Deep learning models for RNA secondary structure prediction (probably) do not generalize across families

JAX-RNAfold: scalable differentiable folding

SUMMARY

Differentiable folding is an emerging paradigm for RNA design in which a probabilistic sequence representation is optimized via gradient descent. However, given the significant memory overhead of differentiating the expected partition function over all RNA sequences, the existing proof-of-concept algorithm only scales to ≤50 nucleotides. We present JAX-RNAfold, an open-source software package for our drastically improved differentiable folding algorithm that scales to 1,250 nucleotides on a single GPU. Our software permits the natural inclusion of differentiable folding as a module in larger deep learning pipelines, as well as complex RNA design procedures such as mRNA design with flexible objective functions.

AVAILABILITY AND IMPLEMENTATION

JAX-RNAfold is hosted on GitHub (https://github.com/rkruegs123/jax-rnafold) and can be installed locally as a Python package. All source code is also archived on Zenodo (https://doi.org/10.5281/zenodo.15003072).

RNA secondary structure prediction by conducting multi-class classifications

Generating valid predictions of RNA secondary structures is challenging. Several deep learning methods have been developed for predicting RNA secondary structures. However, they commonly adopt post-processing steps to adjust the model output to produce valid predictions, which are complicated and could limit the performance. In this study, we propose a simple method by considering RNA secondary structure prediction as multiple multi-class classifications, which eliminates the need for those complicated post-processing steps. Then, we use this method to train and evaluate our model based on the attention mechanism and the convolutional neural network. Besides, we introduce two additional methods, including data augmentation to further improve the within-RNA-family performance and a method to alleviate the performance drop in the cross-RNA-family evaluation. In summary, we could produce valid predictions and achieve better performance without complex post-processing steps, and we show our additional methods are beneficial to the performance in within-RNA-family and cross-RNA-family evaluations.

RNA structure prediction using deep learning - A comprehensive review

In computational biology, accurate RNA structure prediction offers several benefits, including facilitating a better understanding of RNA functions and RNA-based drug design. Implementing deep learning techniques for RNA structure prediction has led tremendous progress in this field, resulting in significant improvements in prediction accuracy. This comprehensive review aims to provide an overview of the diverse strategies employed in predicting RNA secondary structures, emphasizing deep learning methods. The article categorizes the discussion into three main dimensions: feature extraction methods, existing state-of-the-art learning model architectures, and prediction approaches. We present a comparative analysis of various techniques and models highlighting their strengths and weaknesses. Finally, we identify gaps in the literature, discuss current challenges, and suggest future approaches to enhance model performance and applicability in RNA structure prediction tasks. This review provides a deeper insight into the subject and paves the way for further progress in this dynamic intersection of life sciences and artificial intelligence.

Comprehensive benchmarking of large language models for RNA secondary structure prediction

In recent years, inspired by the success of large language models (LLMs) for DNA and proteins, several LLMs for RNA have also been developed. These models take massive RNA datasets as inputs and learn, in a self-supervised way, how to represent each RNA base with a semantically rich numerical vector. This is done under the hypothesis that obtaining high-quality RNA representations can enhance data-costly downstream tasks, such as the fundamental RNA secondary structure prediction problem. However, existing RNA-LLM have not been evaluated for this task in a unified experimental setup. Since they are pretrained models, assessment of their generalization capabilities on new structures is a crucial aspect. Nonetheless, this has been just partially addressed in literature. In this work we present a comprehensive experimental and comparative analysis of pretrained RNA-LLM that have been recently proposed. We evaluate the use of these representations for the secondary structure prediction task with a common deep learning architecture. The RNA-LLM were assessed with increasing generalization difficulty on benchmark datasets. Results showed that two LLMs clearly outperform the other models, and revealed significant challenges for generalization in low-homology scenarios. Moreover, in this study we provide curated benchmark datasets of increasing complexity and a unified experimental setup for this scientific endeavor. Source code and curated benchmark datasets with increasing complexity are available in the repository: https://github.com/sinc-lab/rna-llm-folding/.

