The Washington PostDemocracy Dies in Darkness

An addictive online puzzle game is turning its players into published citizen scientists

(Screen shot from EteRNA)

Three superplayers of an addictive online puzzle game have done something that Stanford University Medical School believes is unprecedented: They’ve become the first authors on a paper published in a peer-reviewed scientific journal based on their discoveries in playing the large-scale, online video game EteRNA. The group of three gamers learned about their subject — the “folding” of RNA into different structures — by playing the game itself.

Jeffrey Anderson-Lee, a co-author on the paper, published Tuesday evening in the Journal of Molecular Biology, said his formal training in biochem amounts to a high school biology class and a chemistry course as an undergrad. He’s a computer scientist by training. He’s also one of the more advanced players of EteRNA, a citizen science game developed to explore the rules governing how different shapes are formed. And while you might assume that EteRNA players are mainly engaged in low-level, repetitive tasks, the new paper demonstrates that, in some cases, the opposite is true: As players have advanced through the game, the game has advanced with them.

“It’s a very fun game to play,” Anderson-Lee, a computer scientist who works in IT, said. “Initially, when you’re not that skilled in the bio sciences and whatnot, you get points and advance quickly.” Introductory levels feature basic instructions, and build gradually as players beat each early level of the game. There’s a cheerful, triumphant animation that plays at the completion of each level, which helps to make the whole thing feel fun. “As you move on,” he added, “you learn about the science, and that has its own rewards.”

How a karma-seeking Redditor uncovered one of the world’s rarest atlases

Although you could always play EteRNA and learn more about this for yourself, the basics are as follows: RNA, a component of living cells that has long been widely misunderstood, can “fold” into different shapes, depending on the arrangement of subunits it has. The structure it forms has lots of implications for how the RNA communicates with other parts of the cell, which itself has implications for research on  RNA-based therapies targeting a whole bunch of different conditions and diseases.

EteRNA concerns itself with the rules that dictate how those structures are formed — conveniently, the problem seems to naturally translate into visual, interactive puzzles. Players can chat with each other in a box on the game’s landing page. That’s how Anderson-Lee and the other two co-authors from the game started working together to solve a new problem, based on their experience in the game: Which RNA “puzzles” are the hardest to solve?

Citizen science projects been around for awhile. Anderson-Lee said he got into them after finding out about SETI@home, which launched in 1999. And there have been plenty of other games, too, like EteRNA’s spiritual predecessor, Foldit, which creates puzzles out of the folding of proteins. Those projects gain their own following, and hardcore players, and scientists have used work from non-scientist participants before to develop research. What’s different here, with EteRNA, is that the idea for the paper itself came from the players, working on their own.

Eli Fisker, who Anderson-Lee said has “always been famous” for pursuing extensive documentation for games like these, created a whole bunch of Google Docs to serve as resources for other players who were exploring that game. That documentation seems to be the catalyst for the work that became the paper. Fisker joined up with another one of the co-authors to try to find out what puzzles were the hardest to solve, and the work attracted the attention of Anderson-Lee, who was playing around with a newly-introduced feature of EteRNA: the ability to write scripts to solve puzzles. He hoped to test his scripts on puzzles that he knew were hard for expert players to solve.

Eventually, Rhiju Das, a Stanford University School of Medicine biochemist who is the the principal investigator for the game and the senior author on the paper, noticed Anderson-Lee’s group had amassed an impressive collection of Google Docs outlining their work on the question of which puzzles pose the biggest challenges. He thought it might make a good topic for a paper and encouraged the group to think about how to frame their findings as one. “I’ve played the role of an adviser, much as I would to my PhD students,” Das said. “I also helped them devise supercomputer ‘experiments’ that would test their ideas — these players supplied 100 targets ordered by difficulty, and folks in my lab (who are the two other co-first authors) independently tested the difficulty of these targets.”

Sequencing the genome creates so much data we don’t know what to do with it

The potential application of their findings, which have essentially created a way to determine the “designability” of many RNA structures, is wide,  Das said. “There are numerous long-term efforts to design medical interventions based on putting together RNAs that fold into specific structure,” he said –for example, RNA interference or CRISPR genome editing. Their observations, which are sometimes counterintuitive to assumptions one might make about the difficulties of particular structures,  “could save RNA engineers quite a bit of cost and time and effort,” Das said.

The players, as outsiders to the scientific disciplines they are participating in, might have had the advantage of a fresh perspective here. As Anderson-Lee put it: “We were looking for patterns that were apparent to us, as non-biochem people.”

But Das challenged the assumption that their work was non-scientific. “In many or most respects the players are scientists,” he said. “Some of the players have been able to devote much more time to thinking about the project than academics — and since the players typically dive into the scientific literature and constantly test their ideas, I do think they should be considered legitimate ‘citizen scientists.’”

Liked that? try these: