EternaBrain: Automated RNA design through move sets and strategies from an Internet-scale RNA videogame

Abstract

Folded RNA molecules underlie emerging approaches to disease detection and gene therapy. These applications require RNA sequences that fold into target base-pairing patterns, but computational algorithms generally remain inadequate for these RNA secondary structure design tasks. The Eterna project has collected over 1 million player moves by crowdsourcing RNA designs in the form of puzzles that reach extraordinary difficulty. Here, we present these data in the eternamoves repository and test their utility in training a multilayer convolutional neural network to predict moves. When pipelined with hand-coded move combinations developed by the Eterna community, the resulting EternaBrain method solves 61 out of 100 independent RNA design puzzles in the Eterna100 benchmark. EternaBrain surpasses all six other prior algorithms that were not informed by Eterna strategies and suggests a path for automated RNA design to achieve human-competitive performance.

Overcoming the design, build, test (DBT) bottleneck for synthesis of nonrepetitive protein-RNA binding cassettes for RNA applications

“…Specifically, the field is still lacking computational methods that will enable users to reliably design their system of choice without going through multiple time-consuming DBT cycles before arriving at the desired solution. Recent studies are emerging to tackle this challenge and to devise such methods for diverse RNA-, DNA-and protein-based applications, with varying degrees of success [1][2][3] . Perhaps the best known methods are the Cello algorithm and RBS calculator, which are limited to bacterial chassis 4,5 at the present time.…”

Engineering molecular translation systems

“…In a different direction, computational methods for 3D RNA design and Eterna, a crowdsourcing/machine-learning platform involving >200,000 global participants, are emerging as potential tools to facilitate ribosome design. Surprisingly, players of Eterna consistently outperform most RNA-folding algorithms and can be used to improve computational RNA design (Koodli et al, 2019;Lee et al, 2014). Recently, Eterna based on player-proposed rules and stringent experimental tests was shown to be the best available algorithm for RNA secondary structure prediction (Wayment-Steele et al, 2020).…”

A Comparative Study of Gameplay of Different Sets of Players in an Engineering Mapping Game

“…Over the years, GWAP has gained extreme popularity in the scientific world. Among various notable games falling under the umbrella of GWAP, Foldit [21], a protein structure prediction and design game and Eterna [22], an RNA molecular design game has fascinated many scientific minds. Both these gaming platforms have used crowdsourcing to tunnel in human intelligence and to use that intelligence in answering some unanswered questions in their respective domain.…”