Into the fold

Posted on Jul 29, 2020 :: Feature
Jessica Thiel
Posted by , Insight on Technology Staff Writer

As he watched the COVID-19 pandemic turn increasingly grave, Adam Wilkins, system administrator at RB Royal Industries in Fond du Lac, realized there was a way his company could help.

A self-described computer nerd, Wilkins follows several tech communities online, where he learned about [email protected] (FAH). Based at the St. Louis School of Medicine at Washington University in St. Louis, FAH is a distributed computing project aimed at helping scientists develop new therapies to fight diseases.

FAH, which launched in 2000, invites individuals and organizations to donate their unused computer power to help research diseases and discover potential cures and treatments. It’s been used to research breast and kidney cancers, infectious diseases such as Ebola and Zika, and neurological diseases including Alzheimer’s, Huntington’s and Parkinson’s. And now it’s harnessing that power to study COVID-19.

By running simulations, FAH aims to learn how COVID-19 functions and find a therapeutic treatment that prevents it from spreading throughout the body.

RB Royal recently purchased a server with extra horsepower, and Wilkins knew it could be put to good use. When COVID-19 emerged, FAH released a program that’s designed for servers like the one RB Royal had just acquired. Wilkins visited the FAH website and installed the free software.

“I wanted to give back and we have the means to do it. It literally costs us nothing,” Wilkins says.

The Fond du Lac manufacturing comp-any wasn’t alone in wanting to help. In late March, when FAH debuted a way for individuals and companies to donate computer power to help run simulations to aid in COVID-19 research, it received an overwhelming response, with 700,000 operators joining — up from 30,000 before the pandemic.

As the FAH website puts it, “Each simulation you run is like buying a lottery ticket. The more tickets we buy, the better our chances of hitting the jackpot.”

Jonathan Gutow, a professor of chemistry at the University of Wisconsin-Oshkosh, says scientists have been using computational tools for more than 100 years, beginning with hand computations and later taking advantage of computers’ powerful microprocessors
starting in the 1960s and picking up steam in the 1970s and ’80s.

In his own work, Gutow studies the rhinovirus, which is responsible for about 50 to 80 percent of common cold cases. In his lab, he and his team used computational tools to figure out how a cold virus binds to a particular molecule on the cell surface, initiating entry of the virus into the cell. These were considered low-level calculations, and they used a group of 15 computers with massive processors running between three and six weeks and using all their processing power, he says.

Computational tools allow scientists to perform more sophisticated calculations, ones that in the past may have taken years to complete and now can be done in weeks, days or even hours, Gutow says. Their power lies in offering the ability to process in parallel, breaking a large calculation into a lot of little pieces.

“What [email protected] is doing is taking advantage of the fact that an awful lot of people and organizations have computers sitting around that are really not using much of their processing power,” Gutow says.

After a user installs the program, FAH sets up a background task on the computer and uses the processing power to try various tasks. The project often begins with the sequence of amino acids, molecules that connect together like beads on a string using the same chemical bond. A protein consists of one or more of these chains folded up in a particular shape, Gutow says.

Knowing the sequence doesn’t tell scientists the shape of the protein, and FAH is trying to figure that out. It sends out tasks for computers to try with the aim of calculating energy and eventually leading to an estimate of the shape of a protein. Proteins have common shapes, and starting with that knowledge and doing some “jiggling,” scientists can begin to make pretty good guesses, Gutow says.

“To get good results, you’ve got to try lots and lots of things. That’s why we’re seeing some quite promising progress on COVID-19,” he says.
Many people and organizations with expertise in bringing drugs and vaccines to market are working simultaneously and sharing information, suggesting the scientific community will make progress faster, Gutow says.

While he says launching a vaccine before the end of the year is an opti-mistic goal, there is some promise for treatments that could provide temporary immunity to heath care workers. One technique causes the body to produce antigens typical of the virus, triggering the production of antibodies to the virus. Another technique directly provides instructions to some cells in the body for producing the antibodies.

“That could make a huge difference in keeping nurses, hospital workers, orderlies and physicians healthy so they can take care of people,” Gutow says.

On the Web