Dataset Information

Identification of pre-microRNAs by characterizing their sequence order evolution information and secondary structure graphs.

ABSTRACT:

Background

Distinction between pre-microRNAs (precursor microRNAs) and length-similar pseudo pre-microRNAs can reveal more about the regulatory mechanism of RNA biological processes. Machine learning techniques have been widely applied to deal with this challenging problem. However, most of them mainly focus on secondary structure information of pre-microRNAs, while ignoring sequence-order information and sequence evolution information.

Results

We use new features for the machine learning algorithms to improve the classification performance by characterizing both sequence order evolution information and secondary structure graphs. We developed three steps to extract these features of pre-microRNAs. We first extract features from PSI-BLAST profiles and Hilbert-Huang transforms, which contain rich sequence evolution information and sequence-order information respectively. We then obtain properties of small molecular networks of pre-microRNAs, which contain refined secondary structure information. These structural features are carefully generated so that they can depict both global and local characteristics of pre-microRNAs. In total, our feature space covers 591 features. The maximum relevance and minimum redundancy (mRMR) feature selection method is adopted before support vector machine (SVM) is applied as our classifier. The constructed classification model is named MicroRNA -NHPred. The performance of MicroRNA -NHPred is high and stable, which is better than that of those state-of-the-art methods, achieving an accuracy of up to 94.83% on same benchmark datasets.

Conclusions

The high prediction accuracy achieved by our proposed method is attributed to the design of a comprehensive feature set on the sequences and secondary structures, which are capable of characterizing the sequence evolution information and sequence-order information, and global and local information of pre-microRNAs secondary structures. MicroRNA -NHPred is a valuable method for pre-microRNAs identification. The source codes of our method can be downloaded from https://github.com/myl446/MicroRNA-NHPred .

SUBMITTER: Ma Y

PROVIDER: S-EPMC6311913 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:BackgroundMicroRNAs (miRNAs) are produced by the sequential processing of a long hairpin RNA transcript by Drosha and Dicer, an RNase III enzymes, and form transitory small RNA duplexes. One strand of the duplex, which incorporates into RNA-induced silencing complex (RISC) and silences the gene expression is called guide strand, or miRNA; while the other strand of duplex is degraded and called the passenger strand, or miRNA*. Predicting the guide strand of miRNA is important for better understanding the RNA interference pathways.ResultsThis paper describes support vector machine (SVM) models developed for predicting the guide strands of miRNAs. All models were trained and tested on a dataset consisting of 329 miRNA and 329 miRNA* pairs using five fold cross validation technique. Firstly, models were developed using mono-, di-, and tri-nucleotide composition of miRNA strands and achieved the highest accuracies of 0.588, 0.638 and 0.596 respectively. Secondly, models were developed using split nucleotide composition and achieved maximum accuracies of 0.553, 0.641 and 0.602 for mono-, di-, and tri-nucleotide respectively. Thirdly, models were developed using binary pattern and achieved the highest accuracy of 0.708. Furthermore, when integrating the secondary structure features with binary pattern, an accuracy of 0.719 was seen. Finally, hybrid models were developed by combining various features and achieved maximum accuracy of 0.799 with sensitivity 0.781 and specificity 0.818. Moreover, the performance of this model was tested on an independent dataset that achieved an accuracy of 0.80. In addition, we also compared the performance of our method with various siRNA-designing methods on miRNA and siRNA datasets.ConclusionIn this study, first time a method has been developed to predict guide miRNA strands, of miRNA duplex. This study demonstrates that guide and passenger strand of miRNA precursors can be distinguished using their nucleotide sequence and secondary structure. This method will be useful in understanding microRNA processing and can be implemented in RNA silencing technology to improve the biological and clinical research. A web server has been developed based on SVM models described in this study (http://crdd.osdd.net:8081/RISCbinder/).

Project description:MotivationThe structure of proteins is organized in a hierarchy among which the secondary structure elements, α-helix, β-strand and loop, are the basic bricks. The determination of secondary structure elements usually requires the knowledge of the whole structure. Nevertheless, in numerous experimental circumstances, the protein structure is partially known. The detection of secondary structures from these partial structures is hampered by the lack of information about connecting residues along the primary sequence.ResultsWe introduce a new methodology to estimate the secondary structure elements from the values of local distances and angles between the protein atoms. Our method uses a message passing neural network, named Sequoia, which allows the automatic prediction of secondary structure elements from the values of local distances and angles between the protein atoms. This neural network takes as input the topology of the given protein graph, where the vertices are protein residues, and the edges are weighted by values of distances and pseudo-dihedral angles generalizing the backbone angles ϕ and ψ. Any pair of residues, independently of its covalent bonds along the primary sequence of the protein, is tagged with this distance and angle information. Sequoia permits the automatic detection of the secondary structure elements, with an F1-score larger than 80% for most of the cases, when α helices and β strands are predicted. In contrast to the approaches classically used in structural biology, such as DSSP, Sequoia is able to capture the variations of geometry at the interface of adjacent secondary structure element. Due to its general modeling frame, Sequoia is able to handle graphs containing only Cα atoms, which is particularly useful on low resolution structural input and in the frame of electron microscopy development.Availability and implementationSequoia source code can be found at https://github.com/Khalife/Sequoia with additional documentation.Supplementary informationSupplementary data are available at Bioinformatics Advances online.

