Computational prediction of microRNA genes

Machine Learning Strategies in MicroRNA Research: Bridging Genome to Phenome

MicroRNAs (miRNAs) have emerged as a prominent layer of regulation of gene expression. This article offers the salient and current aspects of machine learning (ML) tools and approaches from genome to phenome in miRNA research. First, we underline that the complexity in the analysis of miRNA function ranges from their modes of biogenesis to the target diversity in diverse biological conditions. Therefore, it is imperative to first ascertain the miRNA coding potential of genomes and understand the regulatory mechanisms of their expression. This knowledge enables the efficient classification of miRNA precursors and the identification of their mature forms and respective target genes. Second, and because one miRNA can target multiple mRNAs and vice versa, another challenge is the assessment of the miRNA-mRNA target interaction network. Furthermore, long-noncoding RNA (lncRNA)and circular RNAs (circRNAs) also contribute to this complexity. ML has been used to tackle these challenges at the high-dimensional data level. The present expert review covers more than 100 tools adopting various ML approaches pertaining to, for example, (1) miRNA promoter prediction, (2) precursor classification, (3) mature miRNA prediction, (4) miRNA target prediction, (5) miRNA- lncRNA and miRNA-circRNA interactions, (6) miRNA-mRNA expression profiling, (7) miRNA regulatory module detection, (8) miRNA-disease association, and (9) miRNA essentiality prediction. Taken together, we unpack, critically examine, and highlight the cutting-edge synergy of ML approaches and miRNA research so as to develop a dynamic and microlevel understanding of human health and diseases.

MicroRNA Targeting

MicroRNAs (miRNAs) are small noncoding elements that play essential roles in the posttranscriptional regulation of biochemical processes. miRNAs recognize and target multiple mRNAs; therefore, investigating miRNA dysregulation is an indispensable strategy to understand pathological conditions and to design innovative drugs. Targeting miRNAs in diseases improve outcomes of several therapeutic strategies thus, this present study highlights miRNA targeting methods through experimental assays and bioinformatics tools. The first part of this review focuses on experimental miRNA targeting approaches for elucidating key biochemical pathways. A growing body of evidence about the miRNA world reveals the fact that it is not possible to uncover these molecules' structural and functional characteristics related to the biological processes with a deterministic approach. Instead, a systemic point of view is needed to truly understand the facts behind the natural complexity of interactions and regulations that miRNA regulations present. This task heavily depends both on computational and experimental capabilities. Fortunately, several miRNA bioinformatics tools catering to nonexperts are available as complementary wet-lab approaches. For this purpose, this work provides recent research and information about computational tools for miRNA targeting research.

A ceRNA network mediated by LINC00475 in papillary thyroid carcinoma

Papillary thyroid carcinoma (PTC) is the most frequent histological type of differentiated thyroid carcinoma. Long noncoding RNAs (lncRNAs) have been widely reported to play a key role in human malignancies, and PTC is included. This study aimed to find out the functions and mechanism of lncRNA LINC00475 in PTC. LINC00475 was upregulated in PTC cells and was mainly located in the cytoplasm according to reverse-transcription polymerase chain reaction analyses and subcellular fractionation assays. As shown by cell counting kit-8 assays, ethynyl deoxyuridine incorporation assays, wound healing assays, and transwell assays, LINC00475 knockdown suppressed cell viability, proliferation, migration, and invasion. Mechanistically, LINC00475 upregulated the expression of messenger RNA zinc finger CCHC-type containing 12 (ZCCHC12) by binding to miR-376c-3p. ZCCHC12 was a direct target gene of miR-376c-3p in PTC cells. The relationship between miR-376c-3p and LINC00475 (or ZCCHC12) in PTC cells was probed by luciferase reporter assays, RNA pulldown assays, and RNA immunoprecipitation assays. In addition, both mRNA and protein levels of ZCCHC12 were downregulated due to miR-376c-3p overexpression or LINC00475 silencing. ZCCHC12 overexpression partially reversed the suppressive effect of LINC00475 knockdown on malignant behaviors of PTC cells. In conclusion, LINC00475 promotes PTC cell proliferation, migration, and invasion by upregulating ZCCHC12 via the interaction with miR-376c-3p.

Popular Computational Tools Used for miRNA Prediction and Their Future Development Prospects

MicroRNAs (miRNAs) are 19-24 nucleotide (nt)-long noncoding, single-stranded RNA molecules that play significant roles in regulating the gene expression, growth, and development of plants and animals. From the year that miRNAs were first discovered until the beginning of the twenty-first century, researchers used experimental methods such as cloning and sequencing to identify new miRNAs and their roles in the posttranscriptional regulation of protein synthesis. Later, in the early 2000s, informatics approaches to the discovery of new miRNAs began to be implemented. With increasing knowledge about miRNA, more efficient algorithms have been developed for computational miRNA prediction. The miRNA research community, hoping for greater coverage and faster results, has shifted from cumbersome and expensive traditional experimental approaches to computational approaches. These computational methods started with homology-based comparisons of known miRNAs with orthologs in the genomes of other species; this method could identify a known miRNA in new species. Second-generation sequencing and next-generation sequencing of mRNA at different developmental stages and in specific tissues, in combination with a better search and alignment algorithm, have accelerated the process of predicting novel miRNAs in a particular species. Using the accumulated annotated miRNA sequence information, researchers have been able to design ab initio algorithms for miRNA prediction independent of genome sequence knowledge. Here, the methods recently used for miRNA computational prediction are summarized and classified into the following four categories: homology-based, target-based, scoring-based, and machine-learning-based approaches. Finally, the future developmental directions of miRNA prediction methods are discussed.

Predicting novel microRNA: a comprehensive comparison of machine learning approaches

MOTIVATION

The importance of microRNAs (miRNAs) is widely recognized in the community nowadays because these short segments of RNA can play several roles in almost all biological processes. The computational prediction of novel miRNAs involves training a classifier for identifying sequences having the highest chance of being precursors of miRNAs (pre-miRNAs). The big issue with this task is that well-known pre-miRNAs are usually few in comparison with the hundreds of thousands of candidate sequences in a genome, which results in high class imbalance. This imbalance has a strong influence on most standard classifiers, and if not properly addressed in the model and the experiments, not only performance reported can be completely unrealistic but also the classifier will not be able to work properly for pre-miRNA prediction. Besides, another important issue is that for most of the machine learning (ML) approaches already used (supervised methods), it is necessary to have both positive and negative examples. The selection of positive examples is straightforward (well-known pre-miRNAs). However, it is difficult to build a representative set of negative examples because they should be sequences with hairpin structure that do not contain a pre-miRNA.

RESULTS

This review provides a comprehensive study and comparative assessment of methods from these two ML approaches for dealing with the prediction of novel pre-miRNAs: supervised and unsupervised training. We present and analyze the ML proposals that have appeared during the past 10 years in literature. They have been compared in several prediction tasks involving two model genomes and increasing imbalance levels. This work provides a review of existing ML approaches for pre-miRNA prediction and fair comparisons of the classifiers with same features and data sets, instead of just a revision of published software tools. The results and the discussion can help the community to select the most adequate bioinformatics approach according to the prediction task at hand. The comparative results obtained suggest that from low to mid-imbalance levels between classes, supervised methods can be the best. However, at very high imbalance levels, closer to real case scenarios, models including unsupervised and deep learning can provide better performance.

Construction of Multi-Potent MicroRNA Sponge and Its Functional Evaluation

MicroRNA is a small, endogenous RNA that inhibits specific gene expression by interacting mostly on the UTR region of target mRNA. Among various methods to inhibit the microRNA, microRNA sponge is a construct designed to inhibit specific microRNA by providing excess amount of target mRNA. Here, we describe a method to generate multi-potent miRNA sponge which can inhibit multiple microRNAs simultaneously. In addition, two methods to examine the functionality of miRNA sponge will be introduced. This tool can be used to stably inhibit multiple microRNAs either in cell or in vivo, expanding the scope of functional analysis of specific microRNAs.

A compilation of Web-based research tools for miRNA analysis

Since the discovery of microRNAs (miRNAs), a class of noncoding RNAs that regulate the gene expression posttranscriptionally in sequence-specific manner, there has been a release of number of tools useful for both basic and advanced applications. This is because of the significance of miRNAs in many pathophysiological conditions including cancer. Numerous bioinformatics tools that have been developed for miRNA analysis have their utility for detection, expression, function, target prediction and many other related features. This review provides a comprehensive assessment of web-based tools for the miRNA analysis that does not require prior knowledge of any computing languages.

Identification and characterization of microRNAs and their target genes from Nile tilapia (Oreochromis niloticus)

MicroRNAs (miRNAs) are a class of small single-stranded, endogenous 21–22 nt non-coding RNAs that regulate their target mRNA levels by causing either inactivation or degradation of the mRNAs. In recent years, miRNA genes have been identified from mammals, insects, worms, plants, and viruses. In this research, bioinformatics approaches were used to predict potential miRNAs and their targets in Nile tilapia from the expressed sequence tag (EST) and genomic survey sequence (GSS) database, respectively, based on the conservation of miRNAs in many animal species. A total of 19 potential miRNAs were detected following a range of strict filtering criteria. To test the validity of the bioinformatics method, seven predicted Nile tilapia miRNA genes were selected for further biological validation, and their mature miRNA transcripts were successfully detected by stem–loop RT-PCR experiments. Using these potential miRNAs, we found 56 potential targets in this species. Most of the target mRNAs appear to be involved in development, metabolism, signal transduction, transcription regulation and stress responses. Overall, our findings will provide an important foundation for further research on miRNAs function in the Nile tilapia.

