{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,11]],"date-time":"2026-05-11T18:12:06Z","timestamp":1778523126522,"version":"3.51.4"},"reference-count":37,"publisher":"MDPI AG","issue":"5","license":[{"start":{"date-parts":[[2026,5,11]],"date-time":"2026-05-11T00:00:00Z","timestamp":1778457600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Algorithms"],"abstract":"<jats:p>The prediction accuracy of RNA\u2019s tertiary structure remains a core challenge in the field of computational biology. Existing models frequently encounter significant challenges due to the complexities of diverse topologies and the intricate nature of long-range interactions. We introduce RNAFoldDiff, a generative framework that integrates a sequence-aware graph transformer with a geometric diffusion process for end-to-end RNA 3D structure prediction. RNA sequences and secondary structures are converted into graph representations that capture backbone connectivity and base pair topology. The transformer models local motifs and global dependencies, while the diffusion module iteratively denoises coordinates into physically consistent conformations. The model was pretrained on more than 15,000 structural motifs from the RNA 3D Hub and fine-tuned on complete RNAs from the RNA-Puzzles dataset. In benchmarking tests, RNAFold-Diff achieved an average root mean square deviation (RMSD) of 2.64 \u00c5, a Global Distance Test (GDT) score of 68.7%, and a base pair accuracy of 89.5%, reducing RMSD by nearly 30% and improving GDT by 9 points compared to RoseTTAFoldNA. The framework also outperformed FARFAR2, SimRNA, and RNAformer. Ablation experiments confirmed the contributions of diffusion refinement, edge-aware graph encoding, and motif-level pretraining, while qualitative analyses showed biologically plausible folds including helices, junctions, and multiloops. By combining topology-aware graph learning with generative diffusion, RNAFoldDiff advances RNA tertiary structure modeling and provides a practical tool for RNA design, ribozyme analysis, and structure-guided drug discovery.<\/jats:p>","DOI":"10.3390\/a19050381","type":"journal-article","created":{"date-parts":[[2026,5,11]],"date-time":"2026-05-11T17:17:42Z","timestamp":1778519862000},"page":"381","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["RNAFoldDiff-Based Sequence-Aware Graph Diffusion for Accurate RNA 3D Structure Prediction"],"prefix":"10.3390","volume":"19","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-1557-3982","authenticated-orcid":false,"given":"Abdullah","family":"Al-Refai","sequence":"first","affiliation":[{"name":"Software Engineering Department, King Hussein School of Computing Sciences, Princess Sumaya University for Technology, Amman P.O. Box 1438, Jordan"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6884-3715","authenticated-orcid":false,"given":"Mohammad F.","family":"Al-Hammouri","sequence":"additional","affiliation":[{"name":"Department of Computer Engineering, Faculty of Engineering, The Hashemite University, Zarqa 13133, Jordan"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9111-0990","authenticated-orcid":false,"given":"Bandi","family":"Vamsi","sequence":"additional","affiliation":[{"name":"Department of Artificial Intelligence, Madanapalle Institute of Technology and Science, Deemed to Be University, Madanapalle 517326, Andhra Pradesh, India"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9530-7950","authenticated-orcid":false,"given":"Ali","family":"Al Bataineh","sequence":"additional","affiliation":[{"name":"Artificial Intelligence Center, Norwich University, Northfield, VT 05663, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1968","published-online":{"date-parts":[[2026,5,11]]},"reference":[{"key":"ref_1","unstructured":"Yang, S., Pham, N.T., Li, Z., Baik, J.Y., Lee, J., Zhai, T., Yu, W., Hou, B., Shang, T., and He, W. (2025). Advances in RNA secondary structure prediction and RNA modifications: Methods, data, and applications. arXiv."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"bpae097","DOI":"10.1093\/biomethods\/bpae097","article-title":"Robust RNA secondary structure prediction with a mixture of deep learning and physics-based experts","volume":"10","author":"Qiu","year":"2025","journal-title":"Biol. Methods Protoc."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"109845","DOI":"10.1016\/j.compbiomed.2025.109845","article-title":"RNA structure prediction using deep learning\u2014A comprehensive review","volume":"188","author":"Chaturvedi","year":"2025","journal-title":"Comput. Biol. Med."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"102985","DOI":"10.1016\/j.sbi.2025.102985","article-title":"Advancing protein structure prediction beyond AlphaFold2","volume":"90","author":"Park","year":"2025","journal-title":"Curr. Opin. Struct. Biol."