A new approach for detecting riboswitches in DNA sequences

Functional Nucleic Acids Under Unusual Conditions

Functional nucleic acids (FNAs), including naturally occurring ribozymes and riboswitches as well as artificially created DNAzymes and aptamers, have been popular molecular toolboxes for diverse applications. Given the high chemical stability of nucleic acids and their ability to fold into diverse sequence-dependent structures, FNAs are suggested to be highly functional under unusual reaction conditions. This review will examine the progress of research on FNAs under conditions of low pH, high temperature, freezing conditions, and the inclusion of organic solvents and denaturants that are known to disrupt nucleic acid structures. The FNA species to be discussed include ribozymes, riboswitches, G-quadruplex-based peroxidase mimicking DNAzymes, RNA-cleaving DNAzymes, and aptamers. Research within this space has not only revealed the hidden talents of FNAs but has also laid important groundwork for pursuing these intriguing functional macromolecules for unique applications.

Identification and characterisation of thiamine pyrophosphate (TPP) riboswitch in Elaeis guineensis

The oil palm (Elaeis guineensis) is an important crop in Malaysia but its productivity is hampered by various biotic and abiotic stresses. Recent studies suggest the importance of signalling molecules in plants in coping against stresses, which includes thiamine (vitamin B1). Thiamine is an essential microelement that is synthesized de novo by plants and microorganisms. The active form of thiamine, thiamine pyrophosphate (TPP), plays a prominent role in metabolic activities particularly as an enzymatic cofactor. Recently, thiamine biosynthesis pathways in oil palm have been characterised but the search of novel regulatory element known as riboswitch is yet to be done. Previous studies showed that thiamine biosynthesis pathway is regulated by an RNA element known as riboswitch. Riboswitch binds a small molecule, resulting in a change in production of the proteins encoded by the mRNA. TPP binds specifically to TPP riboswitch to regulate thiamine biosynthesis through a variety of mechanisms found in archaea, bacteria and eukaryotes. This study was carried out to hunt for TPP riboswitch in oil palm thiamine biosynthesis gene. Riboswitch detection software like RiboSW, RibEx, Riboswitch Scanner and Denison Riboswitch Detector were utilised in order to locate putative TPP riboswitch in oil palm ThiC gene sequence that encodes for the first enzyme in the pyrimidine branch of the pathway. The analysis revealed a 192 bp putative TPP riboswitch located at the 3' untranslated region (UTR) of the mRNA. Further comparative gene analysis showed that the 92-nucleotide aptamer region, where the metabolite binds was conserved inter-species. The secondary structure analysis was also carried out using Mfold Web server and it showed a stem-loop structure manifested with stems (P1-P5) with minimum free energy of -12.26 kcal/mol. Besides that, the interaction of riboswitch and its ligand was determined using isothermal titration calorimetry (ITC) and it yielded an exothermic reaction with 1:1 stoichiometry interaction with binding affinities of 0.178 nM, at 30°C. To further evaluate the ability of riboswitch to control the pathway, exogenous thiamine was applied to four months old of oil palm seedlings and sampling of spear leaves tissue was carried out at days 0, 1, 2 and 3 post-treatment for expression analysis of ThiC gene fragment via quantitative polymerase chain reaction (qPCR). Results showed an approximately 5-fold decrease in ThiC gene expression upon application of exogenous thiamine. Quantification of thiamine and its derivatives was carried out via HPLC and the results showed that it was correlated to the down regulation of ThiC gene expression. The application of exogenous thiamine to oil palm affected ThiC gene expression, which supported the prediction of the presence of TPP riboswitch in the gene. Overall, this study provides the first evidence on the presence, binding and the functionality of TPP riboswitch in oil palm. This study is hoped to pave a way for better understanding on the regulation of thiamine biosynthesis pathway in oil palm, which can later be exploited for various purposes especially in manipulation of thiamine biosynthesis pathways in combating stresses in oil palm.

