TargetMiner: microRNA target prediction with systematic identification of tissue-specific negative examples

DeepMirTar: a deep-learning approach for predicting human miRNA targets

Motivation

MicroRNAs (miRNAs) are small non-coding RNAs that function in RNA silencing and post-transcriptional regulation of gene expression by targeting messenger RNAs (mRNAs). Because the underlying mechanisms associated with miRNA binding to mRNA are not fully understood, a major challenge of miRNA studies involves the identification of miRNA-target sites on mRNA. In silico prediction of miRNA-target sites can expedite costly and time-consuming experimental work by providing the most promising miRNA-target-site candidates.

Results

In this study, we reported the design and implementation of DeepMirTar, a deep-learning-based approach for accurately predicting human miRNA targets at the site level. The predicted miRNA-target sites are those having canonical or non-canonical seed, and features, including high-level expert-designed, low-level expert-designed and raw-data-level, were used to represent the miRNA-target site. Comparison with other state-of-the-art machine-learning methods and existing miRNA-target-prediction tools indicated that DeepMirTar improved overall predictive performance.

Availability and implementation

DeepMirTar is freely available at https://github.com/Bjoux2/DeepMirTar_SdA.

Supplementary information

Supplementary data are available at Bioinformatics online.

miRBaseMiner, a tool for investigating miRBase content

microRNAs are small non-coding RNA molecules playing a central role in gene regulation. miRBase is the standard reference source for analysis and interpretation of experimental studies. However, the richness and complexity of the annotation is often underappreciated by users. Moreover, even for experienced users, the size of the resource can make it difficult to explore annotation to determine features such as species coverage, the impact of specific characteristics and changes between successive releases. A further consideration is that each new miRBase release contains entries that have had limited review and which may subsequently be removed in a future release to ensure the quality of annotation. To aid the miRBase user, we developed a software tool, miRBaseMiner, for investigating miRBase annotation and generating custom annotation sets. We apply the tool to characterize each release from v9.2 to v22 to examine how annotation has changed across releases and highlight some of the annotation features that users should keep in mind when using for miRBase for data analysis. These include: (1) entries with identical or very similar sequences; (2) entries with multiple annotated genome locations; (3) hairpin precursor entries with extremely low-estimated minimum free energy; (4) entries possessing reverse complementary; (5) entries with 3ʹ poly(A) ends. As each of these factors can impact the identification of dysregulated features and subsequent clinical or biological conclusions, miRBaseMiner is a valuable resource for any user using miRBase as a reference source.

mirDIP 4.1-integrative database of human microRNA target predictions

MicroRNAs are important regulators of gene expression, achieved by binding to the gene to be regulated. Even with modern high-throughput technologies, it is laborious and expensive to detect all possible microRNA targets. For this reason, several computational microRNA-target prediction tools have been developed, each with its own strengths and limitations. Integration of different tools has been a successful approach to minimize the shortcomings of individual databases. Here, we present mirDIP v4.1, providing nearly 152 million human microRNA-target predictions, which were collected across 30 different resources. We also introduce an integrative score, which was statistically inferred from the obtained predictions, and was assigned to each unique microRNA-target interaction to provide a unified measure of confidence. We demonstrate that integrating predictions across multiple resources does not cumulate prediction bias toward biological processes or pathways. mirDIP v4.1 is freely available at http://ophid.utoronto.ca/mirDIP/.

miR-27a-3p regulates proliferation and apoptosis of colon cancer cells by potentially targeting BTG1

microRNA (miR/miRNA)-27a-3p has been reported to be abnormally expressed in various types of cancer, including colorectal cancer (CRC). B-cell translocation gene 1 (BTG1) has also been implicated with CRC. However, the association between miR-27a-3p and BTG1 in CRC, to the best of our knowledge, has not been investigated. In order to assess whether miR-27a-3p is associated with CRC, reverse transcription-quantitative PCR was performed on 20 paired CRC and paracancerous tissues for miRNA analysis. For the screening and validation of miR-27a-3p expression in colon cancer, several colon cancer cell lines (HCT-116, HCT8, SW480, HT29, LOVO and Caco2) and the normal colorectal epithelial cell line NCM460 were examined. The highest expression levels of miR-27a-3p were detected in the HCT-116, which was selected for further experimentation. The HCT-116 cells were divided into control, miR-27a-3p mimic and inhibitor groups, and cell proliferation was tested using an MTT assay. Additionally, miR-27a-3p inhibitor/mimic or BTG1 plasmid were transfected into the HCT-116 cells, and flow cytometry was performed to analyze cell cycle distributions. TUNEL analysis was performed to detect apoptosis. Protein levels of factors in the downstream signaling pathway mediated by miR-27a-3p [ERK/mitogen-activated extracellular signal-regulated kinase (MEK)] were detected. miR-27a-3p was revealed to be overexpressed in human CRC tissues and colon cancer cell lines. Knockdown of miR-27a-3p suppressed proliferation of HCT-116 cells and apoptosis was increased. It further markedly upregulated expression levels of BTG1 and inhibited activation of proteins of the ERK/MEK signaling pathway. In addition, overexpression of BTG1 in HCT-116 cells triggered G₁/S phase cell cycle arrest and increased apoptosis via the ERK/MEK signaling pathway. In conclusion, the present study demonstrated that the effects of miR-27a-3p on colon cancer cell proliferation and apoptosis were similar to those of the tumor suppressor gene BTG1. The miR-27a-3p/BTG1 axis may have potential implications for diagnostic and therapeutic approaches in CRC.

