Assessment of three-dimensional RNA structure prediction in CASP15

Challenges, advances, and opportunities in RNA structural biology by Cryo-EM

RNAs are remarkably versatile molecules that can fold into intricate three-dimensional (3D) structures to perform diverse cellular and viral functions. Despite their biological importance, relatively few RNA 3D structures have been solved, and our understanding of RNA structure-function relationships remains in its infancy. This limitation partly arises from challenges posed by RNA's complex conformational landscape, characterized by structural flexibility, formation of multiple states, and a propensity to misfold. Recently, cryo-electron microscopy (cryo-EM) has emerged as a powerful tool for the visualization of conformationally dynamic RNA-only 3D structures. However, RNA's characteristics continue to pose challenges. We discuss experimental methods developed to overcome these hurdles, including the engineering of modular modifications that facilitate the visualization of small RNAs, improve particle alignment, and validate structural models.

Embracing exascale computing in nucleic acid simulations

This mini-review reports the recent advances in biomolecular simulations, particularly for nucleic acids, and provides the potential effects of the emerging exascale computing on nucleic acid simulations, emphasizing the need for advanced computational strategies to fully exploit this technological frontier. Specifically, we introduce recent breakthroughs in computer architectures for large-scale biomolecular simulations and review the simulation protocols for nucleic acids regarding force fields, enhanced sampling methods, coarse-grained models, and interactions with ligands. We also explore the integration of machine learning methods into simulations, which promises to significantly enhance the predictive modeling of biomolecules and the analysis of complex data generated by the exascale simulations. Finally, we discuss the challenges and perspectives for biomolecular simulations as we enter the dawning exascale computing era.

Comparative analysis of RNA 3D structure prediction methods: towards enhanced modeling of RNA-ligand interactions

Accurate RNA structure models are crucial for designing small molecule ligands that modulate their functions. This study assesses six standalone RNA 3D structure prediction methods-DeepFoldRNA, RhoFold, BRiQ, FARFAR2, SimRNA and Vfold2, excluding web-based tools due to intellectual property concerns. We focus on reproducing the RNA structure existing in RNA-small molecule complexes, particularly on the ability to model ligand binding sites. Using a comprehensive set of RNA structures from the PDB, which includes diverse structural elements, we found that machine learning (ML)-based methods effectively predict global RNA folds but are less accurate with local interactions. Conversely, non-ML-based methods demonstrate higher precision in modeling intramolecular interactions, particularly with secondary structure restraints. Importantly, ligand-binding site accuracy can remain sufficiently high for practical use, even if the overall model quality is not optimal. With the recent release of AlphaFold 3, we included this advanced method in our tests. Benchmark subsets containing new structures, not used in the training of the tested ML methods, show that AlphaFold 3's performance was comparable to other ML-based methods, albeit with some challenges in accurately modeling ligand binding sites. This study underscores the importance of enhancing binding site prediction accuracy and the challenges in modeling RNA-ligand interactions accurately.

FRET-guided modeling of nucleic acids

The functional diversity of RNAs is encoded in their innate conformational heterogeneity. The combination of single-molecule spectroscopy and computational modeling offers new attractive opportunities to map structural transitions within nucleic acid ensembles. Here, we describe a framework to harmonize single-molecule Förster resonance energy transfer (FRET) measurements with molecular dynamics simulations and de novo structure prediction. Using either all-atom or implicit fluorophore modeling, we recreate FRET experiments in silico, visualize the underlying structural dynamics and quantify the reaction coordinates. Using multiple accessible-contact volumes as a post hoc scoring method for fragment assembly in Rosetta, we demonstrate that FRET can be used to filter a de novo RNA structure prediction ensemble by refuting models that are not compatible with in vitro FRET measurement. We benchmark our FRET-assisted modeling approach on double-labeled DNA strands and validate it against an intrinsically dynamic manganese(II)-binding riboswitch. We show that a FRET coordinate describing the assembly of a four-way junction allows our pipeline to recapitulate the global fold of the riboswitch displayed by the crystal structure. We conclude that computational fluorescence spectroscopy facilitates the interpretability of dynamic structural ensembles and improves the mechanistic understanding of nucleic acid interactions.