Identification and differential expression of microRNAs associated with fat deposition in the liver of Wistar rats with nonalcoholic fatty liver disease

The exact mechanism underlying hepatic steatosis in nonalcoholic fatty liver disease (NAFLD) is not clear. Clarifying the full repertoire of microRNAs (miRNAs) in NAFLD rat liver would enhance our understanding of NAFLD pathogenesis. In this study, miRNA expression levels were analyzed in liver tissue from NAFLD Wistar rats, with normal Wistar rats as negative controls. Small RNA libraries were constructed for each sample. A total of 173 conservative miRNAs and 68 potential miRNA candidates were identified. Significant differences in the expression levels of 101 conserved miRNAs were identified between the two groups. The results of GO annotation and KEGG pathway analysis revealed that some miRNAs were likely involved in the process of liver fat deposition. This study represents the first global miRNA profiling of NAFLD Wistar rat livers, and expands the miRNA repertoire for normal livers. Our findings suggest that miRNAs play important roles in liver fat deposition.

Computational Analysis of LncRNA from cDNA Sequences

Based on recent findings, long noncoding (lnc) RNAs represent a potential class of functional molecules within the cell. In this chapter we describe a computational scheme to identify and classify lncRNAs within maize from full-length cDNA sequences to designate subsets of lncRNAs for which biogenesis and regulatory mechanisms may be verified at the bench. We make use of the Coding Potential Calculator and specific Python scripts in our approach.

MatPred: Computational Identification of Mature MicroRNAs within Novel Pre-MicroRNAs

Background. MicroRNAs (miRNAs) are short noncoding RNAs integral for regulating gene expression at the posttranscriptional level. However, experimental methods often fall short in finding miRNAs expressed at low levels or in specific tissues. While several computational methods have been developed for predicting the localization of mature miRNAs within the precursor transcript, the prediction accuracy requires significant improvement. Methodology/Principal Findings. Here, we present MatPred, which predicts mature miRNA candidates within novel pre-miRNA transcripts. In addition to the relative locus of the mature miRNA within the pre-miRNA hairpin loop and minimum free energy, we innovatively integrated features that describe the nucleotide-specific RNA secondary structure characteristics. In total, 94 features were extracted from the mature miRNA loci and flanking regions. The model was trained based on a radial basis function kernel/support vector machine (RBF/SVM). Our method can predict precise locations of mature miRNAs, as affirmed by experimentally verified human pre-miRNAs or pre-miRNAs candidates, thus achieving a significant advantage over existing methods. Conclusions. MatPred is a highly effective method for identifying mature miRNAs within novel pre-miRNA transcripts. Our model significantly outperformed three other widely used existing methods. Such processing prediction methods may provide important insight into miRNA biogenesis.

RNA 3D Modules in Genome-Wide Predictions of RNA 2D Structure

Recent experimental and computational progress has revealed a large potential for RNA structure in the genome. This has been driven by computational strategies that exploit multiple genomes of related organisms to identify common sequences and secondary structures. However, these computational approaches have two main challenges: they are computationally expensive and they have a relatively high false discovery rate (FDR). Simultaneously, RNA 3D structure analysis has revealed modules composed of non-canonical base pairs which occur in non-homologous positions, apparently by independent evolution. These modules can, for example, occur inside structural elements which in RNA 2D predictions appear as internal loops. Hence one question is if the use of such RNA 3D information can improve the prediction accuracy of RNA secondary structure at a genome-wide level. Here, we use RNAz in combination with 3D module prediction tools and apply them on a 13-way vertebrate sequence-based alignment. We find that RNA 3D modules predicted by metaRNAmodules and JAR3D are significantly enriched in the screened windows compared to their shuffled counterparts. The initially estimated FDR of 47.0% is lowered to below 25% when certain 3D module predictions are present in the window of the 2D prediction. We discuss the implications and prospects for further development of computational strategies for detection of RNA 2D structure in genomic sequence.

Principles of miRNA-mRNA interactions: beyond sequence complementarity

MicroRNAs (miRNAs) are small non-coding RNAs that post-transcriptionally regulate gene expression by altering the translation efficiency and/or stability of targeted mRNAs. In vertebrates, more than 50% of all protein-coding RNAs are assumed to be subject to miRNA-mediated control, but current high-throughput methods that reliably measure miRNA-mRNA interactions either require prior knowledge of target mRNAs or elaborate preparation procedures. Consequently, experimentally validated interactions are relatively rare. Furthermore, in silico prediction based on sequence complementarity of miRNAs and their corresponding target sites suffers from extremely high false positive rates. Apparently, sequence complementarity alone is often insufficient to reflect the complex post-transcriptional regulation of mRNAs by miRNAs, which is especially true for animals. Therefore, combined analysis of small non-coding and protein-coding RNAs is indispensable to better understand and predict the complex dynamics of miRNA-regulated gene expression. Single-nucleotide polymorphisms (SNPs) and alternative polyadenylation (APA) can affect miRNA binding of a given transcript from different individuals and tissues, and especially APA is currently emerging as a major factor that contributes to variations in miRNA-mRNA interplay in animals. In this review, we focus on the influence of APA and SNPs on miRNA-mediated gene regulation and discuss the computational approaches that take these mechanisms into account.