miR-192/215-5p act as tumor suppressors and link Crohn's disease and colorectal cancer by targeting common metabolic pathways: An integrated informatics analysis and experimental study

MicroRNAs have emerged as key regulators involved in a variety of biological processes. Previous studies have demonstrated that miR-192/215 participated in progression of Crohn's disease and colorectal cancer. However, their concrete relationships and regulation networks in diseases remain unclear. Here, we used bioinformatics methods to expound miR-192/215-5p macrocontrol regulatory networks shared by two diseases. For data mining and figure generation, several miRNA prediction tools, Human miRNA tissue atlas, FunRich, miRcancer, MalaCards, STRING, GEPIA, cBioPortal, GEO databases, Pathvisio, Graphpad Prism 6 software, etc . are extensively applied. miR-192/215-5p were specially distributed in colon tissues and enriched biological pathways were closely associated with human cancers. Emerging role of miR-192/215-5p and their common pathways in Crohn's disease and colorectal cancer was also analyzed. Based on results derived from multiple approaches, we identified the biological functions of miR-192/215-5p as a tumor suppressor and link Crohn's disease and colorectal cancer by targeting triglyceride synthesis and extracellular matrix remodeling pathways.

micro-RNAs dependent regulation of DNMT and HIF1α gene expression in thrombotic disorders

MicroRNAs (miRNAs) are involved in a wide variety of cellular processes and post-transcriptionally regulate several mechanism and diseases. However, contribution of miRNAs functioning during hypoxia and DNA methylation together is less understood. The current study was aimed to find a shared miRNAs signature upstream to hypoxia (via HIF gene family members) and methylation (via DNMT gene family members). This was followed by the global validation of the hypoxia related miRNA signature using miRNA microarray meta-analysis of the hypoxia induced human samples. We further concluded the study by looking into thrombosis related terms and pathways enriched during protein-protein interaction (PPI) network analysis of these two sets of gene family. Network prioritization of these shared miRNAs reveals miR-129, miR-19band miR-23b as top regulatory miRNAs. A comprehensive meta-analysis of microarray datasets of hypoxia samples revealed 29 differentially expressed miRNAs. GSEA of the interacting genes in the DNMT-HIF PPI network indicated thrombosis associated pathways including “Hemostasis”, “TPO signaling pathway” and “angiogenesis”. Interestingly, the study has generated a novel database of candidate miRNA signatures shared between hypoxia and methylation, and their relation to thrombotic pathways, which might aid in the development of potential therapeutic biomarkers.

Significant improvement of miRNA target prediction accuracy in large datasets using meta-strategy based on comprehensive voting and artificial neural networks

BACKGROUND

Identifying mRNA targets of miRNAs is critical for studying gene expression regulation at the whole-genome level. Multiple computational tools have been developed to predict miRNA:mRNA interactions. Nonetheless, many of these tools are developed in various small datasets, which each represent a limited sample space. Thus, the prediction accuracy of these tools has not been systematically validated at a larger scale. Accordingly, comparing the prediction accuracy of these tools and determining their applicability become challenging. In addition, the accuracy of these tools, especially in large datasets, needs to be improved for broader applications.