SimRNAweb v2.0: a web server for RNA folding simulations and 3D structure modeling, with optional restraints and enhanced analysis of folding trajectories

Research on ribonucleic acid (RNA) structures and functions benefits from easy-to-use tools for computational prediction and analyses of RNA three-dimensional (3D) structure. The SimRNAweb server version 2.0 offers an enhanced, user-friendly platform for RNA 3D structure prediction and analysis of RNA folding trajectories based on the SimRNA method. SimRNA employs a coarse-grained model, Monte Carlo sampling and statistical potentials to explore RNA conformational space, optionally guided by spatial restraints. Recognized for its accuracy in RNA 3D structure prediction in RNA-Puzzles and CASP competitions, SimRNA is particularly useful for incorporating restraints based on experimental data. The new server version introduces performance optimizations and extends user control over simulations and the processing of results. It allows the application of various hard and soft restraints, accommodating alternative structures involving canonical and noncanonical base pairs and unpaired residues, while also integrating data from chemical probing methods. Enhanced features include an improved analysis of folding trajectories, offering advanced clustering options and multiple analyses of the generated trajectories. These updates provide comprehensive tools for detailed RNA structure analysis. SimRNAweb v2.0 significantly broadens the scope of RNA modeling, emphasizing flexibility and user-defined parameter control. The web server is available at https://genesilico.pl/SimRNAweb.

The structure assessment web server: for proteins, complexes and more

The 'structure assessment' web server is a one-stop shop for interactive evaluation and benchmarking of structural models of macromolecular complexes including proteins and nucleic acids. A user-friendly web dashboard links sequence with structure information and results from a variety of state-of-the-art tools, which facilitates the visual exploration and evaluation of structure models. The dashboard integrates stereochemistry information, secondary structure information, global and local model quality assessment of the tertiary structure of comparative protein models, as well as prediction of membrane location. In addition, a benchmarking mode is available where a model can be compared to a reference structure, providing easy access to scores that have been used in recent CASP experiments and CAMEO. The structure assessment web server is available at https://swissmodel.expasy.org/assess.

The landscape of RNA 3D structure modeling with transformer networks

Transformers are a powerful subclass of neural networks catalyzing the development of a growing number of computational methods for RNA structure modeling. Here, we conduct an objective and empirical study of the predictive modeling accuracy of the emerging transformer-based methods for RNA structure prediction. Our study reveals multi-faceted complementarity between the methods and underscores some key aspects that affect the prediction accuracy.

Ribonanza: deep learning of RNA structure through dual crowdsourcing

Prediction of RNA structure from sequence remains an unsolved problem, and progress has been slowed by a paucity of experimental data. Here, we present Ribonanza, a dataset of chemical mapping measurements on two million diverse RNA sequences collected through Eterna and other crowdsourced initiatives. Ribonanza measurements enabled solicitation, training, and prospective evaluation of diverse deep neural networks through a Kaggle challenge, followed by distillation into a single, self-contained model called RibonanzaNet. When fine tuned on auxiliary datasets, RibonanzaNet achieves state-of-the-art performance in modeling experimental sequence dropout, RNA hydrolytic degradation, and RNA secondary structure, with implications for modeling RNA tertiary structure.

Knotted artifacts in predicted 3D RNA structures

Unlike proteins, RNAs deposited in the Protein Data Bank do not contain topological knots. Recently, admittedly, the first trefoil knot and some lasso-type conformations have been found in experimental RNA structures, but these are still exceptional cases. Meanwhile, algorithms predicting 3D RNA models have happened to form knotted structures not so rarely. Interestingly, machine learning-based predictors seem to be more prone to generate knotted RNA folds than traditional methods. A similar situation is observed for the entanglements of structural elements. In this paper, we analyze all models submitted to the CASP15 competition in the 3D RNA structure prediction category. We show what types of topological knots and structure element entanglements appear in the submitted models and highlight what methods are behind the generation of such conformations. We also study the structural aspect of susceptibility to entanglement. We suggest that predictors take care of an evaluation of RNA models to avoid publishing structures with artifacts, such as unusual entanglements, that result from hallucinations of predictive algorithms.

