3D-based RNA function prediction tools in rnaglib

TL;DR

This paper discusses using rnaglib to develop machine learning models for predicting RNA functions based on 3D structural data, addressing challenges in dataset creation and standardization.

Contribution

It introduces a framework for applying rnaglib to train both supervised and unsupervised RNA function prediction models from 3D structure datasets.

Findings

Effective models can be trained using rnaglib for RNA function prediction.

Addresses standardization issues in RNA 3D structure datasets.

Facilitates the development of computational tools for RNA analysis.

Abstract

Understanding the connection between complex structural features of RNA and biological function is a fundamental challenge in evolutionary studies and in RNA design. However, building datasets of RNA 3D structures and making appropriate modeling choices remains time-consuming and lacks standardization. In this chapter, we describe the use of rnaglib, to train supervised and unsupervised machine learning-based function prediction models on datasets of RNA 3D structures.