Robust RNA secondary structure prediction with a mixture of deep learning and physics-based experts

A mixture-of-experts (MoE) approach has been developed to mitigate the poor out-of-distribution (OOD) generalization of deep learning (DL) models for single-sequence-based prediction of RNA secondary structure. The main idea behind this approach is to use DL models for in-distribution (ID) test sequences to leverage their superior ID performances, while relying on physics-based models for OOD sequences to ensure robust predictions. One key ingredient of the pipeline, named MoEFold2D, is automated ID/OOD detection via consensus analysis of an ensemble of DL model predictions without requiring access to training data during inference. Specifically, motivated by the clustered distribution of known RNA structures, a collection of distinct DL models is trained by iteratively leaving one cluster out. Each DL model hence serves as an expert on all but one cluster in the training data. Consequently, for an ID sequence, all but one DL model makes accurate predictions consistent with one another, while an OOD sequence yields highly inconsistent predictions among all DL models. Through consensus analysis of DL predictions, test sequences are categorized as ID or OOD. ID sequences are subsequently predicted by averaging the DL models in consensus, and OOD sequences are predicted using physics-based models. Instead of remediating generalization gaps with alternative approaches such as transfer learning and sequence alignment, MoEFold2D circumvents unpredictable ID-OOD gaps and combines the strengths of DL and physics-based models to achieve accurate ID and robust OOD predictions.

Identification of RNA structures and their roles in RNA functions

The development of high-throughput RNA structure profiling methods in the past decade has greatly facilitated our ability to map and characterize different aspects of RNA structures transcriptome-wide in cell populations, single cells and single molecules. The resulting high-resolution data have provided insights into the static and dynamic nature of RNA structures, revealing their complexity as they perform their respective functions in the cell. In this Review, we discuss recent technical advances in the determination of RNA structures, and the roles of RNA structures in RNA biogenesis and functions, including in transcription, processing, translation, degradation, localization and RNA structure-dependent condensates. We also discuss the current understanding of how RNA structures could guide drug design for treating genetic diseases and battling pathogenic viruses, and highlight existing challenges and future directions in RNA structure research.

Predicting RNA sequence-structure likelihood via structure-aware deep learning

BACKGROUND

The active functionalities of RNA are recognized to be heavily dependent on the structure and sequence. Therefore, a model that can accurately evaluate a design by giving RNA sequence-structure pairs would be a valuable tool for many researchers. Machine learning methods have been explored to develop such tools, showing promising results. However, two key issues remain. Firstly, the performance of machine learning models is affected by the features used to characterize RNA. Currently, there is no consensus on which features are the most effective for characterizing RNA sequence-structure pairs. Secondly, most existing machine learning methods extract features describing entire RNA molecule. We argue that it is essential to define additional features that characterize nucleotides and specific sections of RNA structure to enhance the overall efficacy of the RNA design process.

RESULTS

We develop two deep learning models for evaluating RNA sequence-secondary structure pairs. The first model, NU-ResNet, uses a convolutional neural network architecture that solves the aforementioned problems by explicitly encoding RNA sequence-structure information into a 3D matrix. Building upon NU-ResNet, our second model, NUMO-ResNet, incorporates additional information derived from the characterizations of RNA, specifically the 2D folding motifs. In this work, we introduce an automated method to extract these motifs based on fundamental secondary structure descriptions. We evaluate the performance of both models on an independent testing dataset. Our proposed models outperform the models from literatures in this independent testing dataset. To assess the robustness of our models, we conduct 10-fold cross validation. To evaluate the generalization ability of NU-ResNet and NUMO-ResNet across different RNA families, we train and test our proposed models in different RNA families. Our proposed models show superior performance compared to the models from literatures when being tested across different independent RNA families.

CONCLUSIONS

In this study, we propose two deep learning models, NU-ResNet and NUMO-ResNet, to evaluate RNA sequence-secondary structure pairs. These two models expand the field of data-driven approaches for learning RNA. Furthermore, these two models provide the new method to encode RNA sequence-secondary structure pairs.

