Penn researchers develop flu-prediction method

It comes every year, leaving those in its path sniffling, coughing, feverish — and in some cases, dead. Attempts in guessing the influenza virus’ next move have often been futile, until now.

Researchers at Penn are starting to develop a way to predict the next strain of the flu virus, a development with important public health implications.

Once infected by the flu virus, people are immune to that particular strain for the rest of their lives. However, the flu virus continually evolves, mutating its genes in order to infect new people each year.

“It’s constantly running this uphill race trying to escape the shadow that it has recently cast on the human population,” said Joshua Plotkin, professor of Interdisciplinary Studies at Penn.

Plotkin conducted the flu research with post-doctoral fellow Sergey Kryazhimskiy.

They were joined by researchers at McMaster University and at the Institute for Information Transmission Problems of the Russian Academy of Sciences.

“One of the big questions is to predict the next mutation that will spread,” Kryazhimskiy said. Their research tackles this question by using computer analysis of flu genomes from the last 40 years to observe the genetic “sites” where mutations have occurred and to predict where they may occur next.

“When you think about evolutionary biology, you think about dinosaurs, and you think of going back to the past,” Plotkin said.

But with flu research, “it’s prospective,” Plotkin said.

In the past, researchers focused on the parts of the virus that evolved the quickest. Plotkin’s research, however, noticed that mutations often come in pairs.

“If you know that site ‘A’ is often followed by site ‘B,’ and this year you observe a mutation in site ‘A’, you know pretty well that within the next few years there will also be a mutation in site ‘B’. It gives you some genuine predictability,” Plotkin said.

Influenza kills anywhere from 3,000 to 49,000 people a year in the United States, according to the Center for Disease Control and Prevention. Because of the time it takes to physically grow the vaccine, public health officials have to guess which strain of the virus to use for that year up to eight months in advance.

Each year the Food and Drug Administration specifies which vaccine strains to use based on analysis by the CDC.

Although the CDC has their own personal history for choosing the vaccine strains, Plotkin hopes they will take his research into consideration in coming years.

Right now, Plotkin is testing his computational research in a lab experiment run by Scott Hensley, professor of Immunology at The Wistar Institute. By using genetic techniques to engineer viruses, they can see how the viruses respond to specific mutations at various sites.

“The beauty in Josh’s work is that he really identified mutations that tend to go together. Josh’s study is definitely a really important study, and it’s good to have the computational stuff linked with some of the wet lab stuff,” Hensley said.

Please note All comments are eligible for publication in The Daily Pennsylvanian.