Each of the pandemics appears to have the same pattern when emerging in humans, with different genetic components floating around in people for a few years before a pandemic strain is detected. And the detailed computational analysis showed that different component genes of the viruses seemed to be different ages. "What this suggests is that it's not one virus coming in and mixing with the human seasonal strain to produce a pandemic strain," Smith says. "Rather, there are a number of reassortment events, where one gene comes in and mixes with the human strain, and then another gene comes in and mixes with the human in a stepwise pattern."

If the researchers are right, the 1918 flu may have even more in common with the current swine flu virus than scientists previously believed. And finding such a pattern among known pandemic strains holds implications for future surveillance. By looking backward, at which genes have caused prior influenza strains to turn lethal, the research may one day enable researchers to look forward too. "What this paper is saying is that we're actually in a position now to get hints about these viruses even years in advance," says Greg Poland, a vaccine and infectious-disease expert at the Mayo Clinic, who was not involved in the research. "I think it will inform surveillance efforts, I think it will inform vaccine development efforts, and I think it will eventually inform policy-making decisions."

In addition to keeping an eye out for influenza variants in humans, Poland and Smith believe that it's just as important to start doing deeper surveillance in birds and pigs, and on a much more extensive basis. And, Poland notes, knowing that the strains emerge slowly could help inform vaccine efforts as well.

"There's no reason we can't move away from [creating] a vaccine against what we think we know will circulate this year, toward including proteins from variants we suspect might become problematic in the future," says Poland.

Smith hopes that more full-genome sequencing will provide advance warning of which genes might show up in humans, and that a deeper look at the genomes will provide clues about where and why the animal-to-human transmission occurs. He also hopes that one day, the team's research could help change governmental approaches from pandemic preparedness to pandemic prevention. "But the problem is that we still don't know what it is about a virus that makes it pandemic," Smith says. "Is it mutation? Is it a certain combination of genes? These are things that we still need to look at."