This week's New Scientist reports on work done under the aegis of the US National Institutes of Health  but in both Rotterdam and Wisconsin, in which varieties of H5N1 flu have been synthesised that spread by water droplet between ferrets.  This is very serious news indeed, as if a coughing ferret can spread flu to another ferret, it is likely that humans could do it too.  This is critical, as the previous strains of bird flu could not spread between humans, and only with difficulty and by close contact from chicken to man.

It is serious news, because while the synthesised virus was made in a lab, no doubt after many permutations of the virus genome, the same experiment is going on out there, in chickens and wild birds. The wild experiment has less direction, but far, far more permutations are done every year.  So Nature is quite likely to come up with the jackpot - transmissible human bird flu.   I quote an expert from the article,  "1918 flu would look like nothing if this really got loose"   If the 1918 epidemic killed only 50 million people that was 3% of the world's population.

It is also serious news, because this event has sparked a debate on the justification of doing such work, and how it should be confined.  If the virus escaped from the lab, nothing could stop it infecting the whole world.  This is not idle, alarmist talk.   The SARS virus got out, and so did smallpox, before the only remaining samples were kept in the biological equivalent of a maximum security jail. 

The appropriate level of security for extremely dangerous pathogens is "BioSafety Level 4", but this is so expensive that to do so would confine H5N1 research to only a few labs i the world.     When Nature is likley to relase it anyway, do we need as many people working on it as possible?

See:  http://www.newscientist.com/article/mg21328473.400-we-need-to-fix-the-holey-biosafety-net.html