now the government has recruited a prominent new chief—cardiologist Valentin Fuster of the Mount Sinai School of Medicine in New York City—and signed up five companies to help float the project for 10 years.

    CNIC slipped into limbo in May 2004 following a dispute between the government and acting chief Salvador Moncada, an expert in nitrous oxide who heads the Wolfson Institute for Biomedical Research at University College London. Moncada challenged a court-issued reprimand over travel costs. He and CNIC then parted ways, and almost a year ago, Health Minister Elena Salgado began searching for a new leader, a spokesperson says. On 27 June, she announced that Fuster, a Spanish national, had agreed to return to Spain to help relaunch CNIC and become director on a date to be determined.

    To help revive the project and sustain it through 2015, she said, five big companies with no stake in health products or the drug industry have agreed to kick in €170 million, 35% of a new sustaining fund of €500 million. The remaining 65% will come from the health ministry. Salgado describes the project as “an innovative joint venture between the government and the private sector.” The deal also will make it possible to inaugurate CNIC's new €60 million building on 1 September.

    Double duty.

    Valentin Fuster will lead research in New York City and Madrid.


    Fuster says he plans six major research departments devoted to areas such as tissue regeneration, stem cell studies, heart embryogenesis, and basic genetics and proteomics. He aims to hire 200 scientists on government salaries; they will also receive bonuses based on productivity, to be financed with company money. There will be no place for the unproductive, he says. The involvement of the private sector is a “breakthrough,” adds Fuster, noting that it is a first for Spain. According to the health ministry, the companies will share in rights to CNIC's medical discoveries.

    Fuster says he intends to split his time between Mount Sinai, where he is in charge of 41 basic and clinical scientists, and CNIC. He plans “a highly aggressive scientific interrelation” between CNIC and Mount Sinai—as well with other U.S. and european centers. For example, Salk Institute developmental biologist Juan Carlos Izpisúa-Belmonte is reportedly discussing plans to work with CNIC on embryonic stem cells from a new base at Barcelona's Centre of Regenerative Medicine.

    It is a “very good initiative,” says stem cell researcher Jordi Petriz of Barcelona's IDIBAPS Institute. “But it remains to be seen how CNIC will be sustained at the long term.”


    Ten Centers Chosen to Decode Protein Structures

    1. Jennifer Couzin

    An ambitious and costly plan to churn out protein structures shifted into its second phase last week, as the National Institutes of Health (NIH) announced roughly $300 million in new awards. The Protein Structure Initiative (PSI) aims to deposit up to 5000 new protein structures in a public database.

    Roughly $200 million will go to four large-scale centers that, much like the centers that sequenced the human genome, will crank out protein structures as rapidly as possible. It's a labor-intensive task: Until recently, a single protein structure could take a year to decipher. The rest of the money goes to six “specialized” centers that will focus on how to handle some of the most challenging proteins, including potential drug targets.

    PSI was launched as a pilot project 5 years ago to send protein biology into a new realm. Protein structures can shed light on both normal and deviant molecular pathways and on how divergent species, from bacteria to humans, overlap in their biology (Science, 11 March, p. 1554). But the initiative drew fire from researchers who felt that taking snapshots of an isolated protein's structure reveals little about its function.

    Protein boost.

    Four large-scale and six specialized centers will take part in PSI.


    PSI's expansion hit another snag this year, when a tight NIH budget forced PSI to scale down its current awards from $75 million a year to about $60 million (compared with $68 million for each of the last 2 years of the pilot project). “We had to make reductions in awards to the centers,” says PSI director John Norvell of the National Institute of General Medical Sciences. The four large-scale centers will each receive roughly $9 million to $10 million a year for the next 5 years; specialized centers will garner $3 million to $4 million a year.

    Awardees say they are trying to drive costs downward. “Our goal would be to get to less than $10,000 per protein,” says Lance Stewart, vice president of the company deCODE Biostructures in Bainbridge Island, Washington, and the leader of one of the new specialized centers. Currently, he says, deducing structures of bacterial proteins can cost $100,000; more complex eukaryotic ones can soar to 10 times that.

    Having deciphered structures for more than 1100 proteins, most of them bacterial, PSI is now looking to the “higher hanging fruit,” says Norvell. Researchers agree that won't be easy. Gaetano Montelione of Rutgers University in Piscataway, New Jersey, who directs the Northeast Structural Genomics Consortium, says his success rate for deducing eukaryotic protein structures is 1%, compared with 10% for bacteria; eukaryotic proteins, he says, don't grow well in Escherichia coli bacteria, the method used to purify them.

    Montelione expects to boost his eukaryotic protein yield, though, and will likely need to: His large-scale center will focus on protein networks in cancer biology, and he has already drawn up hit lists of proteins that drive tumor growth.


    New Panel to Offer Guidance on Dual-Use Science

    1. Jocelyn Kaiser

    Most biologists don't spend much time thinking about whether their co-workers in the lab are trustworthy or whether a terrorist might profit from the paper they're about to submit. But a newly formed U.S. committee has begun considering how life scientists should deal with such questions.

    Meeting in Bethesda, Maryland, last week for the first time, the panel hopes to develop guidelines—such as codes of conduct—for “dual use” research in the life sciences that will strike a balance between limiting risks and preserving scientific freedom. “If we don't do this carefully, we run the risk of losing what's really the greatest scientific engine the world has ever seen,” says panelist Paul Keim of Northern Arizona University in Flagstaff.

    The 24-member interagency panel, created in March 2004 and led by the Department of Health and Human Services (HHS), is an outgrowth of a 2004 National Academies report that looked at the potential misuse of biotechnology in the wake of the deadly 2001 anthrax letter attacks. As White House Homeland Security Council official Rajeev Venkayya told the committee last week, 2 years ago “there was an increasing sense of angst” on the council that some newly published studies, such as synthesizing viruses from scratch, could be misused. The academies' report, he said, helped stave off calls for more “draconian” measures.

    Selecting the panelists took more than a year, however, and its membership was only unveiled at last week's meeting. The roster is studded with scientific stars as well as intelligence, biosafety, and bioweapons experts. Harvard University microbiologist Dennis Kasper is chair of the panel, officially the National Science Advisory Board for Bio-security. Its 2-year charter runs out next March but is expected to be extended.

    The panel's goal is to create “a culture of responsibility,” says National Institute of Allergy and Infectious Diseases Director Anthony Fauci, an ex officio member. It is modeled on HHS's Recombinant DNA Advisory Committee, which was created 30 years ago to address concerns about the risks of genetic engineering.

    One big challenge is a definition of dual use. The academies' report included case studies of seven potentially controversial experiments, such as modifying a microbe to make it resistant to drugs. But it did not consider studies that analyze the country's vulnerability to attack, such as a paper modeling use of botulinum toxin to poison the U.S. milk supply that the Proceedings of the National Academies of Sciences published last week despite concerns from HHS. The board also will tackle guidelines for journals, codes of scientific conduct, international collaborations, and advice for studying synthetic genomics.

    Observers are cautiously optimistic about what the board will achieve. “I just don't know if they're going to be able to muster the courage to take steps that are sufficiently strong,” says Ed Hammond of the Sunshine Project, a bioweapons watchdog group based in Austin, Texas.


