• E-mail
• Print
• Comment
• Font Size
• Digg
• del.icio.us
• Discuss article

Grant Allows Researchers to Delve into Visualizing Viruses

Posted on: Tuesday, 30 August 2005, 15:01 CDT

Aug. 30--Two Virginia Tech researchers have been awarded a $400,000 grant from the National Institutes of Health to develop a way to visualize viruses spreading through cells.

If perfected, the technique could allow medical personnel to determine how dangerous a virus is, even without identifying it. In the event of a virus outbreak, it could mean the difference between finding a cure quickly and allowing the virus to spread.

Karen Duca, a research assistant professor at the Virginia Bioinformatics Institute, and Amy Bell, an electrical and computer engineer, have teamed up on a project to view a virus' path of destruction as it spreads through healthy cells. Duca and her team get the raw information, and Bell's group is developing ways to extract the important data.

The grant will pay for them to perfect the technique and pave the way for more research into this kind of virus identification.

But it's first things first. "The goal is to prove that the technique and the methodology work," Duca said.

Initially, that technique will indicate how nasty the virus is. Eventually it could lead to a new way of identifying it by the way human host cells react to the attack.

When a virus invades a body, cells respond to the attack in a variety of ways: by calling for antibodies, for example, and by alerting nearby cells.

They do this by releasing various proteins as messengers. As Duca put it, "The host is signaling to other host cells, 'Virus! Put up your defenses!'"

That means that, during an attack, levels of certain proteins increase in cells under siege. As the virus spreads, the spike in protein levels spreads with it.

"Cells aren't expecting to be attacked," Duca said. "They have a very low level of defense." But when the signal goes out they begin to buzz with activity.

Using a microscope designed for the task, Duca and her team can measure that activity. A single picture of those proteins can show how fast and far the virus moves. That information reveals whether it's minor (a common cold, for example) or something to be concerned about (such as Ebola).

Measuring those proteins involves using special stains that generate a color-coded image. The problem is, that image is far from crystal clear; it has "noise" caused by the microscope used to make it. The result could appear to be a muddy smudge. Getting useful information from that smudge was a problem for Duca.

But it wasn't a biology problem. It was an image-processing problem. And that's where Bell came in.

"I know some stuff about imaging and image processing that helps solve a biological problem," she said.

What she knows is how to extract the useful data from Duca's images.

"Early on, the question was 'How do we extract the good information and get rid of this noise?'" Bell said.

Bell and her students are developing filters -- mathematical algorithms to apply to Duca's images.

Much of the noise in those images comes from the limitations of the equipment. No lens is perfect, and no illumination is perfectly even. But by knowing the equipment, Bell and her team know the errors that are introduced. That means they can correct for them -- the way a hunter corrects for a gun sight he knows is slightly off-center.

Once the errors are worked into the formula, Bell said, "Then we can start to quantify the information in the de-noised image." Translation: Then they can get a clearer picture of the virus's activity.

Different kinds of virus will have different "signal patterns," which will allow them to be identified. At first that identification will be limited to "bad" and "not so bad" -- essentially, whether the virus spreads quickly or not. As the technique is refined, however, researchers may develop a library of virus "signatures" so they can identify a specific culprit in the event of an outbreak.

The two-year grant will fund the initial investigations into the imaging system and hopefully allow Bell and Duca to prove the usefulness of the technique. If they can do that, they hope to get a five-year, $1 million grant in 2007 to take it even further.

-----

To see more of The Roanoke Times, or to subscribe to the newspaper, go to http://www.roanoke.com.

Copyright (c) 2005, The Roanoke Times, Va.

Distributed by Knight Ridder/Tribune Business News.

For information on republishing this content, contact us at (800) 661-2511 (U.S.), (213) 237-4914 (worldwide), fax (213) 237-6515, or e-mail reprints@krtinfo.com.

Source: The Roanoke Times

More News in this Category