Secondary Structure Model Research Articles

Prediction of RNA secondary structure including pseudoknots for long sequences.

RNA structural elements called pseudoknots are involved in various biological phenomena including ribosomal frameshifts. Because it is infeasible to construct an efficiently computable secondary structure model including pseudoknots, secondary structure prediction methods considering pseudoknots are not yet widely available. We developed IPknot, which uses heuristics to speed up computations, but it has remained difficult to apply it to long sequences, such as messenger RNA and viral RNA, because it requires cubic computational time with respect to sequence length and has threshold parameters that need to be manually adjusted. Here, we propose an improvement of IPknot that enables calculation in linear time by employing the LinearPartition model and automatically selects the optimal threshold parameters based on the pseudo-expected accuracy. In addition, IPknot showed favorable prediction accuracy across a wide range of conditions in our exhaustive benchmarking, not only for single sequences but also for multiple alignments.

A new species of Succinea (Gastropoda: Stylommatophora) from the Puna highlands of Argentina

A new species of Succinea Draparnaud, 1801, S. punasp. nov., from the Puna highlands in Catamarca Province, Argentina, is described. Seventeen water bodies were surveyed in the Puna ecoregion of Catamarca, with gastropods being found only at two localities, El Peñón and Guanchín River. We explored the morphological features of shells, reproductive system, radulae, and mandibles of specimens from each site. Although we found slight differences in the shell and the terminal genitalia, with specimens exhibiting an epiphallus folding over itself with terminal globular appearance, whereas others with an epiphallus cylindrical, we conclude that these are intraspecific variabilities. We used DNA sequences of the 16S-rRNA gene to explore genetic distances and genealogical relationships among both populations and other species of Succinea. We found maximum genetic interpopulation distances of up to 2.38%. Mean interspecific divergences among S. punasp. nov. and other Succinea species ranged between 7.14% and 15.21%. A secondary structure model of domains IV and V of the 16S-rRNA gene was developed, with polymorphic sites being concentrated in loop regions. A GIS analysis was carried out using a selection of bioclimatic variables to characterize the environment of the new species in comparison to other close Succinea species from South America. The analysis showed that the particular environmental conditions of the Puna in Catamarca are hardly present in other zones of South America, supporting our hypothesis that S.puna is endemic to that area. The restricted distribution of S. punasp. nov. in a single ecoregion with extreme climatic and physiographic characteristics, high altitude habitat and its microhabitat specialization, makes this species potentially vulnerable to impacts of future anthropogenic changes.

Translation of human Δ133p53 mRNA and its targeting by antisense oligonucleotides complementary to the 5'-terminal region of this mRNA.

The p53 protein is expressed as at least twelve protein isoforms. Within intron 4 of the human TP53 gene, a P2 transcription initiation site is located and this transcript encodes two p53 isoforms: Δ133p53 and Δ160p53. Here, the secondary structure of the 5′-terminal region of P2-initiated mRNA was characterized by means of the SHAPE and Pb2+-induced cleavage methods and for the first time, a secondary structure model of this region was proposed. Surprisingly, only Δ133p53 isoform was synthetized in vitro from the P2-initiated p53 mRNA while translation from both initiation codons occurred after the transfection of vector-encoded model mRNA to HCT116 cells. Interestingly, translation performed in the presence of the cap analogue suggested that the cap-independent process contributes to the translation of P2-initiated p53 mRNA. Subsequently, several antisense oligonucleotides targeting the 5′-terminal region of P2-initiated p53 mRNA were designed. The selected oligomers were applied in in vitro translation assays as well as in cell lines and their impact on the Δ133p53 synthesis and on cell viability was investigated. The results show that these oligomers are attractive tools in the modulation of the translation of P2-initiated p53 mRNA through attacking the 5′ terminus of the transcript. Since cell proliferation is also reduced by antisense oligomers that lower the level of Δ133p53, this demonstrates an involvement of this isoform in tumorigenesis.

Submillisecond Conformational Transitions of Short Single-Stranded DNA Lattices by Photon Correlation Single-Molecule Förster Resonance Energy Transfer.

