This school year I participated in the Kansas Science Olympiad competition and, along with two other Northwest students, competed in the protein modeling event. After this event, my interest in proteins and their structure was heightened, and I delved into learning more about protein. Recently I discovered a game called “Foldit”. Developed in 2010 at the University of Washington (UW), Foldit is an interactive protein visualization environment that can be altered in order to predict the native structure of a protein. This application is an example of a fairly new trend in scientific research known as crowdsourcing, or employing “citizen scientists”, meaning the games help researchers in determining the actual state of the protein.
Protein structure results from the formation of secondary structures, primarily α-helices and β-pleated sheets, and the subsequent tertiary structure, which is the 3-dimensional conformation of the protein determined by side chain interactions. The function of a protein depends entirely on the shape that it takes. This fact is why researchers are so interested in structure. X-ray crystallography is a way to accurately determine structure, but it is time-consuming and expensive, so predictions of the proteins folding are made with the help of technology.
In 2005, a team led by biochemist David Baker at the University of Washington first developed Rosetta@home, which runs protein folding predictive algorithms using the combined processing power of idle home computers. When people noticed that Rosetta was making mistakes they asked Baker to devise an interactive folding environment. In 2008, Baker, with the help of the game science department at UW, created the game Foldit in order to address these concerns. Foldit allows people to manipulate and “solve” proteins using the spatial reasoning capabilities of the human brain. When comparing the efforts of Foldit players and algorithm-generated solutions, it was found that the players often matched or surpassed the accuracy of the computer-generated solutions. In Foldit, players use “recipes” to automatically find preferable manipulations much like Rosetta does but under human guidance and supervision. The most popular recipe in the game is Blue Fuse, which surprised the researchers considering it created solutions faster in the game than the algorithm used in the actual Rosetta project. By 2011 the game was credited in the journal Nature for accurately producing a protein model of a retroviral protease, which helps in the proliferation of the Mason-Pfizer monkey virus. This model could help in the designing of new anti-retroviral drugs to treat HIV.
This power of aggregating human brainpower and finding brilliant solutions from amateur and unlikely volunteers is not limited to just proteins. Crowdsourcing has been applied to research in RNA, neuroscience, astronomy, and more. One site, called Zooniverse, hosts hundreds of projects by researchers needing the collective efforts of the “citizen scientist”. With over 300 years worth of combined effort put into this site, it is an amazing tool for completing research quicker than ever before. The use of the “citizen scientist” in research is increasing public knowledge and improving the scientific literacy of our society. It has become all too common for people to deny the facts and evidence provided by sound scientific methods, and this cultural engagement in scientific matters might change the invalid assumptions people have and create a world in which people combine their efforts to solve the problems plaguing our world today.