Viruses are key infectious agents causing large-scale mortality among humans every year with the recent example of the devastating COVID-19 pandemic. RNA viruses are often associated with higher pathogenicity and their molecular mechanisms of entry into host cells, replication, transmission, and propagation remain poorly understood. To facilitate virus replication and propagation, these viruses regulate the host gene transcription and translation, key cellular processes like apoptosis, autophagy, innate as well as adaptive immune responses, lipid biosynthesis, and other key host signaling processes. They hijack and exploit the host cell machinery by modulating the expression of host proteins and ncRNAs. Some of the regulatory RNAs that these viruses modulate involve lncRNAs, circRNAs and ceRNA networks. RNA viruses alter or control key cellular processes of these RNAs by directly or indirectly interacting with both host and viral proteins and other non-coding RNAs. Through the alteration of regulation of non-coding RNAs, the viruses have been demonstrated to alter several biological processes primarily metabolic pathways, inflammation, and immune pathways like interferon signaling and other innate immune pathways to aid virus replication, pathogenesis and immune evasion. These differentially expressed regulatory RNAs can be further studied to develop potential antiviral therapeutics and vaccines. CeRNA networks involve the role of protein-coding mRNAs and competing ncRNAs like lncRNA, miRNA, and circRNA. This review explores some of the reported modulations of host lncRNAs, circRNAs, and ceRNA networks in RNA virus infections to understand their importance and role in virus pathogenesis.

Viroids are the simplest known infectious agents, consisting of noncoding circular RNAs that replicate autonomously in plant hosts while inducing profound and systemic alterations in host biology. This apparent paradox positions viroids as powerful models for understanding how RNA molecules encode biological information independently of protein translation. A hallmark of viroid infection is the production of viroid-derived small interfering RNAs (vd-siRNAs) by the host RNA silencing machinery. Here, we propose that vd-siRNAs function not merely as antiviroid byproducts, but as active effectors that translate RNA structural information into gene regulatory activity. Through sequence complementarity and selective incorporation into Argonaute complexes, vd-siRNAs influence host gene expression at multiple levels, including transcript abundance, protein accumulation, epigenetic states, and metabolic pathways. Importantly, their effects extend beyond individual targets and are consistent with broader perturbations of interconnected host regulatory networks, potentially contributing to coordinated transcriptional reprogramming and disease-associated phenotypes. By integrating molecular, transcriptomic, epigenetic, and systems-level evidence, we propose a "structure-to-network" perspective in which viroid RNAs function as minimal informational systems capable of interfacing with complex cellular pathways. Within this context, vd-siRNAs emerge as mechanistic links between RNA architecture and multiscale biological responses. Finally, we discuss how this framework expands our understanding of RNA-mediated regulation and highlights the broader regulatory potential of structured noncoding RNAs.

PNLDC1 is a member of the DEDD deadenylases, known for its specificity in poly(A) tail removal during mRNA turnover. While initially characterized for its function in pre-piRNA trimming critical to spermatogenesis and transposon silencing, its broader deadenylase activity which is especially pronounced during early development and silenced during differentiation, remains poorly understood. In this study, we mapped the epigenetic regulation of Pnldc1 during mESC differentiation, uncovering its promoter methylation pattern associated with transcriptional repression. Pnldc1-knockout mESC models established via CRISPR/Cas9 were shown to retain viability and pluripotency. RNA-seq and subsequent polysome profiling revealed that Pnldc1 deficiency leads to the downregulation of gene networks involved in translation and a lower overall translational rate. Notably, PNLDC1 has been linked to epigenetic changes in various cancers and is associated with oncogenic signaling pathways. Using a homology model based on the related PARN deadenylase, we conducted virtual screening of compound libraries and selected 12 candidates for experimental testing. Experimental validation revealed that two compounds significantly impacted genes involved in stress response and signal transduction, while also suppressing aberrant proliferation. These findings highlight PNLDC1's role in regulating mRNA stability during differentiation and proliferation and as a candidate for targeted therapeutic intervention.

G3BP1 is a regulator of stress granule assembly capable of driving condensation of RNA in vitro. Recently, we purified stress granule cores from yeast and mammalian cells, and found that in addition to RNA and proteins, they contain large amounts of circular double-stranded DNA (dsDNA). Since G3BP1 has been implicated in DNA-RNA heteroduplex resolution, we compared the ability of G3BP1 to drive condensation of cytoplasmic RNA, circular cytoplasmic dsDNA, and total nucleic acids (comprising both, circular dsDNA and RNA) extracted from purified stress granule cores of HEK293T cells. Imaging under near-physiologic conditions (1.7 mM MgCl2, no crowding agents, 0.1 ng/μL nucleic acids) revealed that G3BP1 most effectively induces condensation of the DNA-RNA mixture from stress granule cores. Comparison of the disruption of these condensates by treatment with RNase A, DNase, or RNase H further shows that they are especially susceptible to the latter two. The sensitivity to DNase and RNase H of G3BP1-induced condensates of the stress granule-derived nucleic acid mixtures underscores the importance of DNA in these cytoplasmic assemblies.