Accurate structure prediction of biomolecular interactions with AlphaFold 3

The introduction of AlphaFold 21 has spurred a revolution in modelling the structure of proteins and their interactions, enabling a huge range of applications in protein modelling and design2-6. Here we describe our AlphaFold 3 model with a substantially updated diffusion-based architecture that is capable of predicting the joint structure of complexes including proteins, nucleic acids, small molecules, ions and modified residues. The new AlphaFold model demonstrates substantially improved accuracy over many previous specialized tools: far greater accuracy for protein-ligand interactions compared with state-of-the-art docking tools, much higher accuracy for protein-nucleic acid interactions compared with nucleic-acid-specific predictors and substantially higher antibody-antigen prediction accuracy compared with AlphaFold-Multimer v.2.37,8. Together, these results show that high-accuracy modelling across biomolecular space is possible within a single unified deep-learning framework.

State-of-the-RNArt: benchmarking current methods for RNA 3D structure prediction

RNAs are essential molecules involved in numerous biological functions. Understanding RNA functions requires the knowledge of their 3D structures. Computational methods have been developed for over two decades to predict the 3D conformations from RNA sequences. These computational methods have been widely used and are usually categorised as either ab initio or template-based. The performances remain to be improved. Recently, the rise of deep learning has changed the sight of novel approaches. Deep learning methods are promising, but their adaptation to RNA 3D structure prediction remains difficult. In this paper, we give a brief review of the ab initio, template-based and novel deep learning approaches. We highlight the different available tools and provide a benchmark on nine methods using the RNA-Puzzles dataset. We provide an online dashboard that shows the predictions made by benchmarked methods, freely available on the EvryRNA platform: https://evryrna.ibisc.univ-evry.fr/evryrna/state_of_the_rnart/.

Structure prediction of protein-ligand complexes from sequence information with Umol

Protein-ligand docking is an established tool in drug discovery and development to narrow down potential therapeutics for experimental testing. However, a high-quality protein structure is required and often the protein is treated as fully or partially rigid. Here we develop an AI system that can predict the fully flexible all-atom structure of protein-ligand complexes directly from sequence information. We find that classical docking methods are still superior, but depend upon having crystal structures of the target protein. In addition to predicting flexible all-atom structures, predicted confidence metrics (plDDT) can be used to select accurate predictions as well as to distinguish between strong and weak binders. The advances presented here suggest that the goal of AI-based drug discovery is one step closer, but there is still a way to go to grasp the complexity of protein-ligand interactions fully. Umol is available at: https://github.com/patrickbryant1/Umol .

RNA language models predict mutations that improve RNA function

Structured RNA lies at the heart of many central biological processes, from gene expression to catalysis. While advances in deep learning enable the prediction of accurate protein structural models, RNA structure prediction is not possible at present due to a lack of abundant high-quality reference data. Furthermore, available sequence data are generally not associated with organismal phenotypes that could inform RNA function. We created GARNET (Gtdb Acquired RNa with Environmental Temperatures), a new database for RNA structural and functional analysis anchored to the Genome Taxonomy Database (GTDB). GARNET links RNA sequences derived from GTDB genomes to experimental and predicted optimal growth temperatures of GTDB reference organisms. This enables construction of deep and diverse RNA sequence alignments to be used for machine learning. Using GARNET, we define the minimal requirements for a sequence- and structure-aware RNA generative model. We also develop a GPT-like language model for RNA in which triplet tokenization provides optimal encoding. Leveraging hyperthermophilic RNAs in GARNET and these RNA generative models, we identified mutations in ribosomal RNA that confer increased thermostability to the Escherichia coli ribosome. The GTDB-derived data and deep learning models presented here provide a foundation for understanding the connections between RNA sequence, structure, and function.

Structural highlights of macromolecular complexes and assemblies

The structures of macromolecular assemblies have given us deep insights into cellular processes and have profoundly impacted biological research and drug discovery. We highlight the structures of macromolecular assemblies that have been modeled using integrative and computational methods and describe how open access to these structures from structural archives has empowered the research community. The arsenal of experimental and computational methods for structure determination ensures a future where whole organelles and cells can be modeled.

