Pandora's box in mythology was the container from which all the evils of the world were supposed to have escaped – leaving only hope behind. It has become a common metaphor to describe the dangers of unfettered scientific knowledge.

Nuclear physicists know about the dual uses of their research, which can provide nuclear power and inspire nuclear medicine. It can also be used for making weapons of mass destruction.

Good and evil can be seen as two sides of the same coin. Knowlege in itself may be pure, but its application can be used for the benefit, or misused for the detriment, of humanity.

And so it is with the study into the H5N1 bird flu virus carried out by Ron Fouchier at the Erasmus Medical Centre in Rotterdam. Dr Fouchier and his colleagues want to understand how easy it is for this bird virus to mutate into a highly infectious human flu virus – but so might a terrorist.

Dr Fouchier understandably wants to tell his scientific colleagues exactly how he managed to introduce five mutations into two genes of the H5N1 virus, and so convert it into an infectious agent that can spread easily through the air from one experimental animal to another.

Dr Fouchier's finding is the sort of knowledge that can help to design better vaccines and anti-viral drugs. He has also shown that the steps necessary to make the lethal transition from avian flu to human pandemic are far simpler than many experts believed.

Although Dr Fouchier carried out the work on ferrets, which are the best animal model of human flu, observers are in little doubt that an airborne H5N1 in this experiment is highly likely to become an airborne human flu virus. Why else would the study attract such interest from the US government?

Paul Keim, chairman of the US National Science Advisory Board for Biosecurity and anthrax expert at Northern Arizona University in Flagstaff, knows a thing or two about dual-use scientific research. He was after all the scientist who identified the strain of anthrax used in the poisoned letters sent in the aftermath of the 9/11 attacks.

Dr Keim said he cannot comment on the study in question because of the rules of confidentiality under which he has to operate. "We're looking intensely at the H5N1 issue. I can't confirm whether we are looking at the Erasmus study," he said.

But it is nevertheless clear that the biosecurity board wants curbs on what can be published in the open scientific literature about this mutated virus. Not complete censorship, but some kind of redaction of key details that could be critical to terrorists with access to biological research facilities.

Perhaps that could be the hope left behind in Pandora's box.

Alarm as Dutch lab creates highly contagious killer flu