Thermally driven conformational fluctuations (or "breathing") of DNA play important roles in the function and regulation of the "macromolecular machinery of genome expression." Fluctuations in double-stranded (ds) DNA are involved in the transient exposure of pathways to protein binding sites within the DNA framework, leading to the binding of regulatory proteins to single-stranded (ss) DNA templates. These interactions often require that the ssDNA sequences, as well as the proteins involved, assume transient conformations critical for successful binding. Here, we use microsecond-resolved single-molecule Förster resonance energy transfer (smFRET) experiments to investigate the backbone fluctuations of short [oligo(dT)n] templates within DNA constructs that also serve as models for ss-dsDNA junctions. Such junctions, together with the attached ssDNA sequences, are involved in interactions with the ssDNA binding (ssb) proteins that control and integrate the functions of DNA replication complexes. We analyze these data using a chemical network model based on multiorder time-correlation functions and probability distribution functions that characterize the kinetic and thermodynamic behavior of the system. We find that the oligo(dT)n tails of ss-dsDNA constructs interconvert, on submillisecond time scales, between three macrostates with distinctly different end-to-end distances. These are (i) a "compact" macrostate that represents the dominant species at equilibrium; (ii) a "partially extended" macrostate that exists as minority species; and (iii) a "highly extended" macrostate that is present in trace amounts. We propose a model for ssDNA secondary structure that advances our understanding of how spontaneously formed nucleic acid conformations may facilitate the activities of ssDNA-associating proteins.

Structural 3D Domain Reconstruction of the RNA Genome from Viruses with Secondary Structure Models.

Three-dimensional RNA domain reconstruction is important for the assembly, disassembly and delivery functionalities of a packed proteinaceus capsid. However, to date, the self-association of RNA molecules is still an open problem. Recent chemical probing reports provide, with high reliability, the secondary structure of diverse RNA ensembles, such as those of viral genomes. Here, we present a method for reconstructing the complete 3D structure of RNA genomes, which combines a coarse-grained model with a subdomain composition scheme to obtain the entire genome inside proteinaceus capsids based on secondary structures from experimental techniques. Despite the amount of sampling involved in the folded and also unfolded RNA molecules, advanced microscope techniques can provide points of anchoring, which enhance our model to include interactions between capsid pentamers and RNA subdomains. To test our method, we tackle the satellite tobacco mosaic virus (STMV) genome, which has been widely studied by both experimental and computational communities. We provide not only a methodology to structurally analyze the tertiary conformations of the RNA genome inside capsids, but a flexible platform that allows the easy implementation of features/descriptors coming from both theoretical and experimental approaches.

3′UTR SL-IV and DB1 Regions Contribute to Japanese Encephalitis Virus Replication and Pathogenicity

Japanese encephalitis virus (JEV), a mosquito-borne flavivirus that causes fatal neurological disease in humans, is one of the most important emerging pathogens of public health significance. JEV is maintained in an enzootic cycle and causes reproductive failure in pigs. Notably, the shift in JEV genotypes is not fully protected by existing vaccines, so the development of a candidate vaccine is urgently needed. In this study, we compared pathogenicity between Japanese encephalitis virus SA14 and BJB (isolated from humans in the 1970s) strains. We found that the BJB strain was attenuated in mice and that there was no case fatality rate. The growth rate of BJB was higher than SA14 virus in BHK-21 cells. Based on the sequence alignment of the viral genome between the SA14 and BJB virus strains, some mutations at sites 248, 254, 258, and 307 were observed in the 3′ untranslated region (3′UTR). The 3′UTR of JEV plays a very important role in the viral life cycle. Furthermore, using a reverse genetic system, we conducted and rescued the parental JEV strain SA14 (T248, A254, and A258) and the mutant virus rSA14-3′UTRmut (T248C, A254G, A258G, and 307G). Through an analysis of the RNA secondary structure model of the 3′UTR, we discovered that the mutations of T248C, A254G, and A258G reduced the apiculus ring and increased the lateral ring significantly in the stem-loop structures IV (SL-IV) structure region of 3′UTR. Moreover, the insertion of 307G added a ring to the dumbbell structure 1 (DB1) structure region. Strikingly, these RNA secondary structure changes in 3′UTR of rSA14-3′UTRmut increased viral negative chain RNA production and enhanced the replication ability of the virus in BHK-21 cells. However, in vivo mouse experiments illustrated that the rSA14-3′UTRmut virus significantly decreased the neurovirulence of JEV. These results affirmed that the JEV SL-IV and DB1 regions play an important role in viral proliferation and pathogenicity. Taken together, we complement the study of RNA element function in the 3′UTR region of JEV by providing a new target for the rational design of live attenuated candidate vaccines and the increase of virus production.

