Portal de Periódicos da CAPES

Commentary: Crowdsourcing, foldit, and scientific discovery games

2013; Wiley; Volume: 41; Issue: 2 Linguagem: Inglês

10.1002/bmb.20686

1539-3429

Graham R. Parslow,

Scientific Computing and Data Management

The web has created new possibilities for collaboration that fit under the terms crowdsourcing and human-based computation 1. Crowdsourcing applies when a task or problem is outsourced to an undefined public rather than a specific body. Human-based computation refers to ways that humans and computers can work together to solve problems. These two methods come together to accomplish the task of protein structure solving, given the amino acid sequence and the Foldit program 1. A journalistic window on what has happened is provided in the following extract. “Aristides is a typical 13-year-old boy. Under the nickname Cheese, he plays a game that involves twisting, pulling, and wiggling a 3D structure that looks a bit like a tree's root system. He manipulates different lengths of these snaking green tubes until they fit into the smallest volume possible. It may sound like a rather bizarre game—a distant 3D relative of Tetris, perhaps—but it is in fact a brilliant disguise for one of the toughest conundrums facing biologists today: how do proteins fold?” 2. The success of Foldit to harness the creativity of at least 57,000 regular players is shown by achieving the solution to a retroviral protease published in the journal Nature 3. The number of different ways even a small protein can fold is very high because there are so many degrees of freedom. Predicting structures is among the hardest problems in bioscience and easy to get wrong. Foldit exploits the twin challenges of puzzle-solving and the reward of playing competitively to fold proteins. The Foldit program has evolved over five years to exploit both the human psychology of reward for behavior and the power of computers to model systems 4. The twelve authors of Foldit come from multiple US campuses and combine strengths in computer programming, visual interfaces, and protein structure. The guiding principles have been that users should have fun and that they should believe that they are serving a useful function. The game also exploits social grouping and allows players to form teams that can cooperate to achieve better solutions. Some preconceptions were erroneous and “we learned not to expect the way that expert scientists view the problem to be the best way for players” 4. Likewise teachers can have inappropriate expectations and Peter Farley 5 has shared my experience of students showing little enthusiasm for conventional molecular image rendering packages. By contrast Farley reported that Foldit was used with enjoyment by all of his students and the experience improved their understanding of protein structure. Generalizing the outsourcing strategy we can see how in the last decade the world has come to share its intellectual property and individual strengths. Wikipedia has become the world's repository of encyclopedic knowledge. First rank universities are sharing their teaching online through Coursera, YouTube, and other channels. Even so crowdsourcing has had some spectacularly humorous failures 6, but the successes are increasing. As teachers, authors, and researchers the time is coming to divest responsibility and credit for a task from individuals, to achieve superior results through collaboration.