SpanSeq: similarity-based sequence data splitting method for improved development and assessment of deep learning projects

The use of deep learning models in computational biology has increased massively in recent years, and it is expected to continue with the current advances in the fields such as Natural Language Processing. These models, although able to draw complex relations between input and target, are also inclined to learn noisy deviations from the pool of data used during their development. In order to assess their performance on unseen data (their capacity to generalize), it is common to split the available data randomly into development (train/validation) and test sets. This procedure, although standard, has been shown to produce dubious assessments of generalization due to the existing similarity between samples in the databases used. In this work, we present SpanSeq, a database partition method for machine learning that can scale to most biological sequences (genes, proteins and genomes) in order to avoid data leakage between sets. We also explore the effect of not restraining similarity between sets by reproducing the development of two state-of-the-art models on bioinformatics, not only confirming the consequences of randomly splitting databases on the model assessment, but expanding those repercussions to the model development. SpanSeq is available at https://github.com/genomicepidemiology/SpanSeq.

sincFold: end-to-end learning of short- and long-range interactions in RNA secondary structure

MOTIVATION

Coding and noncoding RNA molecules participate in many important biological processes. Noncoding RNAs fold into well-defined secondary structures to exert their functions. However, the computational prediction of the secondary structure from a raw RNA sequence is a long-standing unsolved problem, which after decades of almost unchanged performance has now re-emerged due to deep learning. Traditional RNA secondary structure prediction algorithms have been mostly based on thermodynamic models and dynamic programming for free energy minimization. More recently deep learning methods have shown competitive performance compared with the classical ones, but there is still a wide margin for improvement.

RESULTS

In this work we present sincFold, an end-to-end deep learning approach, that predicts the nucleotides contact matrix using only the RNA sequence as input. The model is based on 1D and 2D residual neural networks that can learn short- and long-range interaction patterns. We show that structures can be accurately predicted with minimal physical assumptions. Extensive experiments were conducted on several benchmark datasets, considering sequence homology and cross-family validation. sincFold was compared with classical methods and recent deep learning models, showing that it can outperform the state-of-the-art methods.

DEBFold: Computational Identification of RNA Secondary Structures for Sequences across Structural Families Using Deep Learning

It is now known that RNAs play more active roles in cellular pathways beyond simply serving as transcription templates. These biological mechanisms might be mediated by higher RNA stereo conformations, triggering the need to understand RNA secondary structures first. However, experimental protocols for solving RNA structures are unavailable for large-scale investigation due to their high costs and time-consuming nature. Various computational tools were thus developed to predict the RNA secondary structures from sequences. Recently, deep networks have been investigated to help predict RNA structures directly from their sequences. However, existing deep-learning-based tools are more or less suffering from model overfitting due to their complicated problem formulation and defective model training processes, limiting their applications across sequences from different structural families. In this research, we designed a two-stage RNA structure prediction strategy called DEBFold (deep ensemble boosting and folding) based on convolution encoding/decoding and self-attention mechanisms to enhance the existing thermodynamic structure models. Moreover, the model training process followed rigorous steps to achieve an acceptable prediction generalization. On the family-wise reserved test sets and the PDB-derived test set, DEBFold achieves better structure prediction performance over traditional tools and existing deep-learning methods. In summary, we obtained a cutting-edge deep-learning-based structure prediction tool with supreme across-family generalization performance. The DEBFold tool can be accessed at https://cobis.bme.ncku.edu.tw/DEBFold/.

MARS and RNAcmap3: The Master Database of All Possible RNA Sequences Integrated with RNAcmap for RNA Homology Search

Recent success of AlphaFold2 in protein structure prediction relied heavily on co-evolutionary information derived from homologous protein sequences found in the huge, integrated database of protein sequences (Big Fantastic Database). In contrast, the existing nucleotide databases were not consolidated to facilitate wider and deeper homology search. Here, we built a comprehensive database by incorporating the non-coding RNA (ncRNA) sequences from RNAcentral, the transcriptome assembly and metagenome assembly from metagenomics RAST (MG-RAST), the genomic sequences from Genome Warehouse (GWH), and the genomic sequences from MGnify, in addition to the nucleotide (nt) database and its subsets in National Center of Biotechnology Information (NCBI). The resulting Master database of All possible RNA sequences (MARS) is 20-fold larger than NCBI's nt database or 60-fold larger than RNAcentral. The new dataset along with a new split-search strategy allows a substantial improvement in homology search over existing state-of-the-art techniques. It also yields more accurate and more sensitive multiple sequence alignments (MSAs) than manually curated MSAs from Rfam for the majority of structured RNAs mapped to Rfam. The results indicate that MARS coupled with the fully automatic homology search tool RNAcmap will be useful for improved structural and functional inference of ncRNAs and RNA language models based on MSAs. MARS is accessible at https://ngdc.cncb.ac.cn/omix/release/OMIX003037, and RNAcmap3 is accessible at http://zhouyq-lab.szbl.ac.cn/download/.