Protein Structure and Function Prediction of SARS-CoV 2: Prospective Antivirus Active Drug Binding Sites

Today a newly emerged corona-virus known as SARS-CoV 2 has become a cause of global health concern and took away the lives of large number of people throughout the world. Corona-viruses are the enveloped virus with positive single stranded genome of 26.4 to 31.7 kb. Envelop of the corona-virus is made up of four structural proteins namely envelop protein (E), membrane protein (M), spike protein (S) and nucleocapsid protein (N). These four proteins are responsible for the overall shape and size of the virus (structure of virus). Envelop protein forms ion channels, membrane protein is responsible for the shape of the virus, spike protein is responsible for the entering inside the target host cell by binding to host receptor and nucleocapsid protein binds to the single stranded RNA genome of the virus forming multiple copies. We investigate the reliability and homogeneity among all the corona-virus species such as MERS CoV, Bat-CoV HKU4, Transmissible gastro-enteritis coronavirus (TGEV), Porcine epidemic diarrhea virus (PEDV), HCoV-229E and Whale-CoV SW1, M-CoV, Hedgehog coronavirus 1, Bulbul-CoV HKU11 etc, using a binary graph which is helpful in the findings of sequence reliability, secondary and tertiary structure model prediction using advanced model builder, we build the model of various protein/gene products by selecting them from SARS-COV 2, which further helpful in the finding of target-ligand binding for future therapeutic applications. Consequently by modeling the structure of the proteins we bring into being that envelop protein have pentameric protein lipid pores that allow ion transportation and were able to depict the active drug binding sites.

Structural evolution of SLA promoter in mosquito-borne flaviviruses: A sequence-structure based phylogenetic framework

All the flaviviruses have a Y-shaped stem-loop secondary structure known as the SLA element, and the structural features of this element are crucial to initiating the infection cycle. The present study particularly investigated how flaviviruses retained the common core SLA element secondary structure during the species evolution by selecting mosquito-borne flaviviruses (MBFVs) as a case study. The detailed search of nucleotide substitutions in species-wise consensus SLA secondary structure models suggested that the compensatory and hemi-compensatory base changes in the helices are crucial to preserving the common core secondary structure. In contrast to the coding region-based phylogeny, the SLA sequence-structure-based phylogenetic tree revealed an intriguing evolutionary relationship among MBFVs. Overall, this paper demonstrated for the first time the efficacy of RNA secondary structures as a phylogenetic marker to study the RNA virus evolution.

Identification of RNAs Engaged in Direct RNA-RNA Interaction with a Long Non-Coding RNA

The growing role attributed nowadays to long non-coding RNAs (lncRNA) in physiology and pathophysiology makes it crucial to characterize their interactome by identifying their molecular partners, DNA, proteins and/or RNAs. The latter can interact with lncRNA through networks involving proteins, but they can also be engaged in direct RNA/RNA interactions. We, therefore, developed an easy-to-use RNA pull-down procedure that allowed identification of RNAs engaged in direct RNA/RNA interaction with a lncRNA using psoralen, a molecule that cross-links only RNA/RNA interactions. Bioinformatics modeling of the lncRNA secondary structure allowed the selection of several specific antisense DNA oligonucleotide probes with a strong affinity for regions displaying a low probability of internal base pairing. Since the specific probes that were designed targeted accessible regions throughout the length of the lncRNA, the RNA-interaction zones could be delineated in the sequence of the lncRNA. When coupled with a high throughput RNA sequencing, this protocol can be used for the whole direct RNA interactome studies of a lncRNA of interest.

Quantum Information in the Protein Codes, 3-Manifolds and the Kummer Surface

Every protein consists of a linear sequence over an alphabet of 20 letters/amino acids. The sequence unfolds in the 3-dimensional space through secondary (local foldings), tertiary (bonds) and quaternary (disjoint multiple) structures. The mere existence of the genetic code for the 20 letters of the linear chain could be predicted with the (informationally complete) irreducible characters of the finite group Gn:=Zn⋊2O (with n=5 or 7 and 2O the binary octahedral group) in our previous two papers. It turns out that some quaternary structures of protein complexes display n-fold symmetries. We propose an approach of secondary structures based on free group theory. Our results are compared to other approaches of predicting secondary structures of proteins in terms of α helices, β sheets and coils, or more refined techniques. It is shown that the secondary structure of proteins shows similarities to the structure of some hyperbolic 3-manifolds. The hyperbolic 3-manifold of smallest volume—Gieseking manifold—some other 3 manifolds and the oriented hypercartographic group are singled out as tentative models of such secondary structures. For the quaternary structure, there are links to the Kummer surface.