Riboflow: Using Deep Learning to Classify Riboswitches With ∼99% Accuracy

Riboswitches are cis-regulatory genetic elements that use an aptamer to control gene expression. Specificity to cognate ligand and diversity of such ligands have expanded the functional repertoire of riboswitches to mediate mounting apt responses to sudden metabolic demands and signal changes in environmental conditions. Given their critical role in microbial life, riboswitch characterisation remains a challenging computational problem. Here we have addressed the issue with advanced deep learning frameworks, namely convolutional neural networks (CNN), and bidirectional recurrent neural networks (RNN) with Long Short-Term Memory (LSTM). Using a comprehensive dataset of 32 ligand classes and a stratified train-validate-test approach, we demonstrated the accurate performance of both the deep learning models (CNN and RNN) relative to conventional hyperparameter-optimized machine learning classifiers on all key performance metrics, including the ROC curve analysis. In particular, the bidirectional LSTM RNN emerged as the best-performing learning method for identifying the ligand-specificity of riboswitches with an accuracy >0.99 and macro-averaged F-score of 0.96. An additional attraction is that the deep learning models do not require prior feature engineering. A dynamic update functionality is built into the models to factor for the constant discovery of new riboswitches, and extend the predictive modeling to new classes. Our work would enable the design of genetic circuits with custom-tuned riboswitch aptamers that would effect precise translational control in synthetic biology. The associated software is available as an open-source Python package and standalone resource for use in genome annotation, synthetic biology, and biotechnology workflows.

A novel riboswitch classification based on imbalanced sequences achieved by machine learning

Riboswitch, a part of regulatory mRNA (50-250nt in length), has two main classes: aptamer and expression platform. One of the main challenges raised during the classification of riboswitch is imbalanced data. That is a circumstance in which the records of a sequences of one group are very small compared to the others. Such circumstances lead classifier to ignore minority group and emphasize on majority ones, which results in a skewed classification. We considered sixteen riboswitch families, to be in accord with recent riboswitch classification work, that contain imbalanced sequences. The sequences were split into training and test set using a newly developed pipeline. From 5460 k-mers (k value 1 to 6) produced, 156 features were calculated based on CfsSubsetEval and BestFirst function found in WEKA 3.8. Statistically tested result was significantly difference between balanced and imbalanced sequences (p < 0.05). Besides, each algorithm also showed a significant difference in sensitivity, specificity, accuracy, and macro F-score when used in both groups (p < 0.05). Several k-mers clustered from heat map were discovered to have biological functions and motifs at the different positions like interior loops, terminal loops and helices. They were validated to have a biological function and some are riboswitch motifs. The analysis has discovered the importance of solving the challenges of majority bias analysis and overfitting. Presented results were generalized evaluation of both balanced and imbalanced models, which implies their ability of classifying, to classify novel riboswitches. The Python source code is available at https://github.com/Seasonsling/riboswitch.

Development of a new oligonucleotide block location-based feature extraction (BLBFE) method for the classification of riboswitches

As knowledge of genetics and genome elements increases, the demand for the development of bioinformatics tools for analyzing these data is raised. Riboswitches are genetic components, usually located in the untranslated regions of mRNAs, that regulate gene expression. Additionally, their interaction with antibiotics has been recently suggested, implying a role in antibiotic effects and resistance. Following a previously published sequential block finding algorithm, herein, we report the development of a new block location-based feature extraction strategy (BLBFE). This procedure utilizes the locations of family-specific sequential blocks on riboswitch sequences as features. Furthermore, the performance of other feature extraction strategies, including mono- and dinucleotide frequencies, k-mer, DAC, DCC, DACC, PC-PseDNC-General and SC-PseDNC-General methods, was investigated. KNN, LDA, naïve Bayes, PNN and decision tree classifiers accompanied by V-fold cross-validation were applied for all methods of feature extraction, and their performances based on the defined feature extraction strategies were compared. Performance measures of accuracy, sensitivity, specificity and F-score for each method of feature extraction were studied. The proposed feature extraction strategy resulted in classification of riboswitches with an average correct classification rate (CCR) of 90.8%. Furthermore, the obtained data confirmed the performance of the developed feature extraction method with an average accuracy of 96.1%, an average sensitivity of 90.8%, an average specificity of 97.52% and an average F-score of 90.69%. Our results implied that the proposed feature extraction (BLBFE) method can classify and discriminate riboswitch families with high CCR, accuracy, sensitivity, specificity and F-score values.