Challenges and best practices in omics benchmarking

Technological advances enabling massively parallel measurement of biological features - such as microarrays, high-throughput sequencing and mass spectrometry - have ushered in the omics era, now in its third decade. The resulting complex landscape of analytical methods has naturally fostered the growth of an omics benchmarking industry. Benchmarking refers to the process of objectively comparing and evaluating the performance of different computational or analytical techniques when processing and analysing large-scale biological data sets, such as transcriptomics, proteomics and metabolomics. With thousands of omics benchmarking studies published over the past 25 years, the field has matured to the point where the foundations of benchmarking have been established and well described. However, generating meaningful benchmarking data and properly evaluating performance in this complex domain remains challenging. In this Review, we highlight some common oversights and pitfalls in omics benchmarking. We also establish a methodology to bring the issues that can be addressed into focus and to be transparent about those that cannot: this takes the form of a spreadsheet template of guidelines for comprehensive reporting, intended to accompany publications. In addition, a survey of recent developments in benchmarking is provided as well as specific guidance for commonly encountered difficulties.

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.

Accurate prediction of RNA secondary structure including pseudoknots through solving minimum-cost flow with learned potentials

Pseudoknots are key structure motifs of RNA and pseudoknotted RNAs play important roles in a variety of biological processes. Here, we present KnotFold, an accurate approach to the prediction of RNA secondary structure including pseudoknots. The key elements of KnotFold include a learned potential function and a minimum-cost flow algorithm to find the secondary structure with the lowest potential. KnotFold learns the potential from the RNAs with known structures using an attention-based neural network, thus avoiding the inaccuracy of hand-crafted energy functions. The specially designed minimum-cost flow algorithm used by KnotFold considers all possible combinations of base pairs and selects from them the optimal combination. The algorithm breaks the restriction of nested base pairs required by the widely used dynamic programming algorithms, thus enabling the identification of pseudoknots. Using 1,009 pseudoknotted RNAs as representatives, we demonstrate the successful application of KnotFold in predicting RNA secondary structures including pseudoknots with accuracy higher than the state-of-the-art approaches. We anticipate that KnotFold, with its superior accuracy, will greatly facilitate the understanding of RNA structures and functionalities.

Frontiers and Challenges of Computing ncRNAs Biogenesis, Function and Modulation

Non-coding RNAs (ncRNAs), generated from nonprotein coding DNA sequences, constitute 98-99% of the human genome. Non-coding RNAs encompass diverse functional classes, including microRNAs, small interfering RNAs, PIWI-interacting RNAs, small nuclear RNAs, small nucleolar RNAs, and long non-coding RNAs. With critical involvement in gene expression and regulation across various biological and physiopathological contexts, such as neuronal disorders, immune responses, cardiovascular diseases, and cancer, non-coding RNAs are emerging as disease biomarkers and therapeutic targets. In this review, after providing an overview of non-coding RNAs' role in cell homeostasis, we illustrate the potential and the challenges of state-of-the-art computational methods exploited to study non-coding RNAs biogenesis, function, and modulation. This can be done by directly targeting them with small molecules or by altering their expression by targeting the cellular engines underlying their biosynthesis. Drawing from applications, also taken from our work, we showcase the significance and role of computer simulations in uncovering fundamental facets of ncRNA mechanisms and modulation. This information may set the basis to advance gene modulation tools and therapeutic strategies to address unmet medical needs.

