A divide has emerged between the United States and the rest of the world on whether to publish or keep secret the details of an engineered mutant bird flu virus that can pass in the air between animals, health experts said on Wednesday.

But that split could be resolved when new data, including how the disease is not as lethal as widely believed, is considered in an upcoming meeting of the US-based advisory panel which initially urged the details be withheld from science journals.

The saga began late last year when a panel of US scientists and biosecurity experts reviewed two US-funded studies that showed how an engineered bird flu, or H5N1 virus, could be transmitted in the air between ferrets in a lab.

The conclusion of the National Science Advisory Board for Biosecurity (NSABB) was that the data was too risky to be issued to the general public and could spark a deadly pandemic if the flu escaped or were unleashed by malevolent players.

However, a meeting of international flu experts in Geneva earlier this month came to the opposite conclusion, agreeing that the data should eventually be published in full, after more consideration is given to the potential risks.

In those talks, there was "new data and a significant amount of time for cross-discussion to clarify issues," said Anthony Fauci, director of the US National Institute of Allergy and Infectious Diseases.

Fauci, who was part of the Geneva talks, told a meeting of the American Society for Microbiology Biodefense and Emerging Diseases meeting in Washington that the NSABB would reconvene to discuss the newest data.

"There was obviously a disagreement in recommendation between the NSABB and the Geneva group and we wanted to make sure all the data and all the discussion is balanced between both," said Fauci.

International experts also pointed to the "futility" of trying to edit out some portions of the studies, he added.

In the meantime, US health authorities stand by the original NSABB stance that the research should not be published in full, and the US endorses just one element of the WHO consensus: to extend a moratorium on such research.

While NSABB members said it was too early to guess what the US advisory panel's next decision will be, one of the lead flu researchers, the Netherlands' Ron Fouchier of Erasmus MC, stressed that the dangers have been overblown in the media.

"The animals get a little bit of flu but they do not drop dead," he said of his lab experiments on ferrets, noting that just one in eight animals got sick from receiving the mutant virus at high doses.

"It is certainly not highly lethal if ferrets start coughing and sneezing on one another," he said.

Also, if ferrets had already been exposed to seasonal flu they were fully protected against the mutant bird flu in his studies.

H5N1 is known to be highly lethal in birds, and among known cases in humans it has killed nearly 60 percent of its victims, according to the World Health Organization.

"There are still serious debates about how fatal this virus is in humans. I think more research is needed," Fouchier said.

"I do not think the 60 percent case fatality rate is correct," he added. "Certainly this is not a virus that would kill half the world population as we have seen in the lay press time and time again."

A US team in Wisconsin made a similar laboratory advance in the mutant flu, which was set to be published in the British journal Nature, while Fouchier's research was slated for publication in the US journal Science.

Science editor-in-chief Bill Alberts said both journals are holding back the papers for now, but that there needs to be some way to get the data to legitimate researchers.

"It is my intention to try to comply with the NSABB recommendation, providing we can have a mechanism for selective access to the redacted information by those who need to know," he said.

"So the future is still unclear."

NSABB board member Michael Osterholm, director of the Center for Infectious Disease Research and Policy at the University of Minnesota School of Public Health, said he supports Fouchier's research but remains concerned.

"I am not personally worried about somebody in a cave somewhere," he said. "I worry about the garage scientist, about the do-your-own scientist, about the person who just wants to see if they can do it."

"We need to know much more about this," he added, calling the papers the "two most famous unpublished manuscripts in history."