Eterna is an open science platform leading to fundamental advances in RNA biomedical research.

Eterna has created current state-of-the-art technologies to aid in complex RNA design, which are now available for licensing

Questions? Contact licensing@eternagame.org.

Ribonanza+

The Ribonanza technology suite leverages ultra-high throughput chemical mapping data for enabling downstream tasks to predict RNA structure, experimental dropout, and degradation. The Ribonanza data set, RibonanzaNet deep learning models, and DNA templates are growing. Initial versions of these are freely available at links in the preprint; higher quality and larger resources are being made available through a license.

RiboTree-mRNA

Software for designing stabilized mRNA sequences using a stochastic Monte Carlo tree search algorithm.

Nucleologic

Design of RNA/DNA switches with complex functions through the MCTS algorithm

EternaBench

EternaBench is a database comprising the diverse high-throughput structural data gathered through the crowdsourced RNA design project Eterna, to evaluate the performance of a wide set of structure algorithms. Freely available on GitHub.

EternaFold

EternaFold is a multitask-learning-based model trained on the EternaBench data, which demonstrates improved predictions both on molecules from Eterna, as well as completely independent datasets of viral genomes and mRNAs. Freely available on GitHub and through the EternaFold webserver.

OpenVaccine-solves

OpenVaccine-solves is a database of mRNA designs for diverse therapeutic design challenges, compiled from Eterna participants, standard methods, and novel algorithms. It also includes tools to predict stability and structure-based metrics. Freely available for non-commercial use; commercial licensing available.

mRNA-hotfix

Patch superfolder COVID vaccines to code for SARS-CoV-2 variant strains. Freely available for non-commercial use; commercial licensing available.

DOWNLOAD

RNAmake

RNAMake is a toolkit for designing and optimizing RNA 3D structure. It allows the alignment between RNA motifs. These motif are small modular peices of RNA that are believed to fold independently, thus attaching them together with helix flanking both sides allows users of RNAMake to build large segments of RNA with a high success rate of forming the predicted structure in vitro. RNAmake v1.0 is freely available for non-commercial use; commercial licensing available.

EternaBrain

The Eterna project has collected over 1 million player moves by crowdsourcing RNA designs in the form of puzzles that reach extraordinary difficulty. EternaBrain is a multilayer convolutional neural network trained on this player data, which 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. Freely available on GitHub.

NEMO

Solving the RNA inverse folding problem, also known as the RNA design problem, is critical to advance several scientific fields like bioengineering, yet existing approaches have had limited success. NEMO combines a different technique, Nested Monte Carlo Search (NMCS), with domain-specific knowledge to create an algorithm that outperforms all prior published methods by wide margins and solves 95 of the 100 puzzles listed in a recently proposed RNA solving difficulty benchmark. Freely available for non-commercial use.

SentRNA

SentRNA is a fully-connected neural network trained using the eternasolves dataset. The agent first predicts an initial sequence for a target using the trained network, and then refines that solution if necessary using a short adaptive walk utilizing a canon of standard design moves. Through this approach, SentRNA can learn and apply human-like design strategies to solve several complex targets previously unsolvable by any computational approach.

EternaBot

A community of 37,000 nonexperts leveraged continuous remote laboratory feedback to learn new design rules that substantially improve the experimental accuracy of RNA structure designs. These rules, distilled by machine learning into an automated algorithm EternaBot, also significantly outperform prior algorithms in a gauntlet of independent tests.


Help Eterna invent medicine. Make a donation today.

TwitterFacebookInstagramYouTubeTwitchGitHub