RESULTS

In this project, a large dataset containing more than 46,600 miRNA:mRNA interactions was assembled and split into eleven subsets based on the availability of prediction scores of four individual predictors, which are miRanda, miRDB, PITA, and TargetScan. In each of these subsets, the predictive results of four individual predictors were integrated using decision-tree based artificial neural networks to make the meta-prediction. The decision-tree is used here to sort the predictive results of four individual predictors, and artificial neural networks are applied to make meta-prediction based on the outputs of individual predictors. In the decision tree, dual-threshold and two-step significance-voting were incorporated, information gain was analysed to select threshold values. The prediction performance of this new strategy was improved significantly in most of the eleven datasets comparing to the individual predictors and other meta-predictors, such as ComiR, under multi-fold cross-validation, as well as in independent datasets. The overall improvement of prediction accuracy in independent datasets is at least 9 percentile points comparing to the other predictors, and the percentage of improvement of F1 and MCC scores is at least 40% compared to the other predictors.

CONCLUSIONS

The combination of dual-threshold, two-step significance-voting, and analysis of information gain is very effective in optimizing the outcome of decision-tree, and further integration with artificial neural networks is critical for further improving the performance of meta-predictor. A new pipeline based on this integration for miRNA target prediction has been developed. A strategy using outputs of individual predictors to reorganize large-scale miRNA:mRNA interaction dataset has also been validated and used to evaluate the prediction accuracy of predictors. The predictor is available at: https://github.com/xueLab/mirTarDANN ).

Identification and characterization of skin color microRNAs in Koi carp (Cyprinus carpio L.) by Illumina sequencing

Background

MicroRNAs (miRNAs) are endogenous, small (21–25 nucleotide), non-coding RNAs that play important roles in numerous biological processes. Koi carp exhibit diverse color patterns, making it an ideal subject for studying the genetics of pigmentation. However, the influence of miRNAs on skin color regulation and variation in Koi carp is poorly understood.

Results

Herein, we performed small RNA (sRNA) analysis of the three main skin colors in Koi carp by Illumina sequencing. The results revealed 330, 397, and 335 conserved miRNAs (belonging to 81 families) and 340, 353, and 351 candidate miRNAs in black, red, and white libraries, respectively. A total of 164 differentially expressed miRNAs (DEMs) and 14 overlapping DEMs were identified, including miR-196a, miR-125b, miR-202, miR-205-5p, miR-200b, and etc. Target prediction and functional analysis of color-related miRNAs such as miR-200b, miR-206, and miR-196a highlighted putative target genes, including Mitf, Mc1r, Foxd3, and Sox10 that are potentially related to pigmentation. Determination of reference miRNAs for relative quantification showed that let-7a was the most abundant single reference gene, and let-7a and miR-26b was the most abundant combination.

Conclusions

The findings provide novel insight into the molecular mechanisms determining skin color differentiation in Koi carp, and serve as a valuable reference for future studies on tissue-specific miRNA abundance in Koi carp.

Electronic supplementary material

The online version of this article (10.1186/s12864-018-5189-5) contains supplementary material, which is available to authorized users.

Combining Supervised and Unsupervised Learning for Improved miRNA Target Prediction

MicroRNAs (miRNAs) are short non-coding RNAs which bind to mRNAs and regulate their expression. MiRNAs have been found to be associated with initiation and progression of many complex diseases. Investigating miRNAs and their targets can thus help develop new therapies by designing anti-miRNA oligonucleotides. While existing computational approaches can predict miRNA targets, these predictions have low accuracy. In this paper, we propose a two-step approach to refine the results of sequence-based prediction algorithms. The first step, which is based on our previous work, uses an ensemble learning approach that combines multiple existing methods. The second step utilizes support vector machine (SVM) classifiers in one- and two-class modes to infer miRNA-mRNA interactions based on both binding features, as well as network features extracted from gene regulatory network. Experimental results using two real data sets from TCGA indicate that the use of two-class SVM classification significantly improves the precision of miRNA-mRNA prediction.

Bioinformatics-based interaction analysis of miR-92a-3p and key genes in tamoxifen-resistant breast cancer cells

The abnormal expression of miR-92a-3p was detected in multiple cancers. However, the biological role and underlying mechanism of miR-92a-3p in tamoxifen-resistant cells are still unknown. The main objective of our study was to find potential miR-92a-3p regulating pathways involved in tamoxifen resistance and to construct their regulatory network using bioinformatics. Four gene expression profiles were retrieved from GEO database and the GEO2R tool was used for analysis. GSE41922 and GSE42072 were applied to investigate aberrant miR-92a-3p expression in breast cancer serum and tissue. We found that miR-92a-3p expression was higher in breast cancer serum or tissue than in healthy volunteer serum or adjacent normal tissue, and high expression of miR-92a-3p could predict poor prognosis of breast cancer patients. In our qRT-PCR validation, we found that miR-92a-3p was upregulated in tamoxifen-resistant cells. MiR-92a-3p might play a role in tamoxifen resistance. In order to find the relationship between miR-92a-3p and some key genes and their potential molecular mechanisms in tamoxifen-resistant cells. The microarray data GSE26459 and GSE28267 were analyzed to determine the differentially expressed genes (DEGs) or miRNAs (DEMs). Furthermore, the related long non-coding RNAs (lncRNAs) were screened with starBase v2.0. Finally,microRNA.org,miRDB, targetminer and targetscan were applied to predict the targets of miR-92a-3p. Through analysis, we find that miR-92a-3p may be used as a potential biomarker for early detection of cancer and monitoring the efficacy of endocrine therapy.