Errors in transcription generate transient RNA variants that are amplified through translation, with potential consequences for cellular phenotypes and RNA-based therapeutics. Accurate detection of these rare events relies on high-fidelity sequencing approaches such as Circle-Seq, but technical artifacts introduced during RNA preparation can distort error profiles and frequencies. Here, we systematically evaluate the impact of bacterial RNA sample preparation on Circle-Seq outputs. We show that RNA extraction and ribosomal RNA depletion introduce artifactual oxidative damage signatures, particularly C>U and G>A substitutions, that inflate apparent transcription error rates. Optimized mRNA preparation conditions substantially reduce these artifacts. In addition, we identify an RNase III fragmentation bias caused by its preference for structured RNAs, resulting in uneven transcriptome coverage. Chaotropic monovalent ions mitigate this bias and improve coverage uniformity. Finally, our optimized protocol captured, with improved accuracy, the frequency and spectrum of transcription errors in absence of the transcription fidelity factor GreA. Together, these results define the technical limitations of Circle-Seq and provide practical solutions to improve the accuracy and interpretability of transcription error measurements in bacterial systems.

Riboswitches are RNA-based genetic control elements that provide a unique mechanism of gene regulation. They function without the participation of proteins and are believed to represent ancient regulatory systems on the evolutionary timescale. To expand the search for novel riboswitches of significant interest, such as eukaryotic riboswitches beyond the purine riboswitch class, the integration of machine learning into RNA design offers considerable potential to improve the effectiveness of traditional search methods. Many riboswitch aptamers that could be targeted in such searches are larger than purine riboswitch aptamers, making their design time-consuming (e.g., SAM, lysine, FMN, glycine, and cobalamin riboswitches). In an example problem unrelated to the discovery of novel riboswitches, namely challenges in the eteRNA game, it has been shown that a transformer encoder-decoder model is effective in solving difficult inverse RNA folding challenges. In the outlined approach, machine learning is used as a preprocessing step before RNA design, rather than relying on random inputs to the Monte Carlo method, and the RNA design step crucially depends on a biophysical model for the forward problem of RNA folding prediction. It is therefore suggested to optimize the training of the machine learning model and then utilize this approach for riboswitch discovery to make searches for riboswitches larger than purine aptamers more efficient and robust.

Only a small minority of annotated long noncoding RNAs (lncRNAs) have a molecular function established by direct experiment, and the gap widens with every new catalog. Existing reviews catalog methods thoroughly but rarely formalize how negative or discordant wet-laboratory results should feed back into computational prioritization, so the iterative refinement that lncRNA biology requires is left implicit. We set out the Hypothesis-Driven Loop, which recasts lncRNA functional analysis as a cycle of three phases: Generation (computational priors), Testing (wet validation), and Refinement (integration that updates the priors for the next round). The cycle itself is Platt's strong inference; what is specific to lncRNAs are the constraints that reshape it-a detectability floor that gates every functional claim, an orders-of-magnitude asymmetry between prediction and validation throughput, and the ambiguity of what a locus-level perturbation actually removes-together with the deliverable each phase must produce. We map existing tools onto the appropriate phase; give a one-page Loop worksheet with an odds-form procedure for updating priors and pre-declared stopping boundaries; and specify how a negative result is triaged as technical or biological before it may revise a hypothesis. Three worked programs-BRIC-seq and the SLiTs paradigm from our laboratory, and HOTAIR, assembled by independent groups from results that disagreed-show that Loops can be entered at different phases and that closure can be a community-level property. The claim is deliberately narrow: iterative dry-wet coupling, recorded so that it can be audited, is one discipline that may help narrow the annotation-function gap.

