The news reports can’t always keep up with the growing totals.
Flu activity was high or widespread in 43 states by early January, prompting the Centers for Disease Control and Prevention to declare an epidemic with peak season still ahead. At least nine people died from flu in North Carolina during Christmas week. Hospitals and doctor’s offices can barely keep up with the number of flu patients.
That’s predominantly the result of H3N2, this season’s dominant flu strain in the United States. Now a report published Jan. 21 in the journal Cladistics provides evidence that influenza A subtype H7 – which is being closely monitored after the China-Taiwan H7N9 outbreak – isn’t just a consistently changing virus but features changing combinations of genetic segments.
Daniel Janies, the Grotnes Belk Distinguished Professor of Bioinformatics at UNC Charlotte and a co-author of the report, says it’s not enough for science to react to outbreaks once they hit because the genes in both H3N2 and H7N9 influenza have recently changed. In an advanced form of flu mapping, he and colleagues use new computing methods to combine viral genomic data and geographic metadata, creating visualizations of viral traffic throughout the world with the goal of getting ahead of these events.
“We’re trying to build the whole concept of medical intelligence – putting all of the information together and asking questions from that information,” he said.
Their “transmission graphs” enable public health scientists to identify hubs for the transport of diseases, as well as where they can take action that will have the most lasting effect on preventing the global spread of a disease.
“We don’t have a crystal ball by any means,” Janies said. “But we’re trying to use the information we have and extrapolate smart maneuvers so we can go from there.”
How mapping is done
This kind of flu mapping provides more comprehensive intelligence than others, Janies said. “It’s putting different kinds of molecular information in a geographic map, not just mapping the genotype (the genetic constitution of an organism).”
He says the map is similar to what’s on the TV news, with fronts of clouds moving across time and space and data coming from satellites or other radar.
His data come from genomics – the study of complex sets of genes and how they work together – and involving viruses and associated variables: “Where they’re isolated from, what animals they’re isolated from, what time they’re isolated, and so forth. We put it in using Google Earth or Google Maps – something that’s similar to a weather map.”
One important component is mapping the risk of known factors. “There are certain point mutations in the virus or in populations of the virus that can affect its phenotype and make it more pathogenic, and can make it also jump from animals to humans. There are other mutations that can make it resist drugs. There’s a whole variety of different aspects of the virus we can understand through genomics.”
Janies said that in a risk map, “it’s sort of layering that kind of information and putting that information in context. One of the first ones we produced was just showing where a specific point mutation was in time and space and showing how it was changing over time. We were able to see avian viruses sort of adapt to mammals as they go westward.”
The CDC is currently mapping H3N2, Janies said, albeit in a more simplified way.
He said the CDC uses molecular biology to determine which virus subtypes there are, and divides the country into regions. “Then they use pie charts to show which subtypes are in each region. We do some of that as well, but that’s still more of a point-to-point concept.
“The CDC is presenting this data in a very accessible way. They have a dashboard concept, such that people can look at it and get the gist. … We decided to use a weather map so we can get a big picture, but also at any point in that map a person who’s responsible for that certain jurisdiction can look at what’s happening in their area.”
That may be very different from what’s happening in, say, the next state over: “You might want to check that out and say, ‘By the way, in South Carolina they have drug-resistant influenza that’s resistant to this kind of drug, so maybe we should prepare in North Carolina.
“So it’s more of a medical, prepared, precise, regionally specific intelligence that we’re doing, rather than a dashboarding concept.”
2 new approaches
Janies talked about two aspects of this more comprehensive form of flu mapping and how it can help identify trends by making concepts of disease transmission networks.
“It’s good to know what’s going on in the other side of the world,” he said. In the recent report about H7N9, “We said, ‘We don’t know how H7N9 is going to break out of China, but we know what it’s made of.’ So let’s ask: Of all the genes that make up H7N9, where historically have those genes participating in those viruses circulated?”
He gave an example. “Your investment adviser might say, ‘Here’s how these stocks have performed in the past. There’s no guarantee that will continue to happen, but this is the past performance.’ So we looked at that in a geospatial sense, and we can identify places where viruses like H7N9 have traveled in the past.”
Another method examines the global transport of viruses or bacteria.
“We did this most recently with salmonella, a food-borne pathogen that causes gastrointestinal illnesses. It’s a bacterium, but the principles are the same. We look at how often the virus is traveling from Place A to Place B to Place C and so forth, and we look at where the hubs are in that network. … In this case, India is a hub for transport of that pathogen.
“So you can look at the periphery of the network and might treat the people on that periphery, but if you want to do something in a public health context and disrupt the global transport of that bacterium, the interventions in India might be most effective.”
Janies reminds us that just because an outbreak is stopped doesn’t mean the virus is gone for good. “The virus can be hanging around in places and hosts at which we are not looking,” such as chickens or pigeons. “Our tools enable researchers to know where to focus efforts for the best chance to break the (disease) transmission network.”