A Distributed Classifier for MicroRNA Target Prediction with Validation Through TCGA Expression Data

BACKGROUND

MicroRNAs (miRNAs) are approximately 22-nucleotide long regulatory RNA that mediate RNA interference by binding to cognate mRNA target regions. Here, we present a distributed kernel SVM-based binary classification scheme to predict miRNA targets. It captures the spatial profile of miRNA-mRNA interactions via smooth B-spline curves. This is accomplished separately for various input features, such as thermodynamic and sequence-based features. Further, we use a principled approach to uniformly model both canonical and non-canonical seed matches, using a novel seed enrichment metric. Finally, we verify our miRNA-mRNA pairings using an Elastic Net-based regression model on TCGA expression data for four cancer types to estimate the miRNAs that together regulate any given mRNA.

RESULTS

We present a suite of algorithms for miRNA target prediction, under the banner Avishkar, with superior prediction performance over the competition. Specifically, our final kernel SVM model, with an Apache Spark backend, achieves an average true positive rate (TPR) of more than 75 percent, when keeping the false positive rate of 20 percent, for non-canonical human miRNA target sites. This is an improvement of over 150 percent in the TPR for non-canonical sites, over the best-in-class algorithm. We are able to achieve such superior performance by representing the thermodynamic and sequence profiles of miRNA-mRNA interaction as curves, devising a novel seed enrichment metric, and learning an ensemble of miRNA family-specific kernel SVM classifiers. We provide an easy-to-use system for large-scale interactive analysis and prediction of miRNA targets. All operations in our system, namely candidate set generation, feature generation and transformation, training, prediction, and computing performance metrics are fully distributed and are scalable.

CONCLUSIONS

We have developed an efficient SVM-based model for miRNA target prediction using recent CLIP-seq data, demonstrating superior performance, evaluated using ROC curves for different species (human or mouse), or different target types (canonical or non-canonical). We analyzed the agreement between the target pairings using CLIP-seq data and using expression data from four cancer types. To the best of our knowledge, we provide the first distributed framework for miRNA target prediction based on Apache Hadoop and Spark.

AVAILABILITY

All source code and sample data are publicly available at https://bitbucket.org/cellsandmachines/avishkar. Our scalable implementation of kernel SVM using Apache Spark, which can be used to solve large-scale non-linear binary classification problems, is available at https://bitbucket.org/cellsandmachines/kernelsvmspark.

circEPSTI1 as a Prognostic Marker and Mediator of Triple-Negative Breast Cancer Progression

Circular RNAs (circRNAs) represent a class of non-coding RNAs that play a vital role in modulating gene expression and several pathological responses. However, the expression profile and function of circRNAs in triple-negative breast cancer (TNBC) remain unknown. In the current study, we investigated the expression profile of human circRNAs in TNBC tissues and identified circEPSTI1 (hsa_ circRNA_000479) as a significantly upregulated circRNA.

Methods: We performed circular RNA microarray assays to screen circular RNA expression profiles of TNBC and further investigated circEPSTI1. We observed the effect of circEPSTI1 on proliferation, clonal formation and apoptosis in TNBC by knocking downcircEPSTI1 in three TNBC cell lines. Based on the MRE analysis and luciferase reporter assay, we found that circEPSTI1 binds to miRNAs as a miRNA sponge and the co-target genes of miRNAs. We performed xenograft experiments in mice to confirm our findings. We evaluated circEPSTI1 levels in 240 TNBC patients by ISH.

Results: Knockdown of circEPSTI1 inhibits TNBC cell proliferation and induces apoptosis. In vitro and in vivo experiments indicated that circEPSTI1 binds to miR-4753 and miR-6809 as a miRNA sponge to regulate BCL11A expression and affect TNBC proliferation and apoptosis. High levels of circEPSTI1 correlate with reduced survival in TNBC patients.