RNA-protein assemblies are fundamental organizers of eukaryotic cell biology, participating in diverse processes such as ribosome biogenesis in nucleoli, mRNA sequestration in nuclear speckles, and genome packaging in viral ribonucleoprotein (vRNP) complexes. Across these systems, multivalent RNA-protein interactions promote self-organization, contributing directly to liquid-liquid phase separation (LLPS) in nucleoli and nuclear speckles, while facilitating assembly and intermediate states in mature structurally organized vRNPs. This review critically examines the molecular grammar underlying these three paradigmatic systems, focusing on how sequence-encoded RNA-IDR chemistry determines condensate morphology, compositional selectivity, and material properties. We then evaluate the experimental and computational approaches used to characterize RNA-protein assemblies, with particular focus on coarse-grained (CG) forcefield development across HPS, Mpipi, CALVADOS, and 1BPA families. We also discuss how emerging multi-domain protein (MDP) models and RNA-compatible forcefields are optimizing the predictive accuracy of molecular simulations. Finally, we explore physics-based and data-driven predictors of LLPS, and reiterate the need for a framework that can predict sequence-based heterotypic IDR-RNA co-phase separation.

The 5' untranslated region (UTR) of mRNAs is crucial to regulate translation initiation. Diversity in the 5' UTR can be created by alternative transcription start site (TSS) selection and alternative splicing. Alternative TSS directly impact the first exon length or alternative first exon usage, while alternative splicing further contributes to generating different 5' UTR isoforms by for example, cassette exons inclusion, intron retention, and alternative 5' or 3' splice sites. Alternative TSS and alternative splicing in the 5' UTR directly affect translation efficiency by generating different isoforms that may include or exclude regulatory sequence elements and secondary structures in the 5' UTR. Among the regulatory elements, 5' UTRs may harbor translation-inhibitory upstream open reading frames (uORFs) as well as internal ribosome entry sites (IRESs), which are able to recruit the ribosome in a cap-independent manner. Both uORF- and IRES-mediated mechanisms contribute to target-specific protein production, especially under stress conditions associated with an inhibition of global translation. Mechanisms generating different 5' UTR isoforms therefore represent an additional sophisticated layer to regulate transcript-selective translation efficiency.

RNAs navigate highly regulated life cycles that include synthesis, processing, ribonucleoprotein complex formation, localization, and degradation. However, most transcriptome-wide methods capture static, cell-wide averages that obscure the spatiotemporal aspects-where and when-of RNA life. Here we review recent advances that bring spatial and temporal resolution to transcriptome and RNA-protein interaction analyses by labeling RNA either in defined subcellular locales (RNA proximity labeling) or during defined transcriptional time windows to capture the kinetics of RNA synthesis, degradation, and processing (RNA metabolic labeling). We further discuss how RNA labeling approaches can be integrated with RNA-protein interaction mapping, opening new possibilities for studying ribonucleoprotein dynamics in space and time. Finally, we highlight emerging developments and outline future directions for spatiotemporal analysis of the RNA life cycle.

The immature HIV-1 virion is assembled by the Gag polyprotein using inositol hexakisphosphate (IP6) as an essential assembly co-factor. Gag binds the genomic RNA Psi packaging signal via the nucleocapsid (NC) domain and associates with the plasma membrane via the matrix (MA) domain. Previous studies revealed that Gag exists in both compact (C) and extended (E) conformational states in solution. Only E-Gag formed virus-like particles with the correct size and IP6 shifted the equilibrium of DNA-bound Gag to the E state. The influence of specific RNA elements on this conformational change is unknown. In this work, a dual dye-labeled Gag was prepared for probing the effect of RNA binding on Gag conformation using Förster resonance energy transfer (FRET). In low salt and in the absence of other factors, Gag was primarily in the C state. Psi RNA binding induced a more significant FRET decrease than binding to non-Psi RNAs, consistent with a shift to E-Gag. IP6 alone also promoted the E-Gag state in the absence and presence of RNA. Atomistic molecular dynamics simulations are consistent with and provide detail into the role of NC-Psi RNA binding in the conformational switch of C-Gag to assembly-competent E-Gag. Simulations also showed that this switch is driven by capsid (CA) linker domain orientational flexibility and MA-CA unbinding dynamics. Thus, the highly flexible multi-domain Gag polyprotein leverages both viral and host cell factors to sample and stabilize distinct conformations, thereby orchestrating the viral assembly process.

