February 10, 2003 | Should scientists meekly sign away the copyright for their research articles upon publication? Can the obscene costs of some institutional subscriptions to specialty journals — a 2003 subscription to Elsevier Science's Brain Research runs $19,971 — be justified? Would scientific research benefit from electronic access to the entire biomedical literature, rather than just brief abstracts?

Two years ago, Stanford University researchers Pat Brown and Michael Eisen (now at University of California at Berkeley) launched the Public Library of Science (PLoS), a nonprofit organization with the ambitious goal of improving access to biomedical literature — and possibly changing the face of scientific publishing in the process. The movement was strongly supported by dignitaries, including former National Institutes of Health director Harold Varmus and National Center for Biotechnology Information director David Lipman, both of whom had helped create PubMed Central as a searchable digital archive of full-text journals at the National Library of Medicine.

In 2001, the PLoS took on the scientific publishing establishment by encouraging scientists to sign an open letter (www.publiclibraryofscience.org) in which they threatened not to subscribe to, publish in, or review for any scientific journal that would not make its full text freely available after six months.

The PLoS movement quickly caught the attention of the biological societies and the major journals, including Science (published by the American Association for the Advancement of Science), Cell (Elsevier Science), and Nature (Verlagsgruppe Georg von Holtzbrinck GmbH). With a few exceptions, however, the major publishers held firm. Releasing content for free so soon, they argued, would jeopardize subscription revenue and would threaten the very existence of many prestigious academic and medical societies. In essence, the publishers bet that the majority of the 30,000 petitioners — especially students and postdocs early in their careers — needed the glamour science journals more than they realized.

A few concessions were made: Proceedings of the National Academy of Sciences initially agreed to release its content just three months after publication, though it has since reverted to six months. Science opted for a 12-month release period — not what the PLoS had lobbied for but better than Nature, the content of which, with a few key exceptions such as the human and mouse genomes, remains accessible to only subscribers. The boldest move was that of BioMed Central, the London-based open-access publishing company, which debuted a wide range of online journals, capped last summer by Journal of Biology, ostensibly designed to challenge the Big Three. So far, however, the Journal experiment has fizzled, publishing a grand total of two (unexceptional) research articles in six months.


Spreading the Word 
Given the puny response to its petition drive, the PLoS has decided to take matters into its own hands. Last December, the Gordon and Betty Moore Foundation — the philanthropy established by the co-founder of Intel Corp. — announced a five-year, $9-million grant to PLoS that will create two online peer-review journals: PLoS Biology and PLoS Medicine. Future titles may include PLoS Chemistry, PLoS Computer Science, and specialized publications in genetics, oncology, and so on. The organization has hired Vivian Siegel, the editor of Cell, to be its executive director.

PLoS authors will have to pay a charge of about $1,500 upon publication. Once published, however, the paper will be freely accessible to all via the Internet. Authors will retain the copyright but can irrevocably license it to third parties to redistribute as they see fit (with proper attribution). The Howard Hughes Medical Institute, which supports more than 300 of the leading biomedical researchers in the United States, has pledged to cover its investigators' publication costs.

Pat Brown offers several persuasive examples of ways in which open access can benefit research and public health: "When a woman learns she has breast cancer," he says, "she deserves to be able to read the results of research on her treatment options that her own tax dollars have funded. A physician in a public clinic in Uganda ought to have the same access to the latest discoveries about AIDS prevention as a professor at Harvard Medical School. And a precocious high school student in Gary, Ind., who wants to read about the latest discoveries from NIH-sponsored research in cell biology shouldn't have to pay thousands of dollars for journal subscriptions."

This utopian goal is a laudable one, for these reasons and more. Bioinformatic analysis of the full biomedical literature could reveal untold insights into signaling pathways and biochemical networks. Some examples have already appeared from Peer Bork's group at the European Molecular Biology Laboratory in Heidelberg, Germany, among others, revealing unappreciated links between proteins that can be verified by experimentation.

PLoS should have a strong impact when it starts publishing later this year. However, no matter how star-studded the editorial board, it remains a perilous proposition for young scientists to publish the results of years of hard work in a journal that might not be around in a few years.

But the publishing world is hearing footsteps. Last month, Nature said it would allow authors to retroactively retain copyright of their published papers. And we can only hope that the PLoS condemns the hundreds if not thousands of arcane specialty journals, which serve little practical purpose other than inflating publishers' bottom lines.

Kevin Davies, Ph.D.
Editor-in-Chief
Bio·IT World