Video game challenges players to solve medical mysteries

If you’ve ever thought playing video games was a waste of time, you probably never considered that those button-mashing skills could help cure the HIV virus or Alzheimer’s disease.

Enter Foldit, the game developed at the University of Washington that looks like a futuristic, 3D version of Tetris.

As the name suggests, online players compete for the high score by folding up unruly shapes.

But these shapes are actually protein strands and the folds are the microscopic curls that hold the key to biochemical expressions that run the gamut from a recipe for a vaccine to the speed of the body’s metabolism.

It’s a new approach to mass collaboration that researchers hope will solve biological mysteries – and actually be fun to play.

Developed in the university’s computer science and engineering department by doctoral student Seth Cooper with the help of associate professor Zoran Popovic, and biochemistry professor David Baker, the game went live to the public last Friday.

“We had a new person joining every 20 seconds,” Baker says. “There were 10,000 people by Sunday morning.”

Players signing up at Foldit’s Web site are not only playing the game, but creating social network-style profiles. Players can chat together over the site and work as a team to solve the jumbled proteins – of which there are more than 100,000 different kinds in the human body.

“The great thing is, the game is actually fun,” Baker says. “Almost everyone is very positive about it. They’re excited that there is a fun game to play that is going to help humanity in general.”

Baker had previously worked on a more passive type of mass collaboration to tackle the medical challenge of protein folding. He was behind Rosetta@home, an application volunteers run as a screen saver.

As the computer crunches away randomly at trying to fold up the protein, users can watch the action on their screens.

“People were e-mailing in telling us the computer was not making the right moves, and they thought they could do better,” Baker says. So the Washington team decided to come up with an interactive game based on Rosetta’s processing engine.

The game debuted at last week’s Games for Health conference in Baltimore, Md.

The conference was organized by Portland, Maine-based Digital Mill Inc. with the hope of combining the seemingly disparate worlds of videogame technology and healthcare.

“You’re bringing two people together at a cocktail party who normally would never have talked together, let alone started working together,” says Ben Sawyer, Digital Mill co-founder.

Sawyer was impressed by Foldit’s unique approach to mass collaboration.

Instead of relying on the brute force of computer processing, the game seeks to make use of the human mind’s intuitive ability.

“Before everything was done with distributed computing power, but some of these problems can be solved by humans too,” he says, “and perhaps better at times because sometimes there is a judgment call to make.”

Computers randomly guess the best solution for folding a protein and calculate the corrections needed from there, Baker explains. But if a human can use some intuition in guessing at the first step, the computer has less work to do.

There are so many possibilities for folding proteins that it would take all of the world’s computers centuries to find them. People’s innate ability to solve puzzles needs to be tapped, Baker adds.

“People with the high scores aren’t scientists, they are people who like to play videogames and have never seen a protein in their life,” the biochemist admits. “My 13-year-old son is much better at folding proteins than I am.”

Getting people engaged through a game they can play with friends is a good approach to tapping what Sawyer calls “human computation systems.”

The Foldit project might be seeking medical advances, but it is also answering questions about mass collaboration methods, he says.

“It’s an important project on the premise of creating frameworks and success points,” he says. “This project starts to try and answer some of the questions about how to get people involved.”

The number of gamers playing Foldit is key to the project’s success, Sawyer points out. The team has been very aware that developing the game play is just as important as developing the game’s ability to solve protein problems.

“They’re at the same level of obsession as most commercial gamers,” Sawyer says. “I see a lot of projects where they have a goal in mind, but they forget how important the means are.”

Still, Foldit is on par with games like Bejeweled.

New players to the game need to go through 20 minutes of tutorials that teach the game’s rules, which are the same laws of physics dictating a protein strand either curves into a squiggle or bends into a corkscrew.

But the rewards for a high score are great.

Players who solve problems in the game will have their protein shapes analyzed by researchers and could be given recognition for a scientific contribution.

“Anything that looks promising, say like a new vaccine candidate, we’ll make those and see if it works,” Baker says. “If they work, we’ll credit the people that found them.”

Right now the proteins players are folding have already been solved by researchers. The team wants to calibrate players’ ability to solve the problems before unleashing unsolved proteins in the near future.

The project is backed by some big name funders. They include the Defense Advanced Research Projects Agency, the Howard Hughes Medical Institute, Microsoft Corp. and Adobe Systems Inc.

Baker compares the project to the time when amateur astronomers helped space agencies track satellites blinking through the blackness of the night sky.

Not since then, he says, has the average Joe been able to contribute to science in a meaningful way.

It makes for some good chat room conversations.

“To see 500 videogame aficionados talking about protein folding is a pretty interesting sociological experience,” he says.

Share on LinkedIn Share with Google+
More Articles