    Potentially More Lethal Variant Hits Migratory Birds in China

    1. Dennis Normile

    When China reported in mid-May that the H5N1 avian influenza virus had caused the deaths of 1000 or more migratory birds at a breeding ground in western China, ornithologists worldwide were alarmed. “It is the biggest and most extensively mortal avian influenza event ever seen in wild birds,” says David Melville, an ornithologist in New Zealand. Now, in a paper published online by Science this week (, Jinhua Liu of the College of Veterinary Medicine in Beijing and colleagues there and at five other Chinese institutions report that the outbreak at Lake Qinghai in western China appears to have been caused by a new H5N1 variant that may be more lethal to wild birds, as well as to experimentally infected mice. Similar findings, from different groups, were published online this week by Nature. The results suggest that the virus is evolving and raise the possibility that surviving birds could spread it over an even wider geographic area, endangering more poultry and increasing the chances of further genetic changes that could spark a deadly human pandemic.

    Liu and colleagues fully sequenced four isolates recovered from various bird species and found them all to be very similar but distinct from any H5N1 sequences posted in GenBank. George Gao, a virologist at the Chinese Academy of Sciences' Institute of Microbiology and the corresponding author, says the evidence suggests that the genetic changes account for the increased mortality, although more data are needed to be certain.

    Breeding ground.

    Flu experts worry that migratory birds infected with a new strain of the H5N1 virus, like the bar-headed goose (left), might carry it far from their breeding ground at Lake Qinghai.


    The researchers also tested the pathogenicity of the virus by using it to infect mice, which succumbed more quickly than mice infected with other H5N1 strains. “This shows that [the virus] is also more pathological for mammals,” says Ilaria Capua, a virologist at the Istituto Zooprofilattico Sperimentale delle Venezie in Legnaro, Italy. This does not necessarily mean that humans will be more easily infected or that the virus can be passed from human to human, she says.

    The outbreak raises other questions, including how the virus got to this sparsely populated corner of China. Since H5N1 appeared, researchers have debated whether migratory birds can spread it. Some aquatic birds are known to host strains of the virus with no or minimal symptoms. But the United Nations' Food and Agriculture Organization says there is no evidence tying outbreaks in poultry to wild birds. Still, Capua suggests that migratory birds from different regions might have carried several less pathogenic H5N1 strains to the “melting pot” environment of the lake, where this new variant emerged. Melville counters that abundant evidence shows that human activity—transporting poultry, poultry products, and even contaminated crates—can spread avian flu viruses over seemingly improbable distances.

    A more pressing question is where these migratory birds might carry the virus next. Melville says that bar-headed geese, one of the infected species, fly several thousand kilometers to wintering grounds in India, potentially dropping the virus along the way. For many other species that breed at Qinghai, the understanding of migration routes “is very rudimentary,” he says.

    But “dead ducks don't fly,” he adds, quoting an essay on wild birds and flu by Hong Kong-based ornithologist Martin Williams—meaning that if this new strain kills all the birds it infects, it is not going to travel very far. A priority, says Melville, should be determining if surviving birds are carrying a weakened strain of the virus, or if some species or individual birds are carrying the same variant with minimal health effects. “These are the important questions,” says Gao, whose team is gearing up to answer them by collecting additional samples from healthy birds over the next couple of months.


    Scientists Say Genome Canada's Cofunding Rules Stymie Good Ideas

    1. Wayne Kondro*
    1. Wayne Kondro is a freelance writer in Ottawa.

    OTTAWA, CANADA—Does it make sense to reject a study of whether poplar trees can help mitigate global warming simply because the trees were going to be planted anyway? That Zen-like question has become a rallying cry for scientists protesting rules about cofunding of research proposals in Canada.

    Last month, Genome Canada rejected a proposal from University of Toronto botanist Malcolm Campbell to team up with the Canadian Forest Service (CFS) on an $18.4 million poplar genomics initiative that would have examined the role of the trees as carbon sinks or feedstock for biofuels. It was one of 27 ideas shot down in the first of a two-stage process that focused on the financial, rather than scientific, merits of each application. Some 66 proposals remain in the running for $132 million in this, the third round of funding from Genome Canada.

    The rejected scientists fell victim to a flawed process, say 39 prominent researchers who last month released a public letter suggesting that cofunding may be undermining the country's ability to support cutting-edge research (Science, 24 June, p. 1867). “It does sound like sour grapes,” admits Campbell, who says he was lured home last fall from Oxford University in the U.K. because of the “promising” environment created by a raft of new Canadian programs such as Genome Canada. “But it's sour because one does not expect when formulating a scientific proposal to have it evaluated first on the grounds of management criteria.”

    But Martin Godbout, president of Genome Canada, says the complaints have no merit. Cofunding is essential for stretching scarce resources, he says, and is an integral part of Genome Canada's mission to collaborate with provincial and local governments, industry, and private foundations. “Cofunding works,” he asserts. He also defends the initial screening, saying that it was needed to cope with the heavy workload and that it won't affect which proposals ultimately receive funding.

    Growing unhappiness.

    Malcolm Campbell and other Canadian scientists don't like how Genome Canada weeds out grant proposals.


    The letter writers, including some whose proposals were rejected, argue that a “committee of accountants” scoured applications for any flaw that might be used as an excuse to whittle the field. In Campbell's case, the agency decided that the CFS contribution amounted to trees that would be planted regardless of whether the project proceeded. “We all sat there, with our mouths agape, literally, for a minute,” says Campbell, describing his team's reaction in a meeting with the due-diligence review committee. “We were at a complete loss as to how this did not qualify,” he added, noting that the project had passed muster with two of Genome Canada's five regional genomics centers.

    John Bergeron, chair of the department of anatomy and cell biology at McGill University in Montreal, couldn't understand why a KPMG accountant who chaired the review committee viewed as an apparent conflict of interest the housing of mice for Bergeron's proteomic studies of liver diseases at a company associated with his team. “It was so weird,” says Bergeron. “You're sitting there, and you're saying: What's going on? This is wacko.”

    Godbout doesn't think so. Most of the projects rejected demonstrated a poor understanding of the goal of cofunding, he says, which is to generate novel funding sources. Another problem, he suggests, is that the results were delivered differently this year: Applicants who failed the financial review were informed immediately that they were out of the running. In previous years they were not notified until the winners had been chosen, leaving some with the impression that they'd failed the scientific review. “Next time, we will again run these two processes in parallel, within the same week,” Godbout announced. But he predicted that “the outcome will be the same.”

    Regardless of which projects are chosen, Lou Siminovitch, an eminence grise within Canadian genetics and professor emeritus at the University of Toronto, fears that cofunding programs put too great an emphasis on grantsmanship and wooing potential investors to the detriment of science. “They're making people spend so much time at their desks that they have no time to innovate,” he frets.


    EPA Draft Rules for Human Subjects Draw Fire

    1. Erik Stokstad

    Efforts by the Environmental Protection Agency (EPA) to adopt ethical guidelines for controversial testing of pesticides on humans have run into trouble.