Classification of Riboswitch Families Using Block Location-Based Feature Extraction (BLBFE) Method

Purpose: Riboswitches are special non-coding sequences usually located in mRNAs' un-translated regions and regulate gene expression and consequently cellular function. Furthermore, their interaction with antibiotics has been recently implicated. This raises more interest in development of bioinformatics tools for riboswitch studies. Herein, we describe the development and employment of novel block location-based feature extraction (BLBFE) method for classification of riboswitches. Methods: We have already developed and reported a sequential block finding (SBF) algorithm which, without operating alignment methods, identifies family specific sequential blocks for riboswitch families. Herein, we employed this algorithm for 7 riboswitch families including lysine, cobalamin, glycine, SAM-alpha, SAM-IV, cyclic-di-GMP-I and SAH. Then the study was extended toward implementation of BLBFE method for feature extraction. The outcome features were applied in various classifiers including linear discriminant analysis (LDA), probabilistic neural network (PNN), decision tree and k-nearest neighbors (KNN) classifiers for classification of the riboswitch families. The performance of the classifiers was investigated according to performance measures such as correct classification rate (CCR), accuracy, sensitivity, specificity and f-score. Results: As a result, average CCR for classification of riboswitches was 87.87%. Furthermore, application of BLBFE method in 4 classifiers displayed average accuracies of 93.98% to 96.1%, average sensitivities of 76.76% to 83.61%, average specificities of 96.53% to 97.69% and average f-scores of 74.9% to 81.91%. Conclusion: Our results approved that the proposed method of feature extraction; i.e. BLBFE method; can be successfully used for classification and discrimination of the riboswitch families with high CCR, accuracy, sensitivity, specificity and f-score values.

Large-scale and significant expression from pseudogenes in Sodalis glossinidius - a facultative bacterial endosymbiont

The majority of bacterial genomes have high coding efficiencies, but there are some genomes of intracellular bacteria that have low gene density. The genome of the endosymbiont Sodalis glossinidius contains almost 50 % pseudogenes containing mutations that putatively silence them at the genomic level. We have applied multiple 'omic' strategies, combining Illumina and Pacific Biosciences Single-Molecule Real-Time DNA sequencing and annotation, stranded RNA sequencing and proteome analysis to better understand the transcriptional and translational landscape of Sodalis pseudogenes, and potential mechanisms for their control. Between 53 and 74 % of the Sodalis transcriptome remains active in cell-free culture. The mean sense transcription from coding domain sequences (CDSs) is four times greater than that from pseudogenes. Comparative genomic analysis of six Illumina-sequenced Sodalis isolates from different host Glossina species shows pseudogenes make up ~40 % of the 2729 genes in the core genome, suggesting that they are stable and/or that Sodalis is a recent introduction across the genus Glossina as a facultative symbiont. These data shed further light on the importance of transcriptional and translational control in deciphering host-microbe interactions. The combination of genomics, transcriptomics and proteomics gives a multidimensional perspective for studying prokaryotic genomes with a view to elucidating evolutionary adaptation to novel environmental niches.

Regulatory context drives conservation of glycine riboswitch aptamers

In comparison to protein coding sequences, the impact of mutation and natural selection on the sequence and function of non-coding (ncRNA) genes is not well understood. Many ncRNA genes are narrowly distributed to only a few organisms, and appear to be rapidly evolving. Compared to protein coding sequences, there are many challenges associated with assessment of ncRNAs that are not well addressed by conventional phylogenetic approaches, including: short sequence length, lack of primary sequence conservation, and the importance of secondary structure for biological function. Riboswitches are structured ncRNAs that directly interact with small molecules to regulate gene expression in bacteria. They typically consist of a ligand-binding domain (aptamer) whose folding changes drive changes in gene expression. The glycine riboswitch is among the most well-studied due to the widespread occurrence of a tandem aptamer arrangement (tandem), wherein two homologous aptamers interact with glycine and each other to regulate gene expression. However, a significant proportion of glycine riboswitches are comprised of single aptamers (singleton). Here we use graph clustering to circumvent the limitations of traditional phylogenetic analysis when studying the relationship between the tandem and singleton glycine aptamers. Graph clustering enables a broader range of pairwise comparison measures to be used to assess aptamer similarity. Using this approach, we show that one aptamer of the tandem glycine riboswitch pair is typically much more highly conserved, and that which aptamer is conserved depends on the regulated gene. Furthermore, our analysis also reveals that singleton aptamers are more similar to either the first or second tandem aptamer, again based on the regulated gene. Taken together, our findings suggest that tandem glycine riboswitches degrade into functional singletons, with the regulated gene(s) dictating which glycine-binding aptamer is conserved.

