A government agricultural officer prepares to vaccinate chickens from the bird flu in Indonesia. (Binsar Bakkara/AP)

Two scientific papers that describe experiments with a virulent and contagious bird flu virus should be published in uncensored form, a committee of scientists advising the federal government said Friday.

That recommendation by the National Science Advisory Board for Biosecurity reverses one the committee made in January, when it asked two journals, Science and Nature, to hold off publishing studies about the lab-engineered strains of the H5N1 influenza virus.

The about-face came after the heads of the research teams — one Dutch, the other American — clarified their work and provided new information on its possible importance at a two-day meeting of the committee in Washington.

While the studies could still be used by terrorists or mischief-makers, the committee said in a written statement that “the additional information changed the board’s risk/benefit calculation.”

Two facts appeared to sway the 18 voting members, according to people on and off the committee not authorized to speak on the record.

One is that the papers don’t provide step-by-step directions for how to make the engineered H5N1 strain. Specifically, they don’t provide a final list of mutations that made the bird flu easily transmissible in mammals, which it isn’t naturally.

“The data described in the revised manuscripts do not appear to provide information that would immediately enable misuse of the research,” the committee wrote.

The second fact is that new surveillance shows that “wild” H5N1 viruses circulating in chicken flocks overseas contain mutations similar to ones in the lab-engineered strains. Consequently, publishing the papers would give public health officials information that would help them identify wild H5N1 strains evolving in an especially dangerous direction.

“Global cooperation . . . is predicated upon the free sharing of information and was a fundamental principle in evaluating these manuscripts,” the statement said.

The committee voted unanimously to recommend publication of the paper by Yoshihiro Kawaoka of the University of Wisconsin. It split 12 to 6 in a vote on the paper by Ron Fouchier of Erasmus Medical Center in Rotterdam, the Netherlands.

Fouchier has submitted his manuscript to Science. That journal’s editor, Bruce Alberts, said Friday he was “pleased” by the committee’s decision to recommend publication of Fouchier’s paper “in an unredacted form.” No publication date has been set. Editors at Nature, published in London, could not be reached.

The H5N1 influenza virus emerged in Hong Kong in 1997 and has circulated indolently since then in East Asia, Central Asia and parts of Africa. The number of confirmed cases is 598, of which 352 (or nearly 60 percent) have been fatal. The latest victim was a 17-year-old Indonesian boy, who died March 9.

Most of the victims have been poultry workers in close contact with sick birds. In only a few cases does it appear that person-to- person infection occurred.

The controversial research was paid for by the National Institutes of Health. The purpose was to determine what changes wild H5N1 flu virus would have to undergo in order to become easily transmissible between human beings — a trait that would make it hugely more dangerous if it remained lethal as well.

Both labs apparently achieved the goal. The experiments were done using ferrets, which is the lab animal that most closely resembles human beings in its response to flu.

In at least one of the experiments, scientists engineered “starter mutations” into wild H5N1 and then repeatedly infected ferrets, where the virus evolved further, gaining new mutations that eventually made it easily transmitted.

When the advisory committee in December asked the journals to hold the papers back, many of the people who had read Fouchier’s paper believed his final virus was lethal to ferrets when transmitted through the air. At a meeting early this month, he clarified his results. The strain was fatal when sprayed into the lungs of the animals, but not when they sneezed and it traveled through the air in microscopic “aerosols” containing much less of the virus.

Although the notion that some research can be put to both beneficial and nefarious use is not new, the question of what to do with the papers on the engineered bird flu virus took much of the scientific community by surprise. As the biosecurity committee debated the matter, Fouchier and Kawaoka agreed to a 60-day moratorium on further experiments. The World Health Organization convened a two-day meeting in February to mull over the issue. The Royal Society, in London, will hold another meeting next week.

Earlier this week, the Obama administration asked federal agencies to inventory all the research they conduct and sponsor that involves 15 specific pathogens that “pose the greatest risk of deliberate misuse with most significant potential for mass casualties or devastating effects to the economy” and report back in 90 days.