RNA can be coaxed to bend, loop, and twist into a diverse array of structures, making it a key player in the growing field of molecular engineering. By designing and synthesizing an RNA molecule that folds into a desired shape, scientists aim to build novel drugs and sensors. Now, using data generated through the online RNA folding game EteRNA, a team of researchers have developed the best-yet algorithm for designing folded RNA molecules (1).
Algorithms to predict basic RNA structures—how simple loops form in RNA, for example—have existed for more than thirty years. But “the issue is that when you use these computational models, there ends up being a lot of trial and error involved,” said Rhiju Das of Stanford University, senior author of the new paper. “Most of the models don’t actually work in practice.”
In 2011, Das and his colleagues wanted to better predict which models replicated real RNA molecules in the lab, so they turned to citizen scientists by launching the online game EteRNA. Here, players are asked to piece together RNA nucleotides to build target shapes—acquiring points based on the complexity of the shapes designed. Das’s team then collects the results and tests players’ computational designs in the lab, letting users know whether particular designs worked.
“Computers got worse at designing structures with every puzzle, as they got more complex, but players were getting better with each challenge,” Das said.
Das started picking up new “design rules” by analyzing the EteRNA players’ strategies for building unique shapes. Most of the rules, he said, are fairly basic and make a lot of sense.
With the new rules in hand, Das set to work creating a new algorithm—dubbed EteRNABot—to predict RNA folds. In testing, the algorithm performed better than previous methods and almost as well as the players’ by-hand work.
The lab has now ramped up EteRNA—synthesizing thousands of players’ RNA designs every week—and moved on to asking new, more complex questions, such as how to design sensors that change structure when bound.
1. Lee, J., Kladwang, W., Lee, M., Cantu, D., Azizyan, M. et al. (2014) RNA design rules from a massive open laboratory. PNAS Early Edition January 23, 2013. doi: 10.1073/pnas.1313039111