    Last week, the Senate, as part of a measure setting the agency's 2006 budget, voted to bar EPA from using any such studies in its regulatory decisions. The House had passed an identical amendment in May, although differences in the two bills must still be reconciled. And a leaked version of draft regulations has already drawn criticism from scientists who say the rules don't go far enough. “This document is not about protecting human subjects,” says toxicologist and environmental activist Ellen Silbergeld of Johns Hopkins University in Baltimore, Maryland.

    The issue of human testing flared up in 1998, when the Environmental Working Group, an advocacy organization in Washington, D.C., released a report questioning whether it was ethical for EPA to use studies based on volunteers being fed pesticides to help determine how to regulate the compounds. In 2001, EPA turned to the National Academies for advice. The academies' study, published last year, concluded that some research was acceptable under certain conditions (Science, 27 February 2004, p. 1272).

    Meanwhile, EPA had begun to work on rules that would extend a federal ethics code for human research to studies not conducted or funded by EPA. Last week, Representative Hilda Solis (D-CA) and Senator Barbara Boxer (D-CA), who introduced the EPA amendments in their respective bodies, made public a copy that was scheduled for release in August.

    In one corner.

    Senator Barbara Boxer offered one of two Senate amendments that send mixed signals to EPA.

    Critics are unhappy with the scope of the rules to protect pregnant women and children. The ethical requirements would only apply to studies conducted to identify or quantify a toxic effect, with the results intended for EPA's use. The agency could still draw upon other studies in which the subjects might have been harmed from exposure to small doses of a substance, says John Hopkins pediatrician Lynn Goldman, who headed EPA's pesticides program from 1993 to 1998.

    Another worry is that EPA is setting the bar too low by declaring that it will reject only those studies that fail to “substantially” comply with ethical guidelines. EPA can still decide to accept a study if it decides that the ethical flaws are outweighed by public health benefits. “That's an enormous loophole,” Goldman says.

    According to the leaked draft regulation, EPA would also consider using research conducted before the rules are put in place, if that research provided useful knowledge not attainable any other way and met the prevailing ethical standards at the time. But that's not good enough, says Goldman: “We need to make sure we're not going down a slippery slope.”

    The critics' biggest concern is that the rules ignore an academies' recommendation to create an outside expert panel to review proposals for pesticide tests and determine if they would be ethically acceptable. EPA believes that approach would “unnecessarily confine EPA's discretion to adopt more effective or efficient approaches in the future,” according to the leaked draft.

    The agency's stance does have its backers in Congress. In addition to Boxer's measure, the Senate passed an amendment offered by Senator Conrad Burns (R-MT) for EPA to stay the course and issue final rules within 6 months. And because Burns chairs the spending panel that oversees EPA's budget, his view could very well prevail when the House and Senate work out differences between the two bills later this summer.


    Radiation Dangerous Even at Lowest Doses

    1. Jocelyn Kaiser

    A new National Research Council (NRC) report* finds that although the risks of low-dose radiation are small, there is no safe level. That conclusion has grown stronger over the past 15 years, says the NRC committee, dismissing the hypothesis that tiny amounts of radiation are harmless or even beneficial.

    The risk of low-level radiation has huge economic implications because it affects standards for protecting nuclear workers and for cleaning up radioactive waste. The Biological Effects of Ionizing Radiation VII (BEIR VII) panel examined radiation doses at or below 0.1 sieverts (Sv), which is about twice the yearly limit for workers and 40 times the natural background amount the average person is exposed to each year. For typical Americans, 82% of exposure stems from natural sources such as radon gas seeping from Earth; the rest is humanmade, coming mostly from medical procedures such as x-rays.

    In its last report on the topic in 1990, a BEIR panel calculated risks by plotting cancer cases and doses for survivors of the two atomic bombs dropped on Japan in World War II. Risks appeared to increase linearly with the dose. Based on evidence that even a single “track” of radiation can damage a cell's DNA, the panel extrapolated this relationship to very low doses to produce what is known as the linear no-threshold model (LNT).

    Some scientists have challenged this LNT model, however, noting that some epidemiological and lab studies suggest that a little radiation is harmless and could even stimulate DNA repair enzymes and other processes that protect against later insults, an idea known as hormesis (Science, 17 October 2003, p. 378).

    Risky business.

    A new review verifies that even radiation levels well below those encountered by nuclear workers can raise cancer risk.


    But the 712-page BEIR VII report finds that the LNT model still holds. The panel had the latest cancer incidence data on the bomb survivors, as well as new dose information. Committee members also reviewed fresh studies on nuclear workers and people exposed to medical radiation, all of which supported the LNT relationship. The model predicts that a single 0.1-Sv dose would cause cancer in 1 of 100 people over a lifetime. Such risks should be taken into account, the report cautions, when people consider full-body computed tomography scans, a recent fad that delivers a radiation dose of 0.012 Sv.

    At the same time, notes panelist Ethel Gilbert, an epidemiologist at the National Cancer Institute in Bethesda, Maryland, “we can't really pinpoint” the risk at the lowest doses. The BEIR VII panel examined the latest evidence for a threshold. But it found that “ecologic” studies suggesting that people in areas with naturally high background radiation levels do not have elevated rates of disease are of limited use because they don't include direct measures of radiation exposures. The panel also concluded that animal and cell studies suggesting benefits or a threshold for harm are not “compelling,” although mechanisms for possible “hormetic effects” should be studied further.

    Toxicologist Ed Calabrese of the University of Massachusetts, Amherst, a vocal proponent of the hormesis hypothesis, says the panel didn't examine enough studies. “It would be better if more of the details were laid out instead of [hormesis] just being summarily dismissed,” he says. The panel's chair, Harvard epidemiologist Richard Monson, acknowledges that the long-running debate over the LNT model won't end with this report, noting that “some minds will be changed; others will not.”


    Are Humans Still Evolving?

    1. Michael Balter

    The goal of much of modern medicine and culture is effectively to stop evolution. Is that happening?


    The news made headlines around the world: Blonds were going extinct. According to CNN and other media, a World Health Organization (WHO) study concluded that the gene for blond hair, which was described as recessive to dominant genes for dark hair, would disappear in 200 years. The BBC announced that the last natural blond would be born in Finland and suggested that those who dyed their hair might be to blame, because “bottle blonds” were apparently more attractive to the opposite sex than natural blonds were and thus had more children.

    Fortunately for blonds, the whole story turned out to be a hoax—“a pigment of the imagination,” as the Times of India later put it. WHO announced that it had never conducted such a study, and hair color is probably determined by several genes that do not act in a simple dominant-recessive relationship. The story, which may have originally sprung from a German women's magazine, apparently simply leaped from one media outlet to another.

    Although the story was untrue, the ease with which it spread reflects popular fascination with the evolutionary future of our species, as well as the media's appetite for evolutionary pop science. Today, Oxford University geneticist Bryan Sykes is receiving voluminous coverage for his book, Adam's Curse, which predicts that continuing degeneration of genes on the Y chromosome will leave men sterile or even extinct in 125,000 years. Many biologists say that the question they most often receive from students and the public is “Are humans still evolving?”

    To many researchers, the answer is obvious: Human biology, like that of all other living organisms on Earth, is the result of natural selection and other evolutionary mechanisms. Some say the question itself betrays a misunderstanding of how evolution works. “The very notion that … we might not be evolving derives from a belief that all other life forms were merely stages on the way to the appearance of humans as the intended end point,” says primatologist Mary Pavelka of the University of Calgary in Canada.

