DENVER, CO — New international standards for publishing papers on security-sensitive biological materials, released by journal editors Saturday (February 15), challenge scientists and their editors to withhold data that could aid the development of biological weapons.

The group of 32 journal editors entreated their peers to deal responsibly with safety and security issues and design new processes to ease effective review of these issues. Editors also should ensure that manuscripts provide sufficient detail for independent verification and remove information deemed dangerous to the public, the group advised.

Ron Atlas, president of the American Society for Microbiology, urged other editors to embrace these proposed guidelines, issued during a press conference at the annual meeting of the American Association for the Advancement of Science in Denver.

"What we've done is international: It can't be done by one country alone," Atlas said. "We are asking all journals to begin to protect [sensitive information]."

The call for standards caps a tense 16-month tussle between scientists and government security agencies since the deaths of five people from anthrax delivered in the US mail, starting in October 2001. Microbiologists have faced scrutiny by federal agents in search of suspects; restrictions in the Patriot Act, passed by Congress in 2002, have changed the requirements for handling 100 "select" biological agents, including anthrax. And a loosely defined new category of materials called "sensitive but not classified," is now subject to government inspection.

The heightened scrutiny worries some researchers, who fear the government will issue blanket publication restraints. To build a rapport with the national security experts likely to issue such restraints, Atlas organized a meeting of the National Academy of Sciences and the Center for Strategic and International Studies in Washington DC on January 9. The new standards result from that meeting. But Atlas and other editors denied that the standards represent an effort to keep government agents at bay.

"It was not intended to deflect any outside restraint…or censorship," Bridget Coughlin, managing editor of the Proceedings of the National Academy of Sciences, told The Scientist. "It was an internal process."

Nevertheless, members of a panel of editors, scientists, and national security experts who released the standards during the AAAS meeting, acknowledged the difficulty in defining what kind of information should be removed from published materials. Atlas compared the process to defining pornography. "I know it when I see it," he said.

Editors examining 14,000 papers to be published in ASM journals in 2001 and 2002 flagged 224 as potentially sensitive; of those, two papers were withdrawn. One author rescinded his own paper after deciding that it could be used to make dangerous weapons. Publication of another paper was withheld to allow the writer to delete a parenthetical comment akin to describing how to make exposure to a biological toxin lethal on a mass scale, Atlas said. In general, what Atlas termed "cookbook" information, which details how to make weapons, should be withheld.

Such scrutiny of individual works will be necessary even if all journal editors subscribe to the new standards, said John Steinbrunner, director of the Center for International and Security Studies at Maryland. "It's very important to set general standards," he added. "But they are not going to be enough unless there are detailed judgments [made by peer review panels]."

Those judgments should balance the benefits of publication against the risks, panelists said. An editorial by Nicholas R. Cozzarelli, editor-in-chief of the Proceedings of the National Academy of Sciences focused on a hypothetical paper describing how to make ciproflaxin-resistant Bacillus anthracis. "The potential for misuse might be argued to preclude publication," Cozzarelli wrote. Yet, because cyproflaxin is used as a prophylactic for anthrax, he continued, it's imperative to understand the resistance mechanism, so the paper should be published. "Any work of value to terrorists will also be of value in countering terrorism."

The standards codify what many editors already do to minimize risks in the publication of scientific material, Atlas said. By publishing them, the group of 32 hopes to assist editors in crafting their own guidelines.

"The real issue for editors is, 'What kind of process will be put in place?' " explained Donald Kennedy, editor-in-chief of Science.

Correction: When originally posted, this story reported that six, not five, people with killed by anthrax sent through the U.S. mail. The Scientist regrets the error.