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"1089","DOI":"10.1038\/s41588-025-02157-7","article-title":"Circular RNA discovery with emerging sequencing and deep learning technologies","volume":"57","author":"Zhang","year":"2025","journal-title":"Nat. Genet."},{"key":"ref_6","unstructured":"Gong, Z., Jiang, Z., Gao, W., Zhuo, D., and Ma, L. (2025). A New Deep-learning-Based Approach For mRNA Optimization: High Fidelity, Computation Efficiency, and Multiple Optimization Factors. arXiv."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"lqae019","DOI":"10.1093\/nargab\/lqae019","article-title":"Unraveling the complex relationship between mRNA and protein abundances: A machine learning-based approach for imputing protein levels from RNA-seq data","volume":"6","author":"Prabahar","year":"2024","journal-title":"NAR Genom. Bioinform."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"2647","DOI":"10.1016\/j.bpj.2024.01.026","article-title":"Machine learning in RNA structure prediction: Advances and challenges","volume":"123","author":"Zhang","year":"2024","journal-title":"Biophys. J."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"2375","DOI":"10.1007\/s11030-023-10771-y","article-title":"Deep learning algorithms applied to computational chemistry","volume":"28","year":"2024","journal-title":"Mol. Divers."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"3381","DOI":"10.1016\/j.bpj.2022.08.017","article-title":"FebRNA: An automated fragment-ensemble-based model for building RNA 3D structures","volume":"121","author":"Zhou","year":"2022","journal-title":"Biophys. J."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"963","DOI":"10.1016\/j.str.2020.05.011","article-title":"FARFAR2: Improved de novo rosetta prediction of complex global RNA folds","volume":"28","author":"Watkins","year":"2020","journal-title":"Structure"},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Yuan, L., Guo, Z.H., Cao, W.J., Luo, Y., and Shi, Y.Z. (2021). An integrated tool for RNA 3D structure prediction and analysis. Proceedings of the 2021 33rd Chinese Control and Decision Conference (CCDC), IEEE.","DOI":"10.1109\/CCDC52312.2021.9602110"},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"D301","DOI":"10.1093\/nar\/gkl971","article-title":"The worldwide Protein Data Bank (wwPDB): Ensuring a single, uniform archive of PDB data","volume":"35","author":"Berman","year":"2007","journal-title":"Nucleic Acids Res."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"102991","DOI":"10.1016\/j.sbi.2025.102991","article-title":"Deep learning for RNA structure prediction","volume":"91","author":"Wang","year":"2025","journal-title":"Curr. Opin. Struct. Biol."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"142","DOI":"10.1016\/j.bpj.2021.11.016","article-title":"rsRNASP: A residue-separation-based statistical potential for RNA 3D structure evaluation","volume":"121","author":"Tan","year":"2022","journal-title":"Biophys. J."},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"168552","DOI":"10.1016\/j.jmb.2024.168552","article-title":"RNA3DB: A structurally-dissimilar dataset split for training and benchmarking deep learning models for RNA structure prediction","volume":"436","author":"Szikszai","year":"2024","journal-title":"J. Mol. Biol."},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"127853","DOI":"10.1016\/j.eswa.2025.127853","article-title":"Enhancing protein structure predictions: DeepSHAP as a tool for understanding AlphaFold2","volume":"286","author":"Sibli","year":"2025","journal-title":"Expert Syst. Appl."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"1449","DOI":"10.1016\/j.csbj.2025.04.001","article-title":"RNA secondary structure prediction by conducting multi-class classifications","volume":"27","author":"Yang","year":"2025","journal-title":"Comput. Struct. Biotechnol. J."},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"1087","DOI":"10.1016\/j.cels.2023.11.006","article-title":"Deep learning and CRISPR-Cas13d ortholog discovery for optimized RNA targeting","volume":"14","author":"Wei","year":"2023","journal-title":"Cell Syst."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"100621","DOI":"10.1016\/j.crmeth.2023.100621","article-title":"Molecular geometric deep learning","volume":"3","author":"Shen","year":"2023","journal-title":"Cell Rep. Methods"},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"7266","DOI":"10.1038\/s41467-023-42528-4","article-title":"trRosettaRNA: Automated prediction of RNA 3D structure with transformer network","volume":"14","author":"Wang","year":"2023","journal-title":"Nat. Commun."