The mitochondrial genome of the semi-slug Omalonyx unguis (Gastropoda: Succineidae) and the phylogenetic relationships within Stylommatophora.

Here we report the first complete mitochondrial genome of the semi-slug Omalonyx unguis (d'Orbigny, 1836) (Gastropoda: Succineidae). Sequencing was performed on a specimen from Argentina. Assembly was performed using Sanger data and Illumina next generation sequencing (NGS). The mitogenome was 13,984 bp in length and encoded the 37 typical Metazoan genes. A potential origin for mitochondrial DNA replication was found in a non-coding intergenic spacer (49 bp) located between cox3 and tRNA-Ile genes, and its secondary structure was characterized. Secondary structure models of the tRNA genes of O. unguis largely agreed with those proposed for other mollusks. Secondary structure models for the two rRNA genes were also obtained. To our knowledge, the 12S-rRNA model derived here is the first complete one available for mollusks. Phylogenetic analyses based on the mitogenomes of O. unguis and 37 other species of Stylommatophora were performed using amino acid sequences from the 13 protein-coding genes. Our results located Succineoidea as a sister group of Helicoidea + Urocoptoidea, similar to previous studies based on mitochondrial genomes. The gene arrangement of O. unguis was identical to that reported for another species of Succineoidea. The unique rearrangements observed for this group within Stylommatophora, may constitute synapomorphies for the superfamily.

An expanded class of histidine-accepting viral tRNA-like structures.

Structured RNA elements are common in the genomes of RNA viruses, often playing critical roles during viral infection. Some viral RNA elements use forms of tRNA mimicry, but the diverse ways this mimicry can be achieved are poorly understood. Histidine-accepting tRNA-like structures (TLSHis) are examples found at the 3′ termini of some positive-sense single-stranded RNA (+ssRNA) viruses where they interact with several host proteins, induce histidylation of the RNA genome, and facilitate processes important for infection, to include genome replication. As only five TLSHis examples had been reported, we explored the possible larger phylogenetic distribution and diversity of this TLS class using bioinformatic approaches. We identified many new examples of TLSHis, yielding a rigorous consensus sequence and secondary structure model that we validated by chemical probing of representative TLSHis RNAs. We confirmed new examples as authentic TLSHis by demonstrating their ability to be histidylated in vitro, then used mutational analyses to imply a tertiary interaction that is likely analogous to the D- and T-loop interaction found in canonical tRNAs. These results expand our understanding of how diverse RNA sequences achieve tRNA-like structure and function in the context of viral RNA genomes and lay the groundwork for high-resolution structural studies of tRNA mimicry by histidine-accepting TLSs.

De novo 3D models of SARS-CoV-2 RNA elements from consensus experimental secondary structures.

The rapid spread of COVID-19 is motivating development of antivirals targeting conserved SARS-CoV-2 molecular machinery. The SARS-CoV-2 genome includes conserved RNA elements that offer potential small-molecule drug targets, but most of their 3D structures have not been experimentally characterized. Here, we provide a compilation of chemical mapping data from our and other labs, secondary structure models, and 3D model ensembles based on Rosetta's FARFAR2 algorithm for SARS-CoV-2 RNA regions including the individual stems SL1-8 in the extended 5′ UTR; the reverse complement of the 5′ UTR SL1-4; the frameshift stimulating element (FSE); and the extended pseudoknot, hypervariable region, and s2m of the 3′ UTR. For eleven of these elements (the stems in SL1–8, reverse complement of SL1–4, FSE, s2m and 3′ UTR pseudoknot), modeling convergence supports the accuracy of predicted low energy states; subsequent cryo-EM characterization of the FSE confirms modeling accuracy. To aid efforts to discover small molecule RNA binders guided by computational models, we provide a second set of similarly prepared models for RNA riboswitches that bind small molecules. Both datasets (‘FARFAR2-SARS-CoV-2’, https://github.com/DasLab/FARFAR2-SARS-CoV-2; and ‘FARFAR2-Apo-Riboswitch’, at https://github.com/DasLab/FARFAR2-Apo-Riboswitch’) include up to 400 models for each RNA element, which may facilitate drug discovery approaches targeting dynamic ensembles of RNA molecules.

In vivo structure of the Ty1 retrotransposon RNA genome.

