The Triple Helix @ UChicago

Spring 2014

"A New Kind of Scientist: The Influence of Crowdsourcing" by Stephanie Diaz

 

How do you analyze large amounts of complex information quickly? Nowadays, with the power of the internet, the answer is videogames. 

Recently, scientists have been using video games to crowdsource people to analyze data. Before the use of video games, scientists relied on the collective computing power of people with internet access with projects like the SETI@home which involved downloading a program that analyzes radio signals from space when a computer is not in use. Now scientists can have people analyze data through videogames. 

Take for example Eyewire, a computer game designed by Sebastian Seung, a neuroscientist from the Massachusetts Institute of Technology. For years, scientists have looked for the elusive answer to the question of how direction selectivity works.[1] Direction selectivity is a phenomenon in which a signal is sent from the retina to the brain only if the movement of the object being seen is aligned with a path from the center of a neuron to a dendrite. After fifty years and with the help of over 120,000 people around the world, scientists have finally been able to propose an answer.[1,2] 

Scientists struggled to explain direction selectivity for uncovering the answer required analysis of high resolution images of retinal tissue to discover the neural pathways, a very labor intensive process. This process was very long and tedious and proved to be too complicated for computers to do.[2] Eyewire provided the solution to this problem by creating a game where players would be given a high-resolution picture of tissue and be tasked with tracing the neural network, specifically the starburst amacrine cells, photoreceptors, bipolar cells, and ganglion cells. Players would gain points based on accuracy and speed with which they analyzed these pictures. This game was incredibly successful - more than 120,000 players from 140 countries have played. 

Eyewire has allowed Seung and his team to create three-dimensional maps of the retinal neural network. This work of thousands of gamers has allowed scientists to discover the positions of certain types of cells in relation to each other and how those positions contribute to direction selectivity. Eyewire has not only helped answer a question that is fifty years old but also made gamers scientists.  

Eyewire is not the only game of its kind. The games Foldit and EteRNA also use similar concepts. Foldit allows gamers to fold together proteins. Points are awarded based on how energetically efficient the gamer synthesizes the protein, meaning creating an enzyme with the least amount of energy results in the most points. Together, gamers have come up with the structure of an AIDS-related enzyme, a Mason-Pfizer monkey virus retroviral protease, within three weeks[4]. These results were published in Nature, and several more papers are currently being written about what algorithmic discoveries have been made by gamers.[3,4] Foldit allows people to test all the possibilities. 

Similarly, EteRNA allows players to work with RNA molecules. Scientists will post specific challenges, creating a specific shape like a key or a specific structure that is a real RNA design problem. EteRNA players build their skill and intuition as to how to synthesize the desire RNA molecules. Every two weeks, the players vote on which enzyme they deem has the best design. Scientists will make it and send a picture of the actual molecule and information on its behavior back to the players.[4] Understanding how RNA works will allow scientists to manipulate cell functions and cure diseases. 

This is not to say that crowdsourcing is a perfect method. People can make mistakes, and scientists have to take this into account. For Eyewire, a team of neuroscientists review the work done by gamers and correct any errors made. For games like Foldit and EteRNA, mistakes are part of the trial-and-error method that allow gamers to create efficient syntheses and replicate enzymes. The games are not intended to replace the need for a scientific authority on a study so much as to get as many people involved in order to collect data quickly or to get as many minds as possible thinking about a problem. As is the cause with Foldit and EteRNA, crowdsourcing allows for people from different types of backgrounds to work on problems that require critical thinking skills and not necessarily complete understandings of the science behind it. 

While crowdsourcing will not replace scientists, it can bring science to a community level and allow for advancements that could not be done otherwise. Eyewire plans to expand into mapping parts of the brain related to the olfactory system, but this is just the beginning of a much larger trend. Other games, such as Phylo which helps identify similar sections of DNA between species through color matching, are allowing anyone to become part of the scientific process. 

References

[1] Boyle, Alan. 2011. "Gamers solve molecular puzzle that baffled scientists." NBC News, September 19. Accessed May 5, 2014. http://cosmiclog.nbcnews.com/_news/2011/09/18/7802623-gamers-solve-molecular-puzzle-that-baffled-scientists. 
[2] Palca, Joe. 2014. "Eyewire: A Computer Game to Map the Eye." NPR, May 5. Accessed May 5, 2014. http://www.npr.org/2014/05/05/309694759/computer-game-aides-scientist-mapping-eye-nerve-cells?utm_source=tumblr.com&utm_medium=social&utm_campaign=skunkbear&utm_term=nprnews&utm_content=20140505.
[3] Sutter, John. 2011. "Why video games are key to modern science." CNN, October 23. Accessed May 5, 2014. http://www.cnn.com/2011/10/23/tech/innovation/foldit-game-science-poptech/. 
[4] Peckham, Matt. 2011. "Foldit Gamers Solve AIDS Puzzle That Baffled Scientists for a Decade" Time, September 19. Accessed May 9. http://techland.time.com/2011/09/19/foldit-gamers-solve-aids-puzzle-that-baffled-scientists-for-decade/.  
[5] Cossins, Dan. 2013. “Games for Science” The Scientist, January 1. Accessed May 17. http://www.the-scientist.com/?articles.view/articleNo/33715/title/Games-for-Science/.

 
UChicago Triple Helix