Research Appraisals
Evidence-based critical appraisals of the latest medical research, systematically evaluated using Oxford CEBM methodology.
Showing 3 appraisals
Molecular biology reports
Recent advances in the detection and functional analysis of circRNAs with short-read RNA sequencing-based methods
Circular RNAs (circRNAs) were first identified approximately 50 years ago in pathogenic viroids as single-stranded, covalently closed RNA molecules. Initially considered by-products of splicing, circRNAs are now recognised as an important class of regulatory RNAs involved in microRNA sponging, RNA-protein interactions, and cellular pathways. Their closed-loop structure, generated through backsplicing, confers resistance to exonucleolytic degradation and contributes to their stability. Owing to their tissue- and disease-specific expression, circRNAs have emerged as promising biomarkers for cancer, neurodegenerative disorders, and cardiovascular disease. Over the past decade, numerous bioinformatics tools utilising RNA-sequencing (RNA-seq) data have been developed for circRNA detection and analysis. Detection methods have evolved from manual split-read inspection to automated identification of the back spliced junction, while annotation pipelines now resolve the genomic origins and structural characteristics of circRNAs. Because individual circRNA callers vary considerably in sensitivity and specificity, a combined usage of tools in circRNA detection has become the preferred strategy for generating high-confidence datasets. Beyond their non-coding functions, increasing evidence suggests that some circRNAs possess protein-coding potential through open reading frames, cap-independent translation mechanisms, internal ribosome entry sites (IRESs), and N6-methyladenosine modifications. A new generation of bioinformatic tools can now assess the protein-coding potential of circRNAs, integrating the above features, as well as machine learning and deep learning approaches refining these predictions. This review summarises recently developed short-read RNA-seq bioinformatics tools for circRNA detection, consensus calling, annotation, and protein-coding potential prediction, with a particular focus on advances from the past five years that facilitate the identification of translatable circRNAs.
5 July 2026
Read appraisal →Journal of chemical information and modeling
SSPSPredictor: A Sequence- and Structure-Based Deep Learning Model for Predicting Phase-Separating Proteins
Liquid-liquid phase separation (LLPS) underlies the formation of biomolecular liquid condensates (also termed membraneless organelles, MLOs), which are essential for spatially organizing various biochemical processes within cells. Proteins that play a key role in driving condensate formation are termed phase-separating proteins (PSPs). Given that experimental identification of PSPs remains labor-intensive and time-consuming, multiple computational tools have been developed based on empirical features or deep learning. In this study, we propose SSPSPredictor, a novel multimodal predictive model for PSPs with folded or intrinsically disordered structures, leveraging the fusion of sequence information from a protein language model ESM-2 and structural insights from a graph neural network GVP. Compared with existing tools, SSPSPredictor achieves a balanced performance in identifying PSPs, predicting relative LLPS propensities, and recognizing key regions that drive LLPS. More importantly, SSPSPredictor exhibits good interpretability in identifying driving regions along protein sequences, although no relevant supervision was provided during training. Further predictive analysis of the human proteome using SSPSPredictor reveals that the proportion of intrinsically disordered proteins (IDPs) undergoing LLPS is significantly higher than that of the folded proteins. In addition, pathogenic variants, especially those located in disordered regions, exhibit a higher LLPS propensity than other mutations, uncovering a link between LLPS and diseases at the amino acid level. An online Web server for SSPSPredictor (https://bio-comp.ucas.ac.cn/SSPSPredictor/) and its source code (https://github.com/ucas-biocomp/SSPSPredictor) are available.
23 June 2026
Read appraisal →Biogerontology
RNA-binding proteins in aging and age-related diseases: roles, mechanisms, and a large language model analysis
Aging is a progressive and irreversible biological process that contributes to the pathogenesis of numerous age-related diseases. Elucidating the molecular mechanisms of aging is crucial for promoting healthy aging and extending healthspan. RNA-binding proteins (RBPs) are pivotal regulators of post-transcriptional gene expression and play essential roles in diverse biological processes. RBPs interact with both coding and non-coding RNAs to regulate RNA metabolism, stability, localization, and translation. Dysregulated RBP-RNA interactions have been closely associated with aging and age-related diseases. This review systematically summarizes the structural characteristics of RBPs and the evolution of methods used to study them. We focus on the molecular mechanisms of six key RBPs, namely HuR, AUF1, TTP, IGF2BP2, QKI, and LARP7, in the context of aging and age-related diseases. In addition, we discuss the regulatory functions of post-translational modifications of RBPs. Furthermore, we provide a multidimensional overview of RBP involvement in aging and age-related diseases through large language model (LLM)-based text-mining analysis. Our study provides a foundation for the comprehensive characterization of RBPs in aging and age-related diseases.
22 June 2026
Read appraisal →