RNA structure: implications in viral infections and neurodegenerative diseases

RNA is an intermediary between DNA and protein, a catalyzer of biochemical reactions, and a regulator of genes and transcripts. RNA structures are essential for complicated functions. Recent years have witnessed rapid advancements in RNA secondary structure probing techniques. These technological strides provided comprehensive insights into RNA structures, which significantly contributed to our understanding of diverse cellular regulatory processes, including gene regulation, epigenetic regulation, and post-transactional regulation. Meanwhile, they have facilitated the creation of therapeutic tools for tackling human diseases. Despite their therapeutic applications, RNA structure probing methods also offer a promising avenue for exploring the mechanisms of human diseases, potentially providing the key to overcoming existing research constraints and obtaining the in-depth information necessary for a deeper understanding of disease mechanisms.

Concurrent prediction of RNA secondary structures with pseudoknots and local 3D motifs in an integer programming framework

MOTIVATION

The prediction of RNA structure canonical base pairs from a single sequence, especially pseudoknotted ones, remains challenging in a thermodynamic models that approximates the energy of the local 3D motifs joining canonical stems. It has become more and more apparent in recent years that the structural motifs in the loops, composed of noncanonical interactions, are essential for the final shape of the molecule enabling its multiple functions. Our capacity to predict accurate 3D structures is also limited when it comes to the organization of the large intricate network of interactions that form inside those loops.

RESULTS

We previously developed the integer programming framework RNA Motifs over Integer Programming (RNAMoIP) to reconcile RNA secondary structure and local 3D motif information available in databases. We further develop our model to now simultaneously predict the canonical base pairs (with pseudoknots) from base pair probability matrices with or without alignment. We benchmarked our new method over the all nonredundant RNAs below 150 nucleotides. We show that the joined prediction of canonical base pairs structure and local conserved motifs (i) improves the ratio of well-predicted interactions in the secondary structure, (ii) predicts well canonical and Wobble pairs at the location where motifs are inserted, (iii) is greatly improved with evolutionary information, and (iv) noncanonical motifs at kink-turn locations.

AVAILABILITY AND IMPLEMENTATION

The source code of the framework is available at https://gitlab.info.uqam.ca/cbe/RNAMoIP and an interactive web server at https://rnamoip.cbe.uqam.ca/.

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.

Tailor made: the art of therapeutic mRNA design

mRNA medicine is a new and rapidly developing field in which the delivery of genetic information in the form of mRNA is used to direct therapeutic protein production in humans. This approach, which allows for the quick and efficient identification and optimization of drug candidates for both large populations and individual patients, has the potential to revolutionize the way we prevent and treat disease. A key feature of mRNA medicines is their high degree of designability, although the design choices involved are complex. Maximizing the production of therapeutic proteins from mRNA medicines requires a thorough understanding of how nucleotide sequence, nucleotide modification and RNA structure interplay to affect translational efficiency and mRNA stability. In this Review, we describe the principles that underlie the physical stability and biological activity of mRNA and emphasize their relevance to the myriad considerations that factor into therapeutic mRNA design.

AliNA - a deep learning program for RNA secondary structure prediction

Nowadays there are numerous discovered natural RNA variations participating in different cellular processes and artificial RNA, e. g., aptamers, riboswitches. One of the required tasks in the investigation of their functions and mechanism of influence on cells and interaction with targets is the prediction of RNA secondary structures. The classic thermodynamic-based prediction algorithms do not consider the specificity of biological folding and deep learning methods that were designed to resolve this issue suffer from homology-based methods problems. Herein, we present a method for RNA secondary structure prediction based on deep learning - AliNA (ALIgned Nucleic Acids). Our method successfully predicts secondary structures for non-homologous to train-data RNA families thanks to usage of the data augmentation techniques. Augmentation extends existing datasets with easily-accessible simulated data. The proposed method shows a high quality of prediction across different benchmarks including pseudoknots. The method is available on GitHub for free (https://github.com/Arty40m/AliNA).

