What some scientists refer to as "viral sex" has been found to be the cause of many flu epidemics through research conducted at Penn State.

Interaction between two viruses leads to the development of new viruses, which contribute to influenza outbreaks, said Martha Nelson, a graduate student in the department of biology and the lead author of the study.

"Unusual genomes are being generated through reassortment, which occurs when multiple viruses co-infect a person," she said. "The viruses actually swap genetic material and create new viruses."

Using a database containing genomes of various influenza viruses, Nelson and fellow researchers studied the evolution of the H1N1 strand of influenza with sequences ranging from 1918 to 2006.

"Basically, we were looking at genomes and different evolutionary patterns. You could see that viruses were related to different genes and that unusual genomes were created by different viruses," she said.

She explained that data from previous years and data concerning which strains are growing or decreasing are also used in predicting reassortment.

"We have to select the vaccine in February, and the season doesn't start until October, which doesn't give much time for research," Nelson said. "What we're finding is that vaccine design is much more complicated than what was formerly believed."

In addition, comparisons with outbreaks in other areas can be used as a predictive factor.

"We can use data from the Southern Hemisphere to predict outbreaks in the Northern Hemisphere," Nelson said. "Often, new strains become globally dominant in Asia before hitting the United States."

Prior to this discovery, it was believed that flu epidemics were mainly caused by mutations; however, this research suggests that the cause of flu epidemics have more genetic diversity than previously thought.

"What we're finding is that we can have people that are infected with multiple kinds of influenza viruses," Nelson said. "In order to understand the virus, you have to expand surveillance."

While surveillance in the United States and Europe is thorough, such research is not conducted as well in areas like Central America and Africa.

"That's part of the problem because the areas where the epidemics start have the poorest surveillance," she said.

While it is known that novel viruses are created through reassortment events, Nelson explained that it is still unclear why certain genomes and combinations result in outbreaks. Through this research and future work, data will be available to enable better predictions.

"The more intensively you look, the more you find and the more complicated evolutionary dynamics are," Nelson said.