Structural introspection of a putative fluoride transporter in plants

The xenobiotic fluoride is a detrimental toxin which is exported by a double-barrelled fluoride channel (Fluc) in prokaryotes and a single-barrelled fluoride exporter (FEX) in lower eukaryotes. The presence of FEX gene in higher plant species has not been reported till date. In this communication, we have identified FEXs across nine plant species belonging to five different families. Homology modelling predicted the sequential and structural conservation of crucial residues in plant FEXs. It was predicted that the N-terminal segments were the main target sites for phosphorylation. Other post-translational modifications included Arg methylation, Lys acetylation and Cys S-nitrosylation, all of which contribute to protein stability and function. FEXs were predicted to contain lipid moieties which anchor the transporters to the plasma membrane. The crystallizability along with associated properties of plant FEXs was also analyzed to facilitate future experimental analyses. The predicted RNA structure of OsFEX formed several pseudoknots, though no riboswitches could be detected from our prediction. The analysis of upstream regions of FEX in representative plants represented that this gene could be responsive to phytohormone-mediated signaling. This is the first novel report highlighting that FEX genes in plants are probably expressed in a tissue-specific pattern and possibly via a phytohormone-dependent pathway to encode functional FEX proteins during fluoride stress.

Classification of riboswitch sequences using k-mer frequencies

Riboswitches are non-coding RNAs that regulate gene expression by altering the structural conformation of mRNA transcripts. Their regulation mechanism might be exploited for interesting biomedical applications such as drug targets and biosensors. A major challenge consists in accurately identifying metabolite-binding RNA switches which are structurally complex and diverse. In this regard, we investigated the classification of 16 riboswitch families using supervised learning algorithms trained solely with sequence-based features. We generated a reduced feature set and proposed a visual representation to explore its components. We induced Support Vector Machine, Random Forest, Naive Bayes, J48, and HyperPipes classifiers with our proposed feature set and tested their performance over independent data. Our best multi-class classifier achieved F-measure values of 0.996 and 0.966 in the training and test phases, respectively, outperforming those of a previous approach. When compared against BLAST, our best classifiers yielded competitive results. This work shows that the classifiers trained with our sequence-based feature set accurately discriminate riboswitches.

Computational approaches for alternative and transient secondary structures of ribonucleic acids

Transient and alternative structures of ribonucleic acids (RNAs) play essential roles in various regulatory processes, such as translation regulation in living cells. Because experimental analyses for RNA structures are difficult and time-consuming, computational approaches based on RNA secondary structures are promising. In this article, we review computational methods for detecting and analyzing transient/alternative secondary structures of RNAs, including static approaches based on probabilistic distributions of RNA secondary structures and dynamic approaches such as kinetic folding and folding pathway predictions.

An Archaeal Fluoride-Responsive Riboswitch Provides an Inducible Expression System for Hyperthermophiles

Robust genetic systems for the hyperthermophilic Thermococcales have facilitated the overexpression of native genes, enabled the addition of sequences encoding secretion signals, epitope, and affinity tags to coding regions, and aided the introduction of sequences encoding new proteins in these fast-growing fermentative heterotrophs. However, tightly controlled and easily manipulated systems facilitating regulated gene expression are limited for these hosts. Here, we describe an alternative method for regulatory control reliant on a cis-encoded functional riboswitch in the model archaeon Thermococcus kodakarensis Despite the hyperthermophilic growth temperatures, the proposed structure of the riboswitch conforms to a fluoride-responsive riboswitch encoded in many bacteria and similarly functions to regulate a component-conserved fluoride export pathway. Deleting components of the fluoride export pathway generates T. kodakarensis strains with increased fluoride sensitivity. The mechanism underlying regulated expression suggested that the riboswitch-encoding sequences could be utilized as a tunable expression cassette. When appended to a reporter gene, the riboswitch-mediated control system provides fluoride-dependent tunable regulatory potential, offering an alternative system for regulating gene expression. Riboswitch-regulated expression is thus ubiquitous in extant life and can be exploited to generate regulated expression systems for hyperthermophiles.IMPORTANCE Gene expression is controlled by a myriad of interconnected mechanisms that interpret metabolic states and environmental cues to balance cell physiology. Transcription regulation in Archaea is known to employ both typical repressors-operators and transcription activators to regulate transcription initiation in addition to the regulation afforded by chromatin structure. It was perhaps surprising that the presumed ancient mechanism of riboswitch-mediated regulation is found in Bacteria and Eukarya, but seemingly absent in Archaea We demonstrate here that a fluoride-responsive riboswitch functions to regulate a detoxification pathway in the hyperthermophilic archaeon Thermococcus kodakarensis The results obtained define a universal role for riboswitch-mediated regulation, adumbrate the presence of several riboswitch-regulated genes in Thermococcus kodakarensis, demonstrate the utility of RNA-based regulation at high temperatures, and provide a novel riboswitch-regulated expression system to employ in hyperthermophiles.