Automatic recognition of complementary strands: lessons regarding machine learning abilities in RNA folding

Introduction: Prediction of RNA secondary structure from single sequences still needs substantial improvements. The application of machine learning (ML) to this problem has become increasingly popular. However, ML algorithms are prone to overfitting, limiting the ability to learn more about the inherent mechanisms governing RNA folding. It is natural to use high-capacity models when solving such a difficult task, but poor generalization is expected when too few examples are available. Methods: Here, we report the relation between capacity and performance on a fundamental related problem: determining whether two sequences are fully complementary. Our analysis focused on the impact of model architecture and capacity as well as dataset size and nature on classification accuracy. Results: We observed that low-capacity models are better suited for learning with mislabelled training examples, while large capacities improve the ability to generalize to structurally dissimilar data. It turns out that neural networks struggle to grasp the fundamental concept of base complementarity, especially in lengthwise extrapolation context. Discussion: Given a more complex task like RNA folding, it comes as no surprise that the scarcity of useable examples hurdles the applicability of machine learning techniques to this field.

Machine learning modeling of RNA structures: methods, challenges and future perspectives

The three-dimensional structure of RNA molecules plays a critical role in a wide range of cellular processes encompassing functions from riboswitches to epigenetic regulation. These RNA structures are incredibly dynamic and can indeed be described aptly as an ensemble of structures that shifts in distribution depending on different cellular conditions. Thus, the computational prediction of RNA structure poses a unique challenge, even as computational protein folding has seen great advances. In this review, we focus on a variety of machine learning-based methods that have been developed to predict RNA molecules' secondary structure, as well as more complex tertiary structures. We survey commonly used modeling strategies, and how many are inspired by or incorporate thermodynamic principles. We discuss the shortcomings that various design decisions entail and propose future directions that could build off these methods to yield more robust, accurate RNA structure predictions.

Recent trends in RNA informatics: a review of machine learning and deep learning for RNA secondary structure prediction and RNA drug discovery

Computational analysis of RNA sequences constitutes a crucial step in the field of RNA biology. As in other domains of the life sciences, the incorporation of artificial intelligence and machine learning techniques into RNA sequence analysis has gained significant traction in recent years. Historically, thermodynamics-based methods were widely employed for the prediction of RNA secondary structures; however, machine learning-based approaches have demonstrated remarkable advancements in recent years, enabling more accurate predictions. Consequently, the precision of sequence analysis pertaining to RNA secondary structures, such as RNA-protein interactions, has also been enhanced, making a substantial contribution to the field of RNA biology. Additionally, artificial intelligence and machine learning are also introducing technical innovations in the analysis of RNA-small molecule interactions for RNA-targeted drug discovery and in the design of RNA aptamers, where RNA serves as its own ligand. This review will highlight recent trends in the prediction of RNA secondary structure, RNA aptamers and RNA drug discovery using machine learning, deep learning and related technologies, and will also discuss potential future avenues in the field of RNA informatics.

Comparative mitochondrial genomics of American nectar-feeding long-nosed bats Leptonycteris spp. with insights into the phylogeny of the family Phyllostomidae

Among leaf-nosed bats (family Phyllostomidae), the genus Leptonycteris (subfamily Glossophaginae), contains three migratory and obligate nectar-dwelling species of great ecological and economic importance; the greater long-nosed bat L. nivalis, the lesser long-nosed bat L. yerbabuenae, and the southern long-nosed bat L. curasoae. According to the IUCN, the three species are categorized as 'vulnerable', 'endangered', and 'near threatened', respectively. In this study, we assembled and characterized in detail the mitochondrial genome of Leptonycteris spp. and examined the phylogenetic position of this genus in the family Phyllostomidae based on protein coding genes (PCGs). The mitogenomes of L. nivalis, L. curasoae, and L. yerbabuenae are 16,708, 16,758, and 16,729 bp in length and each encode 13 PCGs, 22 transfer RNA genes, 2 rRNA genes, and a putative control region (CR). Mitochondrial gene order is identical to that reported before for the family Phyllostomidae. All tRNAs exhibit a 'cloverleaf' secondary structure, except tRNA-Serine-1 that is missing the DHU arm in the three species. All PCGs are exposed to purifying selection with atp8 experiencing the most relaxed purifying selection as the ω ratio was higher for this gene compared to the other PCGs in each species. The CR of each species contains three functional domains: extended termination associated sequence (ETAS), Central, and conserved sequence block domain (CSB). A phylomitogenomic analysis revealed that Leptonycteris is monophyletic and most closely related to the genus Glossophaga. The analysis also supported the monophyly of the family Glossophaginae in the speciose family Phyllostomidae. The mitochondria characterization of these species provides relevant information to develop molecular markers for conservation purposes.