    But other scientists point out that in developed countries, culture, technology, and especially medical advances have changed the evolutionary rules, from survival of the fittest to the survival of nearly everyone. The result, they say, is a “relaxation” of the selective pressures that might have operated 50 or 100 years ago. “Biologically, human beings are going nowhere,” says anthropologist Ian Tattersall of the American Museum of Natural History in New York City. University College London geneticist Steven Jones agrees. “The central issue is what one means by 'evolving,'” Jones says. “Most people when they think of evolution mean natural selection, a change to a different or better adapted state. In that sense, in the developed world, human evolution has stopped.”

    Modern mismatch.

    Overbite is widespread among modern humans, but evolution may not be to blame.


    Yet millions of people in developing countries continue to live under the combined stresses of poverty and disease. Under these conditions, even skeptics of ongoing human evolution agree that natural selection may be favoring genes that confer resistance to disease or enhance reproductive fitness in other ways. Indeed, researchers are now tracking how deadly maladies such as AIDS and malaria exert selective pressure on people today. “As long as some people die before reproducing or reaching reproductive age, selection is likely to be acting,” says geneticist Chris Tyler-Smith of the Sanger Institute near Cambridge, United Kingdom.

    Even in developed countries, where survival tends to be prolonged for almost all, recent studies suggest that there are still genetic differences among people in fertility and reproductive fitness, an indication that natural selection is operating. “The question 'Are humans still evolving?' should be rephrased as 'Do all people have the same number of children?'” says Pavelka. “The answer is that we do not make equal contributions to the next generation, and thus we are still evolving.”

    Over the past few years, a wealth of new data has begun to illuminate how natural selection has shaped—and may still be shaping—humanity. The human genome project and genetic data from people around the world have powered an explosion of research seeking signs of natural selection in human DNA. “A lot of the tools we are now using to search for selection were developed by people working on flies and other organisms,” says evolutionary geneticist Bruce Lahn of the University of Chicago. “But once researchers began to discover examples of ongoing selection in humans, it opened the door and gave them confidence that they could find even more.”

    So far, the number of confirmed cases of genes under recent selective pressure is only “a handful,” says Tyler-Smith. But that is likely to change once the results of the International HapMap Project, a multination effort to determine worldwide variation in the human genome, are released later this year. Because genetic variation is the raw material on which natural selection works, favoring certain alleles over others, Tyler-Smith says the HapMap should “give us an overall view of the regions of the genome that have been under selection.”

    Drifting toward modernity?

    To science-fiction fans, the future of human evolution conjures up visions of dramatic changes in our bodies, such as huge brains and skulls. “Many people see us continuing on the righteous path of increasing intelligence,” says Pavelka. “But we will not head in the direction of larger brains and crania as long as infants are required to pass through a woman's pelvis to get into the world.”

    Whatever lies in our evolutionary future, scientists agree that the modern human body form is largely the result of evolutionary changes that can be traced back millions of years. The uniquely human lineage dates from about 6 million years ago, and many studies have demonstrated that our divergence from chimpanzees was accompanied by strong selective pressure, for example on the human brain. Yet researchers caution that not all morphological changes—the ones we can see in body shape and size—are the result of natural selection; some may not be due to genetic evolution at all. For example, the increase in average height seen in many developed nations over the past 150 years or so is probably due mostly to better diets rather than natural selection.

    Cold adapted.

    Natural selection may have favored the Buriats' broad skulls.


    Even very early evolutionary changes in the hominid line were not necessarily due to natural selection. Take the hominid face, which has changed dramatically in the past 3 million years from the heavy-jawed mugs of the australopithecines to the relatively small and gracile skulls of modern humans. Anthropologist Rebecca Ackermann of the University of Cape Town in South Africa and anatomist James Cheverud of the Washington University School of Medicine in St. Louis, Missouri, analyzed hominid faces over time, using formulas that model natural selection as well as random genetic drift, in which some traits or alleles become more common simply through chance. They concluded last December in the Proceedings of the National Academy of Sciences (PNAS) that natural selection probably drove the evolution of facial form up to the birth of early Homo. But they also found that genetic drift could explain most of the changes in the human face after the birth of Homo about 2.5 million years ago. “Selective pressures on the face may have been released” when humans began using tools and so did less biting and chewing, says Ackermann.

    The take-home lesson, she says, is that “genetic drift has played an important role in shaping human diversity. This is evolution, too.” Drift has continued to shape modern human faces and skulls in the more recent past, according to other studies. For example, researchers have examined regional differences in head shape—parameters such as width of the skull, height of the nose, and length of the jaw—to see whether certain traits were favored by natural selection in response to differences in climate or environment. In most cases, the differences among populations turned out to be no more than expected due to random drift. But there are a few exceptions: Anthropologist Charles Roseman of Stanford University in California last year reported in PNAS that the skulls of the Buriat people of Siberia are broader than predicted by random drift. Broad skulls have smaller surface areas and so may be an adaptation to cold climates. That fits with previous work by anthropologist John Relethford of the State University of New York College at Oneonta. Relethford concludes that random drift and migration can explain cranial differences in “most cases,” with the exception of people like the Buriat and Greenland Eskimos, who live in very cold environments.

    Battle for survival.

    AIDS and other deadly diseases may spur a rise in resistant gene alleles.


    Although the evolution of measurable traits such as modern human skull shape may be due to random drift, some changes in human body form may have more to do with cultural and environmental factors such as diet. “Over the past 10,000 years, there has been a significant trend toward rounder skulls and smaller, more gracile faces and jaws,” notes anthropologist Clark Larsen of Ohio State University in Columbus. Most of the change, says Larsen, is probably due to how we use our jaws rather than genetic evolution. With the rise of farming, humans began to eat much softer food that was easier to chew. The resulting relaxation of stress on the face and jaw triggered changes in skull shape, Larsen says. He adds that the dramatic and worldwide increase in tooth malocclusion, tooth crowding, and impacted molars are also signs of these changes: Our teeth are too big for our smaller jaws. Numerous studies show that non-Western people who eat harder textured foods have very low rates of malocclusion, he notes. Similar changes are found in monkeys fed hard and soft diets. “With the reduction in masticatory stress, the chewing muscles grow smaller, and thus the bone grows smaller,” Larsen says. “It is not genetic but rather reflects the great plasticity of bone. It is a biological change but heavily influenced by culture.”

    Signs of selection

    Even if random drift and other nongenetic forces have helped shape modern humans, there is growing evidence that natural selection has also played an important role, even if its effects have been more subtle. Human evolution researchers are now mining the riches of genomic data to spot genes subject to recent selective pressures (Science, 15 November 2002, p. 1324). Geneticists have a large arsenal of “tests of selection” at their disposal, all of which exploit the genetic diversity of human populations to determine whether individual alleles or larger blocks of the genome—called haplotypes—are behaving as would be expected if they were only subject to random drift and were not under selection.

