Length-Dependent Deep Learning Model for RNA Secondary Structure Prediction

Mao, Kangkun; Wang, Jun; Xiao, Yi

doi:10.3390/molecules27031030

Cited by 15 publications

(5 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Among the methods developed for predicting RNA secondary structure, a minority can accurately predict the secondary structure of RNAs containing both canonical and noncanonical base pairs (e.g., − ). Machine learning and deep learning methods of RNA secondary structure prediction have been proposed (see, for example, references − ), many of which demonstrate proficiency in handling both canonical and noncanonical interactions. ,− While these methods have achieved considerable accuracy, ,,, challenges persist, especially for long RNA sequences (>500 nucleotides). Indeed, the complexity of possible secondary structures in longer sequences makes accurate prediction more challenging, and further improvements are still needed.…”

Section: Computational Methods For Predicting Non-coding Rna–disease ...mentioning

confidence: 99%

Frontiers and Challenges of Computing ncRNAs Biogenesis, Function and Modulation

Rinaldi,

Moroni,

Rozza

et al. 2024

J. Chem. Theory Comput.

View full text Add to dashboard Cite

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, PIWIinteracting 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-theart 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.

show abstract

Section: Computational Methods For Predicting Non-coding Rna–disease ...mentioning

confidence: 99%

Frontiers and Challenges of Computing ncRNAs Biogenesis, Function and Modulation

Rinaldi,

Moroni,

Rozza

et al. 2024

J. Chem. Theory Comput.

View full text Add to dashboard Cite

show abstract

“…Recently, 2dRNA was improved to take the sequence length of RNA as a feature and showed better performance. 79 These ML-based methods all show higher prediction accuracy than traditional methods of RNA secondary structure prediction. However, their prediction accuracy still leaves room for improvement, especially for long sequences, e.g., longer than 500 residues.…”

Section: Recent Advances In Rna 3d Structure Predictionmentioning

confidence: 98%

“…This is good for secondary structure prediction, where the local features add a benefit to the prediction of the local structures like stems or hairpins while the global features about these local structures provide information on the global structure, e.g., for pseudoknots. Recently, 2dRNA was improved to take the sequence length of RNA as a feature and showed better performance …”

Section: Recent Advances In Rna 3d Structure Predictionmentioning

confidence: 99%

Advances in RNA 3D Structure Prediction

Zhang

Xiong

et al. 2022

J. Chem. Inf. Model.

Self Cite

View full text Add to dashboard Cite

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.

show abstract

“…As existing large RNA secondary structure datasets are curated largely via comparative sequence analysis [19,20], this study focuses on the class of single-sequence-based DL models, referred to as de novo DL models. A number of highly successful de novo DL models have been reported, such as 2dRNA [21], ATTfold [22], DMfold [23], E2Efold [24], MXfold2 [25], SPOT-RNA [26], and Ufold [27], among others [28][29][30][31]. These DL models markedly outperform traditional algorithms, with even close-to-perfect predictions in some cases, though questions on the training vs. test similarity have been raised [32,33] and discussed below.…”

Section: Introductionmentioning

confidence: 99%

Sequence similarity governs generalizability of de novo deep learning models for RNA secondary structure prediction

Qiu

2023

PLoS Comput Biol

View full text Add to dashboard Cite

Making no use of physical laws or co-evolutionary information, de novo deep learning (DL) models for RNA secondary structure prediction have achieved far superior performances than traditional algorithms. However, their statistical underpinning raises the crucial question of generalizability. We present a quantitative study of the performance and generalizability of a series of de novo DL models, with a minimal two-module architecture and no post-processing, under varied similarities between seen and unseen sequences. Our models demonstrate excellent expressive capacities and outperform existing methods on common benchmark datasets. However, model generalizability, i.e., the performance gap between the seen and unseen sets, degrades rapidly as the sequence similarity decreases. The same trends are observed from several recent DL and machine learning models. And an inverse correlation between performance and generalizability is revealed collectively across all learning-based models with wide-ranging architectures and sizes. We further quantitate how generalizability depends on sequence and structure identity scores via pairwise alignment, providing unique quantitative insights into the limitations of statistical learning. Generalizability thus poses a major hurdle for deploying de novo DL models in practice and various pathways for future advances are discussed.

show abstract

Length-Dependent Deep Learning Model for RNA Secondary Structure Prediction

Cited by 15 publications

References 42 publications

Frontiers and Challenges of Computing ncRNAs Biogenesis, Function and Modulation

Frontiers and Challenges of Computing ncRNAs Biogenesis, Function and Modulation

Advances in RNA 3D Structure Prediction

Sequence similarity governs generalizability of de novo deep learning models for RNA secondary structure prediction

Contact Info

Product

Resources

About