Project description:BACKGROUND: Residue depth allows determining how deeply a given residue is buried, in contrast to the solvent accessibility that differentiates between buried and solvent-exposed residues. When compared with the solvent accessibility, the depth allows studying deep-level structures and functional sites, and formation of the protein folding nucleus. Accurate prediction of residue depth would provide valuable information for fold recognition, prediction of functional sites, and protein design. RESULTS: A new method, RDPred, for the real-value depth prediction from protein sequence is proposed. RDPred combines information extracted from the sequence, PSI-BLAST scoring matrices, and secondary structure predicted with PSIPRED. Three-fold/ten-fold cross validation based tests performed on three independent, low-identity datasets show that the distance based depth (computed using MSMS) predicted by RDPred is characterized by 0.67/0.67, 0.66/0.67, and 0.64/0.65 correlation with the actual depth, by the mean absolute errors equal 0.56/0.56, 0.61/0.60, and 0.58/0.57, and by the mean relative errors equal 17.0%/16.9%, 18.2%/18.1%, and 17.7%/17.6%, respectively. The mean absolute and the mean relative errors are shown to be statistically significantly better when compared with a method recently proposed by Yuan and Wang [Proteins 2008; 70:509-516]. The results show that three-fold cross validation underestimates the variability of the prediction quality when compared with the results based on the ten-fold cross validation. We also show that the hydrophilic and flexible residues are predicted more accurately than hydrophobic and rigid residues. Similarly, the charged residues that include Lys, Glu, Asp, and Arg are the most accurately predicted. Our analysis reveals that evolutionary information encoded using PSSM is characterized by stronger correlation with the depth for hydrophilic amino acids (AAs) and aliphatic AAs when compared with hydrophobic AAs and aromatic AAs. Finally, we show that the secondary structure of coils and strands is useful in depth prediction, in contrast to helices that have relatively uniform distribution over the protein depth. Application of the predicted residue depth to prediction of buried/exposed residues shows consistent improvements in detection rates of both buried and exposed residues when compared with the competing method. Finally, we contrasted the prediction performance among distance based (MSMS and DPX) and volume based (SADIC) depth definitions. We found that the distance based indices are harder to predict due to the more complex nature of the corresponding depth profiles. CONCLUSION: The proposed method, RDPred, provides statistically significantly better predictions of residue depth when compared with the competing method. The predicted depth can be used to provide improved prediction of both buried and exposed residues. The prediction of exposed residues has implications in characterization/prediction of interactions with ligands and other proteins, while the prediction of buried residues could be used in the context of folding predictions and simulations.

Project description:The E(280)/E(260) ratio was found to be suitable for following the ionization of cytosine residues of polynucleotides on the basis of studies with model compounds such as oligoguanylic acid, oligocytidylic acid, a complex formed between polyadenylic acid and polyuridylic acid, and a copolymer of guanylic acid and cytidylic acid, provided that changes in secondary structure were taken into account. The pK of cytosine residues of a polynucleotide in the amorphous form was found to be 4.70 at 25 degrees in 0.1m-sodium phosphate on the basis of titration at 75-85 degrees and on the assumption that the heat of ionization was the same as the value (5.2kcal./mole) found for CMP. In contrast, the pK of cytosine residues in the double-helical form of DNA was found to be about 3.25. These observations were utilized in estimating the fraction of cytosine residues in helical segments of ribosomal RNA, a copolymer of guanylic acid and cytidylic acid, and a copolymer of adenylic acid, guanylic acid, uridylic acid and cytidylic acid. The ionization of guanine and uracil residues was estimated from changes in the E(270)/E(260) ratio and E(230)/E(260) ratio respectively. In the amorphous form of RNA both residues had the same pK, whereas in the double-helical form ionization was suppressed. The fraction of guanine and uracil residues in amorphous segments may be estimated from the titration curves. The difference in the denaturation spectrum of adenine--uracil and guanine--cytosine base pairs at 280mmu was enhanced in acidic solutions whereas E(260) was hardly affected. Hence a comparison of the increments in E(280) and E(260) obtained on increasing the temperature at constant pH may be used to distinguish the melting ranges of helical domains differing in nucleotide composition. In alkaline solutions comparison of the increments in E(260) and E(270) yields similar information. In acidic solutions the fraction of cytosine residues involved in helical secondary structure, the degree of ionization of cytosine residues and the fraction of adenine--uracil base pairs denatured may be estimated from DeltaE(265) and DeltaE(280). In alkaline solutions the fractions of guanine and uracil residues involved in secondary structure and the degrees of ionization of these residues may be estimated from DeltaE(230), DeltaE(245), DeltaE(260) and DeltaE(280).

Project description:Methods of artificial evolution such as SELEX and in vitro selection have made it possible to isolate RNA and DNA motifs with a wide range of functions from large random sequence libraries. Once the primary sequence of a functional motif is known, the sequence space around it can be comprehensively explored using a combination of random mutagenesis and selection. However, methods to explore the sequence space of a secondary structure are not as well characterized. Here we address this question by describing a method to construct libraries in a single synthesis which are enriched for sequences with the potential to form a specific secondary structure, such as that of an aptamer, ribozyme, or deoxyribozyme. Although interactions such as base pairs cannot be encoded in a library using conventional DNA synthesizers, it is possible to modulate the probability that two positions will have the potential to pair by biasing the nucleotide composition at these positions. Here we show how to maximize this probability for each of the possible ways to encode a pair (in this study defined as A-U or U-A or C-G or G-C or G.U or U.G). We then use these optimized coding schemes to calculate the number of different variants of model stems and secondary structures expected to occur in a library for a series of structures in which the number of pairs and the extent of conservation of unpaired positions is systematically varied. Our calculations reveal a tradeoff between maximizing the probability of forming a pair and maximizing the number of possible variants of a desired secondary structure that can occur in the library. They also indicate that the optimal coding strategy for a library depends on the complexity of the motif being characterized. Because this approach provides a simple way to generate libraries enriched for sequences with the potential to form a specific secondary structure, we anticipate that it should be useful for the optimization and structural characterization of functional nucleic acid motifs.