Machine learning for RNA 2D structure prediction benchmarked on experimental data

Since the 1980s, dozens of computational methods have addressed the problem of predicting RNA secondary structure. Among them are those that follow standard optimization approaches and, more recently, machine learning (ML) algorithms. The former were repeatedly benchmarked on various datasets. The latter, on the other hand, have not yet undergone extensive analysis that could suggest to the user which algorithm best fits the problem to be solved. In this review, we compare 15 methods that predict the secondary structure of RNA, of which 6 are based on deep learning (DL), 3 on shallow learning (SL) and 6 control methods on non-ML approaches. We discuss the ML strategies implemented and perform three experiments in which we evaluate the prediction of (I) representatives of the RNA equivalence classes, (II) selected Rfam sequences and (III) RNAs from new Rfam families. We show that DL-based algorithms (such as SPOT-RNA and UFold) can outperform SL and traditional methods if the data distribution is similar in the training and testing set. However, when predicting 2D structures for new RNA families, the advantage of DL is no longer clear, and its performance is inferior or equal to that of SL and non-ML methods.

How does precursor RNA structure influence RNA processing and gene expression?

RNA is a fundamental biomolecule that has many purposes within cells. Due to its single-stranded and flexible nature, RNA naturally folds into complex and dynamic structures. Recent technological and computational advances have produced an explosion of RNA structural data. Many RNA structures have regulatory and functional properties. Studying the structure of nascent RNAs is particularly challenging due to their low abundance and long length, but their structures are important because they can influence RNA processing. Precursor RNA processing is a nexus of pathways that determines mature isoform composition and that controls gene expression. In this review, we examine what is known about human nascent RNA structure and the influence of RNA structure on processing of precursor RNAs. These known structures provide examples of how other nascent RNAs may be structured and show how novel RNA structures may influence RNA processing including splicing and polyadenylation. RNA structures can be targeted therapeutically to treat disease.

GERONIMO: A tool for systematic retrieval of structural RNAs in a broad evolutionary context

BACKGROUND

While web-based tools such as BLAST have made identifying conserved gene homologs appear easy, genes with variable sequences pose significant challenges. Functionally important noncoding RNAs (ncRNA) often show low sequence conservation due to genetic variations, including insertions and deletions. Rather than conserved sequences, these RNAs possess highly conserved structural features across a broad phylogenetic range. Such features can be identified using the covariance models approach, which combines sequence alignment with a secondary RNA structure consensus. However, running standard implementation of that approach (Infernal) requires advanced bioinformatics knowledge compared to user-friendly web services like BLAST. The issue is partially addressed by RNAcentral, which can be used to search for homologs across a broad range of ncRNA sequence collections from diverse organisms but not across the genome assemblies.

RESULTS

Here, we present GERONIMO, which conducts evolutionary searches across hundreds of genomes in a fully automated way. It provides results extended with taxonomy context, as summary tables and visualizations, to facilitate analysis for user convenience. Additionally, GERONIMO supplements homologous sequences with genomic regions to analyze promoter motifs or gene collinearity, enhancing the validation of results.

CONCLUSION

GERONIMO, built using Snakemake, has undergone extensive testing on hundreds of genomes, establishing itself as a valuable tool in the identification of ncRNA homologs across diverse taxonomic groups. Consequently, GERONIMO facilitates the investigation of the evolutionary patterns of functionally significant ncRNA players, whose understanding has previously been limited to individual organisms and close relatives.

Advances in RNA 3D Structure Prediction

RNA molecules carry out various cellular functions, and understanding the mechanisms behind their functions requires the knowledge of their 3D structures. Different types of computational methods have been developed to model RNA 3D structures over the past decade. These methods were widely used by researchers although their performance needs to be further improved. Recently, along with these traditional methods, machine-learning techniques have been increasingly applied to RNA 3D structure prediction and show significant improvement in performance. Here we shall give a brief review of the traditional methods and recent related advances in machine-learning approaches for RNA 3D structure prediction.