RNA molecules form specific 3D structures that facilitate a variety of functions through interactions with other macromolecules. Many RNA viral genomes maintain these structures to interact with and evade host immunity machinery. One such element, the competitive inhibitor RNA (ciRNA), discovered in the protein coding region of the poliovirus serotype 1 (PV1) genome, inhibits a host antiviral protein, ribonuclease L (RNase L). Although some functionally essential structural motifs of the PV1 ciRNA have been studied, the extent of its evolutionary conservation and other structural requirements remained unexplored. Here we combined bioinformatic and biochemical techniques to further define the requirements of a functional ciRNA and assess its phylogenetic distribution. We systematically mutated ciRNA structural features, verifying that ciRNA inhibitory activity requires a conserved loop E motif and a long-range base-pairing interaction, but its peripheral stems are dispensable and in fact a circularly permuted version maintains function. A structure-based homology search identified potential ciRNAs across the Picornaviridae family, but only a subset of those tested were functional-all are in Enterovirus coxsackiepol. When structural features needed for function were transposed from PV1 ciRNA to an RNA unable to inhibit RNase L, the chimeric RNAs did not gain wild-type function, and chemical probing data revealed that these nonfunctional RNAs are unable to form the correct secondary structure. Overall, the dual constraints of encoding a protein and forming a specific functional structure appear to not only limit the sequence diversity, but also the phylogenetic distribution, of ciRNAs.

A relevant percentage of genetic diseases (~10%) are caused by premature termination codon (PTC) mutations, including cystic fibrosis (CF). CF patients carrying a nonsense mutation in the CFTR gene are insensitive to CFTR modulators and cannot benefit from the pharmacological therapy with Kaftrio®. The rescue of CFTR with PTCs faces two limiting steps: 1) the nonsense-mediated RNA decay (NMD) mechanism that, by degrading mRNAs carrying PTCs, strongly reduces CFTR transcript levels; 2) PTCs cause the arrest of protein synthesis, with the production of a truncated and nonfunctional CFTR protein. The combination of NMD inhibitors with readthrough agents, which overcome the translation arrest, can maximize rescue of CFTR with PTCs. NMD inhibitors can also be effective in the absence of a readthrough agent for PTCs localized at the carboxy-terminus of the CFTR sequence such as W1282X. In this work, we screened a chemical library with a CFTR functional assay to find novel compounds able to modulate NMD mechanism. At the end of the screening, we identified and validated by secondary assays three possible candidates acting as NMD inhibitors/modulators that could increase CFTR rescue in patients with PTCs, namely CC-115, samotolisib and elimusertib. The study also provides new insights into the changes occurring to cells transcriptome after the inhibition of the physiological NMD pathway, improving our knowledge of possible side effects of NMD inhibitors. This could help in the development of new compounds with more efficacy and less toxicity, that could be useful for the treatment of different genetic disorders caused by PTCs.

Numbers of research articles studying circRNAs have increased rapidly since 2017. Previous analyses of human circRNA articles in two high impact factor cancer research journals identified papers with wrongly identified nucleotide sequence reagents and circRNAs whose identities could not be independently verified. In the present study, verification of human nucleotide sequence reagent and cell line identities in retracted circRNA articles published from 2017-2021 in high impact factor journals found wrongly identified nucleotide sequences and/or cell lines in all 13 retracted papers. Similar analyses of human circRNA papers published in high impact factor journals in 2022 found at least one wrongly identified reagent in 51% (60/118) papers, where 52/118 (44%) papers described wrongly identified nucleotide sequence reagents and 22/118 (19%) described wrongly identified cell lines. In addition, some 2022 circRNA papers described circRNA probes of uncertain targeting specificity (34/118, 29%), non-verifiable nucleotide sequence reagents (25/118, 21%), claimed miRNA reagents whose miRNA targets were not indexed by MirGeneDB (13/118, 11%) and/or non-verifiable cell line identifiers (6/118, 5%). Overall, wrongly identified and/or queried reagents were noted in 73% (86/118) 2022 circRNA papers in high IF journals, highlighting the need for critical engagement with the human circRNA literature.

N[6]-methyladenosine (m[6]A) plays an important role in translation control and, particularly, in cap-independent initiation. The impact of m[6]A is usually studied in the context of its location in the transcripts, but the global impact of m[6]A along mRNAs remains unclear. Here, we combined ribosome profiling under conditions of mTOR inhibition and m[6]A mapping to identify distinct subsets of mRNAs that differ in their sensitivity to the suppression of cap-dependent initiation. We found that the sensitivity strongly correlated with the total m[6]A methylation of the transcripts. Further, upon decreased mTOR activity, m[6]A methylation facilitated the enhanced association of mRNAs with components of the eIF4F complex. Thus, the efficiency of cap-independent translation initiation is primarily defined not by precise localization but by the total level of m[6]A methylation, and m[6]A has a compensatory role in maintaining translation when the canonical cap-dependent pathway is impaired. Our findings underscore the significance of contemplating global m[6]A methylation status as a pivotal element in translational control, particularly under stress or signaling perturbations.