For a virus to evade antibodies, all five antigenic sites would have to disguise themselves by mutating. The new finding led the researchers to believe the receptor-binding residues would also have to mutate, but not so much that the binding no longer works. "If the binding is abolished, the virus dies," said Ma, a Rice professor in bioengineering with a joint appointment at BCM.

Such dual-function residues are a likely bottleneck for the virus, he said, because they're under the tightest restrictions. Thus, they could be easier to track over time and may chart a path to predict future mutations that will aid in vaccine design.

"It becomes a weak link and provides us with a window into the virus that we can monitor," Ma said. "The virus's bottleneck is our opportunity."

Wang, an assistant professor of biochemistry and molecular biology at BCM who has long studied the structure and function of HA, has been involved in the project since it began and is now working to verify the results in vitro. She hopes confirming the computations will lead to more efficiency in creating vaccines not only for H1N1 but also for other strains of the flu.

"An underlying implication is that this may not be restricted to H1N1," Wang said. "It may apply to other influenza viruses as well. If studying viral evolution can help predict what will cause a severe problem in humans, you can actually pre-stock vaccines, which will save time."