    Some tests look for evidence that mutations in an allele that alter the protein it codes for have been favored over those that cause no change; others examine whether certain alleles are more common than expected. A fairly new and powerful approach compares the frequency of an allele in a population with the genetic diversity within a haplotype to which it belongs. If the allele is common due to random drift over a long time, the adjacent region of the genome should show considerable variation due to genetic recombination, the exchange of DNA between chromosomes during meiotic cell divisions. But if the variation is less than expected, the allele may have risen to high frequency in a much shorter period of time—a telltale sign of selection. “These tools are powerful,” says Lahn. “Where we are lagging behind is in good data.”

    By deploying such methods, geneticists have identified more than two dozen genes that appear to have come under selective pressures since the rise of Homo, and several of them may still be subject to such pressures today. Some of these favored alleles apparently arose at highly critical periods in human evolution. Such is the case of FOXP2, the so-called speech gene, which is implicated in the ability to talk, shows signs of strong selection, and arose no more than 200,000 years ago, coinciding closely with the first appearance of Homo sapiens (Science, 16 August 2002, p. 1105). Other genes under selection are linked to cognition and behavior, and still others are involved in defense against diseases such as hypertension, malaria, and AIDS (see table, below).


    In some cases, the new tests for selection have helped nail down long-suspected cases of evolutionary adaptation. One classic example is lactase persistence, the inverse condition of so-called lactose intolerance. Most adults cannot drink milk because they produce little lactase, the enzyme that breaks down lactose, which is the major sugar in milk. But a sizable number of people can, and their geographical distribution correlates closely with the spread of domesticated cattle out of the Near East. Thus, more than 70% of Europeans, who have a long history of drinking milk, have lactase persistence, as do some African pastoralists. In contrast, the percentage is very low in most of sub-Saharan Africa and Southeast Asia.

    Last year, researchers clinched the case for selection at the lactase gene. A team led by genome researcher Joel Hirschhorn of Harvard Medical School in Boston identified a haplotype more than 1 million nucleotide base pairs long that includes the lactase gene and confers lactase persistence on people who carry it. This form of the haplotype is found in nearly 80% of Europeans and Americans of European ancestry but is absent in the Bantu of South Africa and most Chinese populations. Hirschhorn and colleagues concluded from the unusual length of the DNA block that it is young, because it has not yet been broken up by genetic recombination. They calculate in the June 2004 issue of the American Journal of Human Genetics that this haplotype came under very strong selective pressure beginning between 5000 and 10,000 years ago, corresponding to the rise of dairy farming. Thus a cultural and technological change apparently fostered a genetic one. “This is one of the best examples of recent selection in humans,” says Tyler-Smith.

    Although being able to drink milk as an adult has its pleasant side, as any chocolate-shake lover can testify, most people in the world get along fine without the beverage. Yet in some cases, having a certain allele can be a matter of life or death. Thus, the genes most likely to be under strong selective pressure today are probably those involved in providing resistance to infectious disease, says Sarah Tishkoff, a geneticist at the University of Maryland, College Park. “In Africa, people are dying daily [of infectious disease], and those who have genotypes that confer some resistance are going to have more offspring. That is natural selection in action.”

    AIDS and malaria are arguably the worst scourges of humankind today, and they may both be exerting selective pressure on African genomes. Several genes have alleles that provide resistance to malaria, including those that code for hemoglobin C and an allele of the so-called Duffy blood group found only in sub-Saharan Africa; accumulating evidence suggests that they have both been under recent selective pressure. Four years ago, Tishkoff and colleagues showed that two different alleles of a gene called glucose-6-phosphate dehydrogenase (G6PD) have also been favored by strong selective pressure. The mutant alleles, A and Med, are found only where malaria is or recently was a problem and offer resistance against malaria, although they can cause blood diseases.

    Tishkoff and her co-workers used the known geographical variations in the G6PD gene to estimate that the A allele probably arose in Africa about 6300 years ago and then spread rapidly across the continent; the Med allele, found in southern Europe, the Middle East, and India, is estimated to be only about 3300 years old (Science, 20 July 2001, pp. 442 and 455). These estimates are consistent with archaeological evidence that malaria only became a major health problem after the invention of farming, when the clearing of forests left standing pools of water in which the vector for the disease, the Anopheles mosquito, could breed. Thus a cultural change again led to a genetic one.

    The case of AIDS, and the virus that causes it, HIV, suggests that the selective advantage of a gene can shift over time. As HIV infects T cells in the blood, it docks onto a cell surface receptor called CCR5. In the mid-1990s researchers discovered that a mutation in the CCR5 gene provides strong protection against AIDS in homozygotes, people who have two copies of the protective allele. The mutation, called delta 32, is found in up to 13% of European populations but is extremely rare in other groups, including Africans. Researchers dated the origins of the delta 32 mutation in humans to about 700 years ago and concluded that a strong selective event resulted in its spread; this finding was confirmed in 2001 using sophisticated selection tests.

    Yet because the AIDS epidemic dates only from the late 1970s at the earliest, researchers believe that the selective pressure on the delta 32 mutation must have been from some other factor. Researchers have debated whether the plague or smallpox, both of which ravaged European populationsithe past, is more likely, although some recent studies have leaned toward smallpox.

    Baby boost.

    Women with an inversion in this region of chromosome 17 have more children.

    CREDIT: NATURE GENETICS 37, 129-137 (2005)

    Icelanders evolving?

    Although researchers scouring the human genome for signs of natural selection have uncovered a few examples, direct evidence that a particular allele actually boosts reproduction—the sine qua non of natural selection—is hard to come by in humans. But that's just what researchers were able to do in one dramatic study in Iceland. For the past several years, scientists at deCODE Genetics, a biotechnology company based in Reykjavik, Iceland, have been gathering genetic information on the nation's 270,000 citizens, in a government-approved effort to isolate disease genes (Science, 24 October 1997, p. 566). In the course of this research, deCODE researchers discovered a variant of human chromosome 17 in which a 900,000-nucleotide-base-pair stretch of DNA was inverted; this inversion was associated with a previously identified haplotype called H2, which they estimate arose 3 million years ago. H2 carriers make up about 17.5% of Icelanders and 21% of Europeans, but only about 6% of Africans and 1% of Asians.

    To see whether the relatively high frequencies in Europeans represented natural selection, the team genotyped 29,137 Icelanders born between 1925 and 1965. When these data were correlated with the island's extensive genealogical database, the evidence for positive selection was stunning: As the team reported in the February 2005 issue of Nature Genetics, female H2 carriers had about 3.5% more children than H1 carriers. “This study has large implications,” says anthropologist Osbjorn Pearson of the University of New Mexico, Albuquerque. “The European version of the H2 haplotype could sweep the entire human population if it conveyed the same reproductive advantage in other people and environments.” But deCODE CEO and research team co-leader Kári Stefánsson says the low frequencies of H2 outside Europe suggest that for some reason, its advantages are limited to that continent. “Why, I can't tell you,” he says.

    There are several genes in the H2 region, but it is not at all clear which ones cause H2 carriers to have more children; one nearby gene is implicated in pregnancy complications. The deCODE team is looking at the genes to see whether differences in expression might create the selective advantage. One lead, Stefánsson says, is that H2 carriers also show a higher rate of recombination during meiosis. In an earlier study, his team found that mothers with high oocyte recombination rates also tend to have more children, possibly because this genetic shuffling helps protect against errors in meiosis, which are a major cause of miscarriage in older mothers. H2 carriers also appear to live longer on average. “It is fascinating to think that there might be an advantage associated with a DNA variant at both ends of life,” Stefánsson says.