RNA3DB: A structurally-dissimilar dataset split for training and benchmarking deep learning models for RNA structure prediction

With advances in protein structure prediction thanks to deep learning models like AlphaFold, RNA structure prediction has recently received increased attention from deep learning researchers. RNAs introduce substantial challenges due to the sparser availability and lower structural diversity of the experimentally resolved RNA structures in comparison to protein structures. These challenges are often poorly addressed by the existing literature, many of which report inflated performance due to using training and testing sets with significant structural overlap. Further, the most recent Critical Assessment of Structure Prediction (CASP15) has shown that deep learning models for RNA structure are currently outperformed by traditional methods. In this paper we present RNA3DB, a dataset of structured RNAs, derived from the Protein Data Bank (PDB), that is designed for training and benchmarking deep learning models. The RNA3DB method arranges the RNA 3D chains into distinct groups (Components) that are non-redundant both with regard to sequence as well as structure, providing a robust way of dividing training, validation, and testing sets. Any split of these structurally-dissimilar Components are guaranteed to produce test and validations sets that are distinct by sequence and structure from those in the training set. We provide the RNA3DB dataset, a particular train/test split of the RNA3DB Components (in an approximate 70/30 ratio) that will be updated periodically. We also provide the RNA3DB methodology along with the source-code, with the goal of creating a reproducible and customizable tool for producing structurally-dissimilar dataset splits for structural RNAs.

Multiple sequence alignment-based RNA language model and its application to structural inference

Compared with proteins, DNA and RNA are more difficult languages to interpret because four-letter coded DNA/RNA sequences have less information content than 20-letter coded protein sequences. While BERT (Bidirectional Encoder Representations from Transformers)-like language models have been developed for RNA, they are ineffective at capturing the evolutionary information from homologous sequences because unlike proteins, RNA sequences are less conserved. Here, we have developed an unsupervised multiple sequence alignment-based RNA language model (RNA-MSM) by utilizing homologous sequences from an automatic pipeline, RNAcmap, as it can provide significantly more homologous sequences than manually annotated Rfam. We demonstrate that the resulting unsupervised, two-dimensional attention maps and one-dimensional embeddings from RNA-MSM contain structural information. In fact, they can be directly mapped with high accuracy to 2D base pairing probabilities and 1D solvent accessibilities, respectively. Further fine-tuning led to significantly improved performance on these two downstream tasks compared with existing state-of-the-art techniques including SPOT-RNA2 and RNAsnap2. By comparison, RNA-FM, a BERT-based RNA language model, performs worse than one-hot encoding with its embedding in base pair and solvent-accessible surface area prediction. We anticipate that the pre-trained RNA-MSM model can be fine-tuned on many other tasks related to RNA structure and function.

CASP15 cryo-EM protein and RNA targets: Refinement and analysis using experimental maps

CASP assessments primarily rely on comparing predicted coordinates with experimental reference structures. However, experimental structures by their nature are only models themselves-their construction involves a certain degree of subjectivity in interpreting density maps and translating them to atomic coordinates. Here, we directly utilized density maps to evaluate the predictions by employing a method for ranking the quality of protein chain predictions based on their fit into the experimental density. The fit-based ranking was found to correlate well with the CASP assessment scores. Overall, the evaluation against the density map indicated that the models are of high accuracy, and occasionally even better than the reference structure in some regions of the model. Local assessment of predicted side chains in a 1.52 Å resolution map showed that side-chains are sometimes poorly positioned. Additionally, the top 118 predictions associated with 9 protein target reference structures were selected for automated refinement, in addition to the top 40 predictions for 11 RNA targets. For both proteins and RNA, the refinement of CASP15 predictions resulted in structures that are close to the reference target structure. This refinement was successful despite large conformational changes often being required, showing that predictions from CASP-assessed methods could serve as a good starting point for building atomic models in cryo-EM maps for both proteins and RNA. Loop modeling continued to pose a challenge for predictors, and together with the lack of consensus amongst models in these regions suggests that modeling, in combination with model-fit to the density, holds the potential for identifying more flexible regions within the structure.