Using RNA Sequence and Structure for the Prediction of Riboswitch Aptamer: A Comprehensive Review of Available Software and Tools

RNA molecules are essential players in many fundamental biological processes. Prokaryotes and eukaryotes have distinct RNA classes with specific structural features and functional roles. Computational prediction of protein structures is a research field in which high confidence three-dimensional protein models can be proposed based on the sequence alignment between target and templates. However, to date, only a few approaches have been developed for the computational prediction of RNA structures. Similar to proteins, RNA structures may be altered due to the interaction with various ligands, including proteins, other RNAs, and metabolites. A riboswitch is a molecular mechanism, found in the three kingdoms of life, in which the RNA structure is modified by the binding of a metabolite. It can regulate multiple gene expression mechanisms, such as transcription, translation initiation, and mRNA splicing and processing. Due to their nature, these entities also act on the regulation of gene expression and detection of small metabolites and have the potential to helping in the discovery of new classes of antimicrobial agents. In this review, we describe software and web servers currently available for riboswitch aptamer identification and secondary and tertiary structure prediction, including applications.

Application of supervised machine learning algorithms for the classification of regulatory RNA riboswitches

Riboswitches, the small structured RNA elements, were discovered about a decade ago. It has been the subject of intense interest to identify riboswitches, understand their mechanisms of action and use them in genetic engineering. The accumulation of genome and transcriptome sequence data and comparative genomics provide unprecedented opportunities to identify riboswitches in the genome. In the present study, we have evaluated the following six machine learning algorithms for their efficiency to classify riboswitches: J48, BayesNet, Naïve Bayes, Multilayer Perceptron, sequential minimal optimization, hidden Markov model (HMM). For determining effective classifier, the algorithms were compared on the statistical measures of specificity, sensitivity, accuracy, F-measure and receiver operating characteristic (ROC) plot analysis. The classifier Multilayer Perceptron achieved the best performance, with the highest specificity, sensitivity, F-score and accuracy, and with the largest area under the ROC curve, whereas HMM was the poorest performer. At present, the available tools for the prediction and classification of riboswitches are based on covariance model, support vector machine and HMM. The present study determines Multilayer Perceptron as a better classifier for the genome-wide riboswitch searches.

Comparative genomics and phylogenomic analyses of lysine riboswitch distributions in bacteria

Riboswitches are cis-regulatory elements that regulate the expression of genes involved in biosynthesis or transport of a ligand that binds to them. Among the nearly 40 classes of riboswitches discovered so far, three are known to regulate the concentration of biologically encoded amino acids glycine, lysine, and glutamine. While some comparative genomics studies of riboswitches focusing on their gross distribution across different bacterial taxa have been carried out recently, systematic functional annotation and analysis of lysine riboswitches and the genes they regulate are still lacking. We analyzed 2785 complete bacterial genome sequences to systematically identify 468 lysine riboswitches (not counting hits from multiple strains of the same species) and obtain a detailed phylogenomic map of gene-specific lysine riboswitch distribution across diverse prokaryotic phyla. We find that lysine riboswitches are most abundant in Firmicutes and Gammaproteobacteria where they are found upstream to both biosynthesis and/or transporter genes. They are relatively rare in all other prokaryotic phyla where if present they are primarily found upstream to operons containing many lysine biosynthesis genes. The genome-wide study of the genetic organisation of the lysine riboswitches show considerable variation both within and across different Firmicute orders. Correlating the location of a riboswitch with its genomic context and its phylogenetic relationship with other evolutionarily related riboswitch carrying species, enables identification and annotation of many lysine biosynthesis, transporter and catabolic genes. It also reveals previously unknown patterns of lysine riboswitch distribution and gene/operon regulation and allows us to draw inferences about the possible point of origin of lysine riboswitches. Additionally, evidence of horizontal transfer of riboswitches was found between Firmicutes and Actinobacteria. Our analysis provides a useful resource that will lead to a better understanding of the evolution of these regulatory elements and prove to be beneficial for exploiting riboswitches for developing targeted therapies.

