RNA-Protein Interaction Classification via Sequence Embeddings

Dominika Matus, Frederic Runge, Jörg K.H. Franke, Lars Gerne, Michael Uhl, Frank Hutter, Rolf Backofen

Track: Machine learning: computational method and/or computational results

Keywords: RNA-protein Interactions, Deep Learning, Sequence Embeddings, Interaction Classification, RPI Classification

TL;DR: RNA-Protein Interaction Classification via Sequence Embeddings

Abstract:

RNA-protein interactions (RPI) are ubiquitous in cellular organisms and essential for gene regulation.
In particular, protein interactions with non-coding RNAs (ncRNAs) play a critical role in these processes. Experimental analysis of RPIs is time-consuming and expensive, and existing computational methods rely on small and limited datasets. This work introduces $\textit{RNAInterAct}$, a comprehensive RPI dataset, alongside $\textit{RPIembeddor}$, a novel transformer-based model designed for classifying ncRNA-protein interactions. By leveraging two foundation models for sequence embedding, we incorporate essential structural and functional insights into our task. We demonstrate RPIembeddor's strong performance and generalization capability compared to state-of-the-art methods across different datasets and analyze the impact of the proposed embedding strategy on the performance in an ablation study.

Submission Number: 32