Long terminal repeat (LTR)-retrotransposons constitute a significant part of eukaryotic genomes and influence their function and evolution. Like other RNA viruses, LTR-retrotransposons efficiently utilize their RNA genome to interact with host cell machinery during replication. Here, we provide the first genome-wide RNA secondary structure model for a LTR-retrotransposon in living cells. Using SHAPE probing, we explore the secondary structure of the yeast Ty1 retrotransposon RNA genome in its native in vivo state and under defined in vitro conditions. Comparative analyses reveal the strong impact of the cellular environment on folding of Ty1 RNA. In vivo, Ty1 genome RNA is significantly less structured and more dynamic but retains specific well-structured regions harboring functional cis-acting sequences. Ribosomes participate in the unfolding and remodeling of Ty1 RNA, and inhibition of translation initiation stabilizes Ty1 RNA structure. Together, our findings support the dual role of Ty1 genomic RNA as a template for protein synthesis and reverse transcription. This study also contributes to understanding how a complex multifunctional RNA genome folds in vivo, and strengthens the need for studying RNA structure in its natural cellular context.

Relative Information Gain: Shannon entropy-based measure of the relative structural conservation in RNA alignments.

Structural characterization of RNAs is a dynamic field, offering many modelling possibilities. RNA secondary structure models are usually characterized by an encoding that depicts structural information of the molecule through string representations or graphs. In this work, we provide a generalization of the BEAR encoding (a context-aware structural encoding we previously developed) by expanding the set of alignments used for the construction of substitution matrices and then applying it to secondary structure encodings ranging from fine-grained to more coarse-grained representations. We also introduce a re-interpretation of the Shannon Information applied on RNA alignments, proposing a new scoring metric, the Relative Information Gain (RIG). The RIG score is available for any position in an alignment, showing how different levels of detail encoded in the RNA representation can contribute differently to convey structural information. The approaches presented in this study can be used alongside state-of-the-art tools to synergistically gain insights into the structural elements that RNAs and RNA families are composed of. This additional information could potentially contribute to their improvement or increase the degree of confidence in the secondary structure of families and any set of aligned RNAs.

Characterization of Euplotes lynni nov. spec., E. indica nov. spec. and description of E. aediculatus and E. woodruffi (Ciliophora, Euplotidae) using an integrative approach

Four species belonging to the genus Euplotes have been investigated, namely: E. lynni nov. spec., E. indica nov. spec., E. aediculatus, and E. woodruffi. All populations are from India and were investigated using morphological and molecular markers. The phylogenetic relationships were inferred from small subunit ribosomal rRNA gene (SSU rRNA), internal transcribed spacer (ITS) region, and mitochondrial cytochrome c oxidase subunit I (COI) gene. Predicted secondary structure models for two new species using the hypervariable region of the SSU rRNA gene and ITS2 region support the distinctness of both species. Morphological characters were subjected to principal component analysis (PCA) and genetic variations were studied in-depth to analyze the relatedness of the two new species with their congeners. An integrative approach combining morphological features, molecular analysis, and ecological characteristics was carried out to understand the phylogenetic position of the reported species within the different clades of the genus Euplotes.

Secondary Structural Model of MALAT1 Becomes Unstructured in Chronic Myeloid Leukemia and Undergoes Structural Rearrangement in Cervical Cancer.

Long noncoding RNAs (lncRNAs) influence cellular function through binding events that often depend on the lncRNA secondary structure. One such lncRNA, metastasis-associated lung adenocarcinoma transcript 1 (MALAT1), is upregulated in many cancer types and has a myriad of protein- and miRNA-binding sites. Recently, a secondary structural model of MALAT1 in noncancerous cells was proposed to form 194 hairpins and 13 pseudoknots. That study postulated that, in cancer cells, the MALAT1 structure likely varies, thereby influencing cancer progression. This work analyzes how that structural model is expected to change in K562 cells, which originated from a patient with chronic myeloid leukemia (CML), and in HeLa cells, which originated from a patient with cervical cancer. Dimethyl sulfate-sequencing (DMS-Seq) data from K562 cells and psoralen analysis of RNA interactions and structure (PARIS) data from HeLa cells were compared to the working structural model of MALAT1 in noncancerous cells to identify sites that likely undergo structural alterations. MALAT1 in K562 cells is predicted to become more unstructured, with almost 60% of examined hairpins in noncancerous cells losing at least half of their base pairings. Conversely, MALAT1 in HeLa cells is predicted to largely maintain its structure, undergoing 18 novel structural rearrangements. Moreover, 50 validated miRNA-binding sites are affected by putative secondary structural changes in both cancer types, such as miR-217 in K562 cells and miR-20a in HeLa cells. Structural changes unique to K562 cells and HeLa cells provide new mechanistic leads into how the structure of MALAT1 may mediate cancer in a cell-type specific manner.