    Our evolutionary future

    To many researchers, the limited but growing evidence that natural selection is currently acting on the human genome means that humans are still evolving, even if in subtle ways. But can we actually predict the course of future evolution, à la Sykes's disappearing males or the vanishing blonds? Most researchers' predictions are considerably more narrow and cautious and are tied to known selective pressures.

    For example, researchers predict that delta 32 and other protective CCR5 mutations may become more common in populations widely infected with HIV, especially in Africa. “If there are no more advances in the treatment of AIDS and people continue to die, we would expect selection pressure to increase [the mutations'] frequency over time,” says Tyler-Smith, who adds that he sees “no reason why they should not go to fixation”—that is, replace all other alleles of the gene.

    Whether or not these patterns will make a significant difference in the way humans look or live is another question. “There will be minor fluctuations over time and space in the makeup of local human gene pools as humans respond to local conditions,” predicts Tattersall, “but they won't be directional. I find it hard to foresee that under current conditions a qualitatively new kind of human is ever likely to emerge. But if conditions change, all bets are off.”

    Evolutionary predictions are tied to speculation about just what kind of environment we may face. Some researchers suggest that changing climate conditions may diminish the benefits of culture and medicine, creating a new era of natural selection. “There has been a relaxation in selective pressures in industrialized societies,” says evolutionary geneticist Peter Keightley of the University of Edinburgh, U.K. “But our ability to sustain that relaxation is probably temporary. We are using up our energy resources, our population is growing, and the climate is changing. All this is bound to lead to greater difficulties and renewed selective pressures.”

    Despite such concerns, however, most scientists remain leery of long-term forecasts, in part because of the way evolution works. “Evolution is not directed towards a goal,” says Tyler-Smith. “It always takes the short-term view, operating just on what allows us to survive and reproduce better in this generation.” For now, predicting humanity's evolutionary future may be little more than crystal ball gazing—better suited to science fiction than scientific research.


    Teaching Qubits New Tricks

    1. Charles Seife

    A novel approach to storing information could give computers with near-magic powers a boost toward reality

    Quantum computers will shatter the encryption that makes Internet commerce safe, search databases at unthinkable speeds, and crank out ciphers that nature itself guarantees secure—if they can be built. For years, scientists thought that would never happen because the same laws of physics that make quantum computers so powerful seemed to make a practical prototype impossible. But in 1995, when they discovered a means of preserving fragile quantum information despite those laws, quantum computing took a step closer to reality. The heart of the discovery was a way to correct errors in quantum information without destroying the information itself. These so-called quantum error correcting codes lie at the heart of quantum-computer research.

    Now physicist Ray Laflamme and colleagues at the University of Waterloo in Ontario, Canada, have mathematically reframed quantum error correction in a way that shows that seemingly distinct approaches to it are really the same. This insight could make quantum error correction more efficient and may well push the field toward a much deeper understanding of the limits of quantum information.

    “I think this is a very nice advance,” says Peter Shor, a mathematician and physicist at the Massachusetts Institute of Technology. “Whether it's a giant leap or just a substantial step forward remains to be seen.”

    Information, whether it's classical bits stored on silicon or quantum “qubits” inscribed on a cluster of atoms, is extremely perishable. Nature spreads it throughout the environment, diluting it, filling it with errors, and making it unreadable. The ravages of time tend to flip bits and turn precious information into useless gobbledygook. For classical computers, the solution is simple: Make a backup. Then, if nature corrupts your original data, you can restore it from the copy. This is the most rudimentary form of error-correction, and every computer and digital communications device bristles with ever more sophisticated ways of ensuring that data gets stored or moves from place to place without being corrupted. The packets your computer sends over the Internet are padded with error-correcting information; the files on your hard drive are flush with extra data to protect from random bit flips; even your cell phone has means of detecting and compensating for damaged data that it receives.

    Better way.

    Ray Laflamme and colleagues showed that “qubits” of data last longer when not stored on quantum objects such as atoms.


    But in quantum mechanics, copying is impossible, thanks to the “no-cloning rule”: You can't duplicate information with perfect fidelity. The act of measuring a quantum object—such as an atom in a delicate state of superposition—destroys the original as you transfer its information to another medium. Any attempt to clone a chunk of quantum information is doomed to failure.

    As a result, many theorists believed that it would be impossible to correct errors in quantum information. Laflamme was one of the naysayers. “I tried to write a paper about it, saying that you would not be able to build a quantum computer,” he says. But a colleague scooped Laflamme and published first. “I was upset, so I decided to poke some holes in the argument,” he says.

    In the mid-1990s, Shor, Laflamme, and a number of other physicists began to realize that there was a way to correct errors without violating the laws of quantum theory. “What we were thinking at the time was that the way we encode information in physical systems—a qubit upon an atom or a photon—was not very reliable,” says Laflamme. Instead, scientists realized, they could spread a qubit over several quantum objects such as atoms or photons at once. The key was to store the information not on a single object but in the relationship among those objects; technically, the collection of objects shares a single quantum state that encodes the information. Unlike information stored on a single object, information inscribed upon such a collection can be made error-resistant without running afoul of the no-cloning rule, because it doesn't need to be copied or read.

    In a paper recently published in Physical Review Letters, Laflamme and colleagues took the principle of abstraction a step further. Instead of storing information on relationships between quantum objects, they argued, one should store it on the relationship among the relationships. “It's getting more abstract, getting further away from the physical system,” Laflamme acknowledges. “But the usual quantum gates can do this easily, and it has some very neat applications.”

    Using this “operator” formalism, Laflamme says, physicists can make error-correcting codes with smaller ensembles of atoms (or photons or other quantum objects) than ever before, thanks to the improved efficiency that the method allows. The new mathematical structure also enabled Laflamme to prove that several seemingly different quantum-computational methods for controlling errors are really the same. “Some other methods of error corrections were proposed that are more passive,” says William Wootters, a physicist at Williams College in Williamstown, Massachusetts. Instead of actively correcting errors as they occur, physicists can pick a setup in which, under certain conditions, the information they inscribe on the system is immune from errors. “It seemed to be a different approach,” Wootters says. “This paper shows that you can reduce the passive kind to the active kind.” That means that physicists might now be able to borrow powerful tools from each of these areas and apply them to the others.

    “We're not sure yet what the real power of this technique is,” says Laflamme. “We haven't found the killer application.” Nevertheless, it's clear that the abstract approach will give theorists a concrete ability to explore new facets of a decade-old idea. “It allows us to understand that quantum error correction is much richer than we had thought,” he says.


