RNA structure prediction using deep learning — A comprehensive review

Computers in Biology and Medicine, Journal Year: 2025, Volume and Issue: 188, P. 109845 - 109845

Published: Feb. 20, 2025

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

Language: Английский

---

Advances and Mechanisms of RNA–Ligand Interaction Predictions

Life, Journal Year: 2025, Volume and Issue: 15(1), P. 104 - 104

Published: Jan. 15, 2025

The diversity and complexity of RNA include sequence, secondary structure, tertiary structure characteristics. These elements are crucial for RNA's specific recognition other molecules. With advancements in biotechnology, RNA-ligand structures allow researchers to utilize experimental data uncover the mechanisms complex interactions. However, determining these complexes experimentally can be technically challenging often results low-resolution data. Many machine learning computational approaches have recently emerged learn multiscale-level features predict Predicting interactions remains an unexplored area. Therefore, studying is essential understanding biological processes. In this review, we analyze interaction characteristics by examining structure. Our goal clarify how specifically recognizes ligands. Additionally, systematically discuss methods predicting guide future research directions. We aim inspire creation more reliable prediction tools.

Language: Английский

---

Designing small molecules that target a cryptic RNA binding site via base displacement

Research Square (Research Square), Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 29, 2025

Most RNA-binding small molecules have limited solubility, weak affinity, and/or lack of specificity, restricting the medicinal chemistry often required for lead compound discovery. We reasoned that conjugation these unfavorable ligands to a suitable “host” molecule can solubilize “guest” and deliver it site-specifically an RNA interest resolve issues. Using this framework, we designed library was hosted by cobalamin (Cbl) interact with Cbl riboswitch through common base displacement mechanism. Combining in vitro binding, cell-based assays, chemoinformatic modeling, structure-based design, unmasked cryptic binding site within exploited discover compounds affinity exceeding native ligand, antagonize function, or bear no resemblance Cbl. These data demonstrate how privileged biphenyl-like scaffold effectively targets optimizing π-stacking interactions pocket.

Language: Английский

---

Prediction and Explainable Analysis of Molecular Weight Distribution of Polystyrene Based on Machine Learning and SHAP

Macromolecular Reaction Engineering, Journal Year: 2025, Volume and Issue: unknown

Published: March 25, 2025

Abstract Molecular weight distribution (MWD) is crucial for the product performance of polymers. In order to explore how process conditions affect molecules with different chain lengths, this study conducts a large number polystyrene simulations based on polymerization kinetics and validates them through pilot plant data generate reliable dataset. Machine learning methods are employed predict average molecular weights conversion rates. Compared extreme gradient boosting (XGBoost) support vector regression (SVR), fully connected neural network (FCNN) shows best performance. Furthermore, an improved FCNN model feature extractor residual structure developed MWD accurately. The polymer divided into 10 bins length, influence revealed SHapley Additive exPlanations (SHAP). Notably, reducing feed mass fraction ethylbenzene increasing charging coefficient second pre‐polymerization reactor will lead increase low Raising temperature promote decrease in proportion small molecule polymers ultra‐large polymers, thereby narrowing MWD. addition, specific target can be effectively predicted by machine learning.

Language: Английский

---

The prediction of RNA-small molecule binding sites in RNA structures based on geometric deep learning

International Journal of Biological Macromolecules, Journal Year: 2025, Volume and Issue: 310, P. 143308 - 143308

Published: April 21, 2025

Language: Английский

---

Pharmacokinetics–Pharmacodynamics Modeling for Evaluating Drug–Drug Interactions in Polypharmacy: Development and Challenges

Clinical Pharmacokinetics, Journal Year: 2024, Volume and Issue: 63(7), P. 919 - 944

Published: June 18, 2024

Language: Английский

---