Abstract
ABSTRACT
Protein-RNA interactions play a central role in many cellular processes, such as gene regulation, protein synthesis and viral infections. Although a few thousand protein-RNA complexes have been structurally characterized, they represent only a small fraction of all interactions. This lack of experimental data limits the ability of deep-learning based approaches such as AlphaFold3 to predict structures of protein-RNA interactions. In this study, we investigate protein-RNA binding specificity by enriching experimental structures with omics data. To this end, we present a scoring approach and associated web resource quantifying the agreement between experimental structures of protein-RNA interfaces and their omics-derived binding preferences, allowing us to identify interaction motif cores. Through key structural and evolutionary features, we further highlight that these motif cores correspond to important interface regions. We then leverage the dataset of protein-RNA complexes for which structural information can be combined with binding preferences in order to benchmark the ability of AlphaFold3 to predict protein-RNA interaction specificity. To this aim, we run AlphaFold3 predictions using different RNA inputs, from the exact sequence present in the experimental structure to a non-specific sequence, including sequences embedding the consensus binding motif from
in vitro
experiments. We show that despite good prediction quality, the sensitivity of AlphaFold3 to the exact RNA sequence used as input indicates signs of memorization. We examine some particular complexes and uncover challenges encountered by our scoring workflow and AlphaFold3, especially in the case of alternative binding modes. Altogether, this work evidences both the promise and current limitations of deep learning approaches for protein-RNA structure prediction, and provides a resource to guide their further development.
AUTHOR SUMMARY
Proteins and RNA interact constantly within cells, driving processes such as gene expression and protein synthesis. Understanding the mechanisms underlying specific protein-RNA recognition matters for both biological knowledge and therapeutic applications, yet remains difficult. Far fewer experimentally resolved structures exist for protein-RNA complexes than for proteins alone, and structure prediction tools based on artificial intelligence, such as AlphaFold3, still struggle with RNA and protein-RNA interactions. In this study, we combine experimental data from different sources to design a scoring method quantifying their agreement and we show our method’s ability to highlight relevant binding regions. These results are available for interactive exploration through a web resource. We also benchmark AlphaFold3 for its ability to predict protein-RNA interactions in a specific manner. This benchmark highlights difficulties of AlphaFold3 in producing specific predictions and signs of memorization. Overall, this work showcases the interest of combining different data modalities and points to possible guidelines for future improvement of specific structural predictions of protein-RNA interactions.