    Biologist Helps Students Get a Leg Up on Scientific Inquiry

    1. Yudhijit Bhattacharjee

    Scott McRobert's research on how temperature affects the development of toads gives second-graders a chance to dip into real science


    PHILADELPHIA, PENNSYLVANIA—Elizabeth places a small plastic cup filled with water on an electronic weighing machine and presses a button to adjust its reading to zero. With help from amphibian biologist Scott McRobert, the 7-year-old dips a fishnet into a bucket swarming with tadpoles and brings up a wriggling specimen that she then nudges into her cup. After recording the animal's weight on a white sheet, she holds the cup aloft against the light to examine the tadpole's body as it swims around. “It's in stage 2 of development,” she announces proudly to McRobert, observing that the animal's hind legs have just begun to bud.

    Elizabeth and her fellow second-graders at Friends' Central School here are helping McRobert investigate the effect of temperature on the metamorphosis of toad tadpoles. A professor at nearby Saint Joseph's University, McRobert has pursued the relationship between temperature and breeding success among amphibians for more than a decade. His assistants are usually graduate students. This spring, however, during a visit to the school to pick up his daughter, a second-grader here, he noted the toads in a small, shallow pond on campus. That discovery led him to pursue a second, educational goal in addition to his scientific one. “I want these students to not only learn about science,” he says, “but also give them the opportunity to be scientists.”

    The project began in April, when students transferred some of the eggs laid in the pond into two tanks kept indoors and began monitoring their development. As the eggs hatched and the tadpoles grew, developing first tiny hind legs, then knees and front legs, the students kept daily journals of the animals' size, weight, and stage of development. The day the first two tadpoles of the study completed metamorphosis and clambered to the rim of their individual cups, “you would have thought we'd elected a new president,” says Barbara Cole, one of two second-grade science teachers at the school. “It became a subject of hallway conversation among the staff.”

    As those tadpoles developed, McRobert monitored two other batches of eggs at his lab kept at two different temperatures. Last month he shared his findings with the students: The tadpoles in the school tanks had taken an average of 23 days to transform into tiny, frail toadlets, whereas those in the cooler tank at his lab had metamorphosed in 29 days and were, as a result, much larger.

    Those results weren't surprising. Nonetheless, McRobert says that the correlation between temperature and developmental rate for this species is significant scientifically. “This is a study I would have done with my graduate students,” says McRobert, who conducted a similar study on poison dart frogs in Costa Rica 10 years ago to document the optimum temperature for their breeding success.

    Catching them young.

    Scott McRobert and second-grade students with their research subject.


    Cole and McRobert say most of the 60 students involved in the experiment seem to have absorbed its fundamental message, namely, that animals undergo change after they are born and that environmental conditions can affect development. That insight showed up as the children learned about African animals in their regular science class, say their teachers. “When learning about crocodiles, the kids wanted to know if babies hatched in the sand stayed there or moved into the water,” says Loren Ratinoff. “They were clearly reflecting on what they'd learned during the toad project.” Previous classes, she noted, limited their questions to the size and color of adult animals.

    The project may have also nurtured deductive reasoning skills. When a toadlet failed to climb out of its cup after completing metamorphosis and drowned, one student speculated about a larger phenomenon at work. “Maybe the toadlets grow front legs because they need to get out of the water when they develop lungs,” he said. Another student wondered if tadpoles developed faster in warmer environments because “when it's warm, the water dries up, and they have no choice but to breathe in the air.” McRobert—who conducted the study on his own time and without additional resources—says he “would have been happy if one of my graduate students had said that.”

    The teachers say the study may have also given the students a more concrete understanding of what science is. “Scientists observe things and write things down,” according to one student. Another says she wants to become a scientist “because scientists make stuff to help the environment, and it's fun to make new stuff.”

    Regardless of what the students learn, the project is likely to help them academically, says Arthur White, a professor of science education at Ohio State University in Columbus. “The memory of the tadpole study could help many of these kids stick through the difficult science classes they will encounter in middle school and high school,” White says.

    McRobert is hoping for exactly that kind of outcome. “I am in science today because my fifth-grade science teacher allowed me to take care of a big tank of turtles he had at the back of his office,” he says. Perhaps one day, his daughter and some of her classmates will be able to say the same thing about their tadpoles.


    Embryo-Free Techniques Gain Momentum

    1. Gretchen Vogel
    1. Held in San Francisco, 23–27 June.

    Ethical concerns about research involving embryos have been driving the search for other ways to derive stem cells, and results may soon be on the horizon

    SAN FRANCISCO, CALIFORNIA—As Californians work to get money flowing into the stem cell initiative that the state's voters approved last fall (see sidebar, p. 241), political pressure is growing in Washington, D.C., to find ways to conduct such research without involving embryos. In theory, the problem is straightforward: A skin cell has all the same genes as an embryonic stem (ES) cell, but different patterns of them are turned on. Scientists would like to be able to control gene expression with enough precision to turn a skin cell, say, directly into a genetically matched line of ES cells.

    The perfect answer remains elusive, but many scientists believe that sometime in the coming decade, they will know enough about cellular “reprogramming” to bypass some of the steps required today. “In 10 to 15 years, we will induce transformation directly and will no longer need embryos or oocytes at all,” predicts Kevin Eggan of Harvard University.

    It may come even sooner, given mounting congressional support and recent scientific advances. At the June meeting of the International Society for Stem Cell Research, Eggan presented his team's latest work using human ES cells to reprogram the gene expression of human fibroblast cells—moving toward the goal of creating genetically matched pluripotent cell lines without using oocytes or creating a new embryo. The team used polyethylene glycol to fuse the two kinds of cells, forming so-called tetraploid cells with twice the normal number of chromosomes. When grown into cell lines, the fused cells behaved like ES cells, Eggan reported, expressing characteristic genes, differentiating into embryoid bodies in culture, and forming so-called teratomas in immune-compromised mice—even forming patches of hair on the normally bald animals.

    Scientists have known for several years that ES cells can fuse with somatic cells to produce stem cell-like hybrids (Science, 15 March 2002, p. 1989), and previous studies had shown that several key tissue-specific genes turned off in the fused nucleus while key embryonic genes turned on. But Eggan and Chad Cowan of Doug Melton's group at Harvard went a step further, using gene expression arrays for detailed analyses of the hybrid cell lines. The cells, they found, had an almost identical expression profile to that of normal ES cells and one very different from that of fibroblast cells. “There is no longer transcription of fibroblast genes,” Eggan says, “and there aren't any deficits of ES cell genes.” Apparently, he says, “the ES cell nucleus can win the battle” between the two sets of chromosomes.

    Eggan says the fused cell lines he described at the meeting were made with ES cells that Melton derived, and his work with them uses no NIH funding. He adds that the group has generated other lines using an NIH-approved ES cell line, and work with those cells would be eligible for NIH funding.

    “The data that he's generated are beautiful,” says George Daley of Harvard Medical School and Children's Hospital in Boston. “It establishes the principle that there are factors in the human ES cell that will reprogram. But the devil is in the details,” such as whether the ES cell's DNA is required to accomplish the reprogramming—and if so, whether it might be removed afterward to create a cell line with just the genome of the original somatic cell.

    New and improved?

    Markus Grompe hopes an alternate approach to cloning might be more efficient and avoid creating an embryo—pleasing scientists and politicians alike.


