# Diverse database and machine learning model to narrow the generalization gap in RNA structure prediction

## TL;DR

This paper introduces a diverse RNA structure database and a deep learning model, eFold, to improve RNA structure prediction accuracy.

## Contribution

The paper introduces eFold, a deep learning model and a diverse RNA database to enhance RNA structure prediction.

## Key findings

- eFold outperforms state-of-the-art methods on challenging RNA structure test sets.
- Incorporating diverse and complex RNA structures improves model performance more than simply increasing database size.
- The database includes 1098 microRNA and 1456 human mRNA secondary structures determined by chemical probing.

## Abstract

Understanding macromolecular structures of proteins and nucleic acids is critical for discerning their functions and biological roles. Advanced techniques—crystallography, nuclear magnetic resonance, and cryo–electron microscopy—have facilitated the determination of more than 180,000 protein structures, all cataloged in the Protein Data Bank. This comprehensive repository has been pivotal in developing deep learning algorithms for predicting protein structures directly from sequences. In contrast, RNA structure prediction has lagged and suffers from a scarcity of structural data. Here, we present the secondary structure models of 1098 primary microRNAs and 1456 human messenger RNA regions determined through chemical probing. We develop a deep learning architecture inspired from the Evoformer model of Alphafold and traditional architectures for secondary structure prediction. This model, eFold, was trained on our newly generated database and more than 300,000 secondary structures across multiple sources. We benchmark eFold on two challenging test sets of long and diverse RNA structures and show that our dataset and architecture contribute to increasing the prediction performance, compared to similar state-of-the-art methods. Together, our results reveal that merely expanding the database size is insufficient for generalization across families, whereas incorporating a greater diversity and complexity of RNA structures allows for enhanced model performance.

Length matters—secondary structures of long, diverse RNAs in RNAndria and deep learning model eFold improve prediction accuracy.