Kirigami: large convolutional kernels improve deep learning-based RNA secondary structure prediction

TL;DR

This paper presents a novel convolutional neural network architecture with large kernels that significantly improves the accuracy of RNA secondary structure prediction, especially for complex pseudoknots.

Contribution

The introduction of a fully convolutional neural network with 11-pixel kernels tailored for RNA structure prediction is a novel approach that outperforms existing methods.

Findings

Achieves MCC over 11-40% higher than state-of-the-art methods.

Significantly improves pseudoknot prediction accuracy by 58-400%.

Outperforms current software on a standard test set of 1,305 molecules.

Abstract

We introduce a novel fully convolutional neural network (FCN) architecture for predicting the secondary structure of ribonucleic acid (RNA) molecules. Interpreting RNA structures as weighted graphs, we employ deep learning to estimate the probability of base pairing between nucleotide residues. Unique to our model are its massive 11-pixel kernels, which we argue provide a distinct advantage for FCNs on the specialized domain of RNA secondary structures. On a widely adopted, standardized test set comprised of 1,305 molecules, the accuracy of our method exceeds that of current state-of-the-art (SOTA) secondary structure prediction software, achieving a Matthews Correlation Coefficient (MCC) over 11-40% higher than that of other leading methods on overall structures and 58-400% higher on pseudoknots specifically.