Riboswitches: From living biosensors to novel targets of antibiotics

Riboswitches are generally located in 5'-UTR region of mRNAs and specifically bind small ligands. Following ligand binding, gene expression is controlled mostly by transcription termination, translation inhibition or mRNA degradation processes. More than 30 classes of known riboswitches have been identified by now. Most riboswitches consist of an aptamer domain and an expression platform. The aptamer domain of each class of riboswitch is a conserved structure and stabilizes specific structures of the expression platforms through binding to specific compounds. In this review, we are highlighting most aspects of riboswitch research including the novel riboswitch discoveries, routine methods for discovering and investigating riboswitches along with newly discovered classes and mechanistic principles of riboswitch-mediated gene expression control. Moreover, we will give an overview about the potential of riboswitches as therapeutic targets for antibiotic design and also their utilization as biosensors for molecular detection.

Riboswitch Scanner: an efficient pHMM-based web-server to detect riboswitches in genomic sequences

Riboswitches are non-coding RNA located in the 5' untranslated regions where they bind a target metabolite used to specify the riboswitch class and control the expression of associated genes. Accurate identification of riboswitches is the first step towards understanding their regulatory and functional roles in the cell. In this article, we describe a new web application named Riboswitch Scanner which provides an automated pipeline for pHMM-based detection of riboswitches in partial as well as complete genomic sequences rapidly, with high sensitivity and specificity.

AVAILABILITY AND IMPLEMENTATION

Riboswitch Scanner can be freely accessed on the web at http://service.iiserkol.ac.in/∼riboscan/

CONTACT

mukherjee.sumit89@gmail.com

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Multilevel Regulation and Translational Switches in Synthetic Biology

In contrast to the versatility of regulatory mechanisms in natural systems, synthetic genetic circuits have been so far predominantly composed of transcriptionally regulated modules. This is about to change as the repertoire of foundational tools for post-transcriptional regulation is quickly expanding. We provide an overview of the different types of translational regulators: protein, small molecule and ribonucleic acid (RNA) responsive and we describe the new emerging circuit designs utilizing these tools. There are several advantages of achieving multilevel regulation via translational switches and it is likely that such designs will have the greatest and earliest impact in mammalian synthetic biology for regenerative medicine and gene therapy applications.

RNAPattMatch: a web server for RNA sequence/structure motif detection based on pattern matching with flexible gaps

Searching for RNA sequence-structure patterns is becoming an essential tool for RNA practitioners. Novel discoveries of regulatory non-coding RNAs in targeted organisms and the motivation to find them across a wide range of organisms have prompted the use of computational RNA pattern matching as an enhancement to sequence similarity. State-of-the-art programs differ by the flexibility of patterns allowed as queries and by their simplicity of use. In particular-no existing method is available as a user-friendly web server. A general program that searches for RNA sequence-structure patterns is RNA Structator. However, it is not available as a web server and does not provide the option to allow flexible gap pattern representation with an upper bound of the gap length being specified at any position in the sequence. Here, we introduce RNAPattMatch, a web-based application that is user friendly and makes sequence/structure RNA queries accessible to practitioners of various background and proficiency. It also extends RNA Structator and allows a more flexible variable gaps representation, in addition to analysis of results using energy minimization methods. RNAPattMatch service is available at http://www.cs.bgu.ac.il/rnapattmatch. A standalone version of the search tool is also available to download at the site.

Computational prediction of riboswitches

Riboswitches present a ubiquitous genetic regulatory mechanism for prokaryotes and have been found in HIV1, fungi, plants, and even H. sapiens. We present an overview of approaches to predict riboswitch aptamers and, more generally, RNA conformational switches.