Foldit Players Outperform Computer Models On Viral Protein Modeling

Players of the online game Foldit have been able to determine the optimal structure of a protein that could prove key to the development of antiretroviral drugs to combat diseases like AIDS. A report from the University of Washington [PDF] outlines how, in just three weeks, Foldit players generated a model for a protein in the Mason-Pfizer monkey virus (M-PMV), responsible for a simian form of AIDS, that was more accurate than multiple computer-generated refinements that have been in development for years. Foldit players worked together with teammates while competing against other teams to "fold" the highest-scoring 3D model of the protein, usually without any specific knowledge of the underlying biochemistry behind the concept. After the three-week online competition, researchers found that several of the solutions created by the Foldit players produced better results than all previous approaches, leading to the creation of a final refined model for the protein only a few days later. "The critical role of Foldit players in the solution of the M-PMV PR structure shows the power of online games to channel human intuition and three-dimensional pattern-matching skills to solve challenging scientific problems," the researchers wrote. "Although much attention has recently been given to the potential of crowdsourcing and game playing, this is the first instance that we are aware of in which online gamers solved a longstanding scientific problem." Despite this success, researchers warned that the game was not always so successful at finding optimal protein models. Since players often make only small changes to the initial models given, players sometimes centered on improved "local minima" that were far from the best possible result. The Foldit project is separate from Sony and Stanford's Folding@Home project, which uses spare PS3 computing cycles in a fully automated effort to find optimal protein structures.