},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"493","DOI":"10.1038\/s41586-024-07487-w","article-title":"Accurate structure prediction of biomolecular interactions with AlphaFold 3","volume":"630","author":"Abramson","year":"2024","journal-title":"Nature"},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"101229","DOI":"10.1016\/j.cels.2025.101229","article-title":"Geometric deep learning and multiple-instance learning for 3D cell-shape profiling","volume":"16","author":"Dent","year":"2025","journal-title":"Cell Syst."},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"104940","DOI":"10.1016\/j.rineng.2025.104940","article-title":"A Deep Learning based Multiple RNA Methylation Sites Prediction Across Species","volume":"26","author":"Shah","year":"2025","journal-title":"Results Eng."},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"101150","DOI":"10.1016\/j.patter.2024.101150","article-title":"A deep learning model for characterizing protein-RNA interactions from sequences at single-base resolution","volume":"6","author":"Shen","year":"2025","journal-title":"Patterns"},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"100053","DOI":"10.1016\/j.aichem.2024.100053","article-title":"Advances in machine-learning approaches to RNA-targeted drug design","volume":"2","author":"Zhou","year":"2024","journal-title":"Artif. Intell. Chem."},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"402","DOI":"10.1016\/j.slasd.2023.10.002","article-title":"Graph neural networks for the identification of novel inhibitors of a small RNA","volume":"28","author":"Haga","year":"2023","journal-title":"SLAS Discov."},{"key":"ref_28","first-page":"1","article-title":"DeepFoldRNA: A Graph Neural Network for RNA 3D Structure Prediction","volume":"1","author":"Thill","year":"2025","journal-title":"Bioinform. Code"},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"102548","DOI":"10.1016\/j.sbi.2023.102548","article-title":"Structure-based drug design with geometric deep learning","volume":"79","author":"Isert","year":"2023","journal-title":"Curr. Opin. Struct. Biol."},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"102566","DOI":"10.1016\/j.sbi.2023.102566","article-title":"Deep generative models for 3D molecular structure","volume":"80","author":"Baillif","year":"2023","journal-title":"Curr. Opin. Struct. Biol."},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"108865","DOI":"10.1016\/j.compbiomed.2024.108865","article-title":"PIDiff: Physics informed diffusion model for protein pocket-specific 3D molecular generation","volume":"180","author":"Choi","year":"2024","journal-title":"Comput. Biol. Med."},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"101267","DOI":"10.1016\/j.jpha.2025.101267","article-title":"Combining transformer and 3DCNN models to achieve co-design of structures and sequences of antibodies in a diffusional manner","volume":"15","author":"Hu","year":"2025","journal-title":"J. Pharm. Anal."},{"key":"ref_33","doi-asserted-by":"crossref","first-page":"101339","DOI":"10.1016\/j.jpha.2025.101339","article-title":"Equivariant graph neural network-based accurate and ultra-fast virtual screening of small molecules targeting miRNA-protein complex","volume":"16","author":"Wang","year":"2025","journal-title":"J. Pharm. Anal."},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"101257","DOI":"10.1016\/j.jpha.2025.101257","article-title":"3D-EDiffMG: 3D equivariant diffusion-driven molecular generation to accelerate drug discovery","volume":"15","author":"Xu","year":"2025","journal-title":"J. Pharm. Anal."},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"100074","DOI":"10.1016\/j.aichem.2024.100074","article-title":"Conf-GEM: A geometric information-assisted direct conformation generation model","volume":"2","author":"Yang","year":"2024","journal-title":"Artif. Intell. Chem."},{"key":"ref_36","unstructured":"RNA-Puzzles Community (2025, June 16). RNA-Puzzles: A CASP-like Evaluation of RNA 3D Structure Prediction. Available online: http:\/\/www.rnapuzzles.org."},{"key":"ref_37","unstructured":"Leontis, N.B., Zirbel, E.N., and the RNA 3D Hub Team (2025, June 16). RNA 3D Hub: A Database of RNA 3D Motifs and Structures. Bowling Green State University, Available online: https:\/\/rna.bgsu.edu\/rna3dhub."}],"container-title":["Algorithms"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1999-4893\/19\/5\/381\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,11]],"date-time":"2026-05-11T17:32:06Z","timestamp":1778520726000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1999-4893\/19\/5\/381"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5,11]]},"references-count":37,"journal-issue":{"issue":"5","published-online":{"date-parts":[[2026,5]]}},"alternative-id":["a19050381"],"URL":"https:\/\/doi.org\/10.3390\/a19050381","relation":{},"ISSN":["1999-4893"],"issn-type":[{"value":"1999-4893","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5,11]]}}}