– Despite the heat of summer, some researchers are still preoccupied by cold and flu season, especially Scott Hensley of Philadelphia's Wistar Institute, who hopes that vaccines being developed for the 2015-16 flu season will provide more protection than last year's versions.

Last year's vaccine was only 19 percent effective in preventing medical visits due to flu-related complications, said the U.S. Centers for Disease Control and Prevention. Until now, researchers did not know why.

Hensley's group identified a hot spot on the flu virus where mutations in the virus' genome occurred last year. His team's work suggests that vaccine developers should look closely at this spot when creating the annual flu vaccine.

"Viruses don't accumulate mutations for the fun of it," said Hensley, whose work was published in the journal Cell Reports. "They acquire mutations in order to deal with the body's immune response."

Seasonal flu is a problem every year and contributes to more than 30,000 annual deaths in the United States alone.

One reason flu is so dangerous is that the body's immune system and the viruses that try to override it engage in an evolutionary arms race. Researchers develop vaccines that, when injected, produce antibodies that bind to viruses, protecting us from illness.

In response, the virus mutates to prevent the immune system from recognizing it, and the cycle continues. The flu virus especially can mutate very quickly, which is why people need a seasonal flu shot every year.

Hensley, an assistant professor at the Wistar medical center, said the flu strains from last season had at least five new mutations. One in particular significantly reduced the immune system's ability to detect the virus and protect people from the flu, he said.

"The thing about the flu is that it's so unpredictable, so selection of vaccine strains becomes an educated guess," Hensley said.

After the flu season peaks in December and January each year, the World Health Organization surveys what strains have developed. In February, experts decide where to focus their efforts on the next winter's vaccine. This decision is typically based on the most prevalent virus strains.

Last year, however, the problem was that strains with the mutations Hensley's group identified were present, but in such small amounts that they flew under officials' radar.

Something about those low-frequency strains caused them to predominate in the virus population. That something was their enhanced ability to "avoid human antibody responses generated against past strains" of the flu, Hensley said.

He added that developing vaccines "is not a perfect process, but by and large flu vaccines work well" and save many lives.