    Another option for reprogramming somatic cells that might be eligible for NIH funding comes from Markus Grompe of Oregon Health and Science University in Portland. It is a slightly different take on the idea of “altered nuclear transfer” that Stanford physician and bioethicist William Hurlbut proposed to the President's Council on Bioethics last fall (Science, 24 December 2004, p. 2174). In Hurlbut's proposal, a gene required for early embryonic development would be deleted or knocked out so that the nuclear transfer would produce a cell incapable of developing into a fetus. Some people objected to the idea, saying it would create disabled embryos rather than the “nonembryonic entity” Hurlbut described.

    In an editorial in the Wall Street Journal on 20 June, Grompe and bioethicist Robert George of Princeton University in New Jersey propose that instead of knocking out a critical gene, scientists could overexpress a gene, such as nanog, that is crucial to ES cells, either in the somatic cell or in the oocyte. The resulting fusion of the two would, in theory, produce a cell with the expression pattern of an ES cell rather than that of a just-fertilized egg—essentially, going directly from a somatic cell to a pluripotent stem cell without forming anything resembling an early embryo.

    Although some scientists dismiss this idea as “mere semantics” not worth the extra trouble, Grompe says the strategy might have practical benefits beyond its political appeal. Studies have shown that ES cells are better cloning donors than are more mature cells, he notes, so perhaps boosting the level of a key pluripotency gene in the somatic cell would prime the cell and make the process even more efficient.

    The idea has won the support of a number of conservative bioethicists, including those who expressed reservations about Hurlbut's technique. If Grompe's idea works, says Tadeusz Pacholczyk, a molecular biologist and priest at the National Catholic Bioethics Center in Philadelphia, Pennsylvania, the somatic nucleus would only be reprogrammed to a pluripotent state—able to become all tissue types in the body—without reaching the totipotent state in which a cell can form a complete new embryo.

    Support seems to be growing in Washington as well. On 30 June, Representative Roscoe Bartlett (R-MD) introduced a bill that would fund animal studies to test the ideas, and Rick Santorum (R-PA), one of the Senate's strongest opponents of embryo research, has said he might include funding for such work in an omnibus spending bill.

    “I would welcome any infusion of resources,” says Daley, “as long as it's not used as an excuse to further delay funding for the methodology we know works today,” such as the nuclear transfer techniques reported by scientists in South Korea. “You move ahead on all fronts. Scientists will in the end use what works best.”


    California Institute: Most Systems Go

    1. Constance Holden

    SAN FRANCISCO, CALIFORNIA—Hounded by lawsuits and threatened by legislation that some fear could cripple it, the California Institute for Regenerative Medicine (CIRM), created by the state's voters last fall, is nonetheless proceeding apace—if not exactly on schedule. “We're not going to flame out,” neuroscientist Zach Hall, the interim director, assured Science last week during the third annual meeting of the International Society for Stem Cell Research.

    “We have made tremendous progress,” claims Hall, who's been on the job for 4 months. CIRM now has some 15 employees, including neuroscientist and stem cell expert Arlene Chiu, recruited from the National Institutes of Health (NIH) in Bethesda, Maryland. Hall says CIRM, which is currently advertising for program and review officers, hopes to have 15 scientists on board within the next couple of years. Blood stem cell expert Stuart Orkin of Harvard University has agreed to chair the peer review working group that includes 15 non-California scientists and seven patient advocates.

    CIRM backers were clearly relieved last month when state democratic Senator Deborah Ortiz agreed not to press for a vote—at least for now—on a proposal to amend the rules governing the institute in ways that many believe would make it impossible to run. But hers is not the only impediment; the courts also have to rule on two lawsuits before the state can begin selling the bonds that will finance the initiative. Both suits—one that claims that CIRM is unconstitutional, and a second that argues that fertilized eggs should be treated as “persons”—are thought to have dim prospects of success. Nonetheless, the suits mean substantial delays. “I've heard anything from 6 months to 2 years,” says Hall.

    Although the $3 billion initiative appears to be mired down, its leaders are steadfastly upbeat. The deadline for applications for the first round of training grants was 1 July, and Hall says that CIRM still plans to award 200 3-year fellowships at 18 institutions in November, to the tune of $45 million. “We're going to go ahead and award them even if there's no [bond] money,” says Hall. Bay Area real estate mogul Robert Klein, who spearheaded Proposition 71 and continues to direct start-up efforts, told Science that the CIRM is actively looking for $100 million in “bridge” funding. He points out that he raised $28 million to pass Prop. 71, and now that he can promise that donations will actually go to research, he is confident he can drum up a lot more.

    Handout imminent.

    CIRM Director Zach Hall says the first grants will go out this fall.


    Meanwhile, the institute's governing board has set up a new legislative subcommittee to come up with “policy enhancements” that they hope will satisfy Ortiz and her backers. Recommendations are scheduled to go to the full board on 12 July. Hall believes most of the differences with Ortiz—who is concerned about conflicts of interest, public access to decision-making, and Californians' access to the fruits of the research—can be resolved. Originally, Ortiz wanted outside peer reviewers to publicly disclose all financial ties to biotech-related ventures. She appears to have relented on this point, says Hall, who notes that CIRM policies already go beyond NIH requirements by asking reviewers to list any companies in which they have more than a $5000 investment. But that information would not be made public.

    Ortiz also pushed for some grant deliberations to be made public—an action that horrified many scientists. Tampering with peer review “would cripple [CIRM's] ability to operate,” said Stanford biologist Paul Berg. But now everyone, including Ortiz, agrees that peer-review meetings should be closed, says Hall. He cautions, however, that if the real business goes on in closed meetings and the public meeting of the Independent Citizens' Oversight Committee just looks like a rubber stamp, the public may object. The solution, Hall believes, will be for the peer reviewers to give scores, just like NIH does, to grants recommended for funding and to worthy grants that are not recommended for funding. The final funding decisions will then be made at open meetings, allowing a patient advocate, for example, to make a case for a project that wasn't recommended for funding.

    The biggest sticking point is how to satisfy intellectual-property concerns while heeding Ortiz's demand that CIRM “ensure” that any new treatments be “accessible and affordable to low-income residents.” According to Hall, this concept goes beyond the institute's mandate—and in any case, no one knows how to ensure that a treatment will be affordable. The latest wording is that CIRM will “seek to” ensure affordability. But that is still problematic, says Hall, as it “presents another target for litigation.”

    But optimism reigns as institutions all over California ramp up their stem cell capabilities. Stanford University's 3-year-old center, for instance, plans to hire a half-dozen scientists and just lured ear stem cell researcher Stefan Heller of Harvard. The University of California (UC), San Francisco, which distributes a number of cell lines, is planning to establish an “embryo bank” to supply excess embryos, eggs, and sperm from fertility clinics to California researchers; it will also be sending scientists to South Korea to learn the nuclear transfer techniques of Woo Suk Hwang. UC Los Angeles plans to spend $20 million in the next 5 years to establish the Institute for Stem Cell Biology and Medicine, with 12 new faculty positions. Hong Kong philanthropist Li Ka Shing just donated $40 million to UC Berkeley for a new research center focused on emerging scientific fields including stem cell biology. And in southern California, four institutions—UC San Diego, the Burnham Institute, the Salk Institute, and the Scripps Research Institute—have formed the La Jolla Stem Cell Initiative. All will be vying for money from CIRM, which aspires to become the world epicenter of stem cell research.