Classifying Conserved RNA Secondary Structures With Pseudoknots by Vector-Edit Distance

Secondary structures have been proved to relate with the great functional diversity of RNA. There have been many studies to predict and compare the RNA secondary structures. However, fast and accurate comparison of RNA secondary structures with arbitrary pseudoknots is a challenging issue due to the hidden but important structural properties, such as the distribution of stems and branches. In this paper, we construct a novel RNA secondary structure model called modified adjoining grammars binary tree ( $\mathrm {BTMG_{CSP}}$ ). It can not only represent the complex RNA secondary structure including arbitrary pseudoknots intuitively, but also reserve RNA secondary structure properties. Further, we propose vector-edit distance to measure the structure similarity between $\mathrm {BTMG_{CSP}}$ trees converted from RNA sequences and their secondary structures for classifying conserved stem pattern. The experimental results show that our method substantially reduces the memory and time consumption in contrast to previous algorithms, such as $O((n/k)^{2}) $ and $O(n/k) $ for time and space complexity, respectively. In particular, the AUC value of our method achieves 0.949 in PseudoBase.

Molecular and <em>in silico</em> Study of TP53 Codon 72 Polymorphism (rs1042522) in a Population Exposed to High Background Radiation in Mamuju-West Sulawesi

The evaluation of the tumor protein p53 (TP53) codon 72 polymorphism (rs1042522) status in a population exposed to high background radiation was performed in this study. Real-time polymerase chain reaction (Q-PCR) was used to genotype the rs1042522 polymorphism in 100 subjects from Takandeang, Salleto, and Ahu villages in Mamuju district, West Sulawesi. An in silico study was then conducted to identify the potential effects of the proline substitution associated with this polymorphism on protein stability. The in silico analysis was performed using three different computational tools, namely I-Mutant Suite, iStable, and Protein Variation Effect Analyzer (PROVEAN). Secondary and three-dimensional (3D) structural models for wild-type (WT) and variant TP53 were generated to predict potential structural changes in the protein. Electrostatic surface charge calculations were also performed to identify changes in the electrostatic charge of codon 72. The Pro72 and Arg72 frequencies among the inhabitants in the Takandeang, Salleto, and Ahu villages, who are Mandarese Austronesian-speaking ethnic group members, were 0.56 and 0.44, respectively. The in silico analysis revealed no negative effects on protein stability due to the proline substitution at codon 72, although molecular modeling showed several differences in the secondary and 3D structures of the TP53 variant compared with the structure of WT TP53. To determine the impact of the proline-coding allele on individual sensitivity to radiation exposure, we compared the micronucleus (MN) frequencies of each genotype. The results showed that the MN frequencies in individuals harboring the proline-coding allele were not significantly higher than those expressing the arginine-coding allele. In conclusion, this study revealed that the proline-coding allele for codon 72 of the TP53 gene represented the predominant genotype among Takandeang, Salleto, and Ahu villages inhabitants. The present study also demonstrated that individuals who carried the proline-coding allele were not more sensitive to radiation compared with those expressing the arginine-coding allele for codon 72 of the TP53 gene.

IRIS: A method for predicting in vivo RNA secondary structures using PARIS data

BackgroundRNA secondary structures play a pivotal role in posttranscriptional regulation and the functions of non‐coding RNAs, yet in vivo RNA secondary structures remain enigmatic. PARIS (Psoralen Analysis of RNA Interactions and Structures) is a recently developed high‐throughput sequencing‐based approach that enables direct capture of RNA duplex structures in vivo. However, the existence of incompatible, fuzzy pairing information obstructs the integration of PARIS data with the existing tools for reconstructing RNA secondary structure models at the single‐base resolution.MethodsWe introduce IRIS, a method for predicting RNA secondary structure ensembles based on PARIS data. IRIS generates a large set of candidate RNA secondary structure models under the guidance of redistributed PARIS reads and then uses a Bayesian model to identify the optimal ensemble, according to both thermodynamic principles and PARIS data.ResultsThe predicted RNA structure ensembles by IRIS have been verified based on evolutionary conservation information and consistency with other experimental RNA structural data. IRIS is implemented in Python and freely available at http://iris.zhanglab.net.ConclusionIRIS capitalizes upon PARIS data to improve the prediction of in vivo RNA secondary structure ensembles. We expect that IRIS will enhance the application of the PARIS technology and shed more insight on in vivo RNA secondary structures.