Conclusions: The circEPSTI1-miR-4753/6809-BCL11A axis affect the proliferation and apoptosis of triple-negative breast cancer through the mechanism of competing endogenous RNAs (ceRNA). In addition, our results identify circEPSTI1 as an independent prognostic marker for survival in patients with TNBC.

A survey on database resources for microRNA-disease relationships

The relationships between microRNAs (miRNAs) and diseases are garnering greater interest in the biological research fields. Recently, miRNA-disease databases have emerged as powerful tools for bioinformatics studies of these relationships. However, guidelines for comparing the features of this type of database have not yet been established. In this article, the details of popular miRNA-disease databases are analyzed, and their features are compared from several different aspects, including database scale, disease classification, miRNA targets, miRNA detection technique, miRNA regulation, quantitative scores, study design and tissue/cell lines. Then, guidelines for choosing a suitable database for specific research interests are provided. This survey will guide computational biology or biological researchers as well as medical and clinical researchers in making better use of miRNA-disease data resources.

The "melanoma-enriched" microRNA miR-4731-5p acts as a tumour suppressor

We previously identified miR-4731-5p (miR-4731) as a melanoma-enriched microRNA following comparison of melanoma with other cell lines from solid malignancies. Additionally, miR-4731 has been found in serum from melanoma patients and expressed less abundantly in metastatic melanoma tissues from stage IV patients relative to stage III patients. As miR-4731 has no known function, we used biotin-labelled miRNA duplex pull-down to identify binding targets of miR-4731 in three melanoma cell lines (HT144, MM96L and MM253). Using the miRanda miRNA binding algorithm, all pulled-down transcripts common to the three cell lines (n=1092) had potential to be targets of miR-4731 and gene-set enrichment analysis of these (via STRING v9.1) highlighted significantly associated genes related to the 'cell cycle' pathway and the 'melanosome'. Following miR-4731 overexpression, a selection (n=81) of pull-down transcripts underwent validation using a custom qRT-PCR array. These data revealed that miR-4731 regulates multiple genes associated with the cell cycle (e.g. CCNA2, ORC5L, and PCNA) and the melanosome (e.g. RAB7A, CTSD, and GNA13). Furthermore, members of the synovial sarcoma X breakpoint family (SSX) (melanoma growth promoters) were also down-regulated (e.g. SSX2, SSX4, and SSX4B) as a result of miR-4731 overexpression. Moreover, this down-regulation of mRNA expression resulted in ablation or reduction of SSX4 protein, which, in keeping with previous studies, resulted in loss of 2D colony formation. We therefore speculate that loss of miR-4731 expression in stage IV patient tumours supports melanoma growth by, in part; reducing its regulatory control of SSX expression levels.

Bioinformatics facilitating the use of microarrays to delineate potential miRNA biomarkers in aristolochic acid nephropathy

Aristolochic acid nephropathy (AAN) is a rapidly progressive acute or chronic tubulointerstitial nephritis (TIN). The present study attempted to explore the molecular mechanisms underlying the miRNA-directed development of AAN. Our differentially expressed analysis identified 11 DE-miRNAs and retrieved the target genes of these DE-miRNAs; then, network analysis and functional analysis further identified 6 DE-miRNAs (has-miR-192, has-miR-194, has-miR-542-3p, has-miR-450a, has-miR-584, has-miR-33a) as phenotypic biomarkers of AAN. Surprisingly, of has-miR-192 has been reported to be associated with the pathogenesis of AAN, and has-miR-194, has-miR-542-3p and has-miR-450a was first-time identified to link to the development of AAN. In addition, the expressional changes of has-miR-584 and has-miR-33a may be associated with the development of AAN as well, which must be further confirmed by the associated experiments. Taken together, our work reveals for the first time the regulatory mechanisms of miRNAs in the development of AAN and this will contribute to miRNA-based diagnosis and treatment of AAN.

Unveiling MYCN regulatory networks in neuroblastoma via integrative analysis of heterogeneous genomics data

MYCN, an oncogenic transcription factor of the Myc family, is a major driver of neuroblastoma tumorigenesis. Due to the difficulty in drugging MYCN directly, revealing the molecules in MYCN regulatory networks will help to identify effective therapeutic targets for neuroblastoma therapy. Here we perform ChIP-sequencing and small RNA-sequencing of neuroblastoma cells to determine the MYCN-binding sites and MYCN-associated microRNAs, and integrate various types of genomic data to construct MYCN regulatory networks. The overall analysis indicated that MYCN-regulated genes were involved in a wide range of biological processes and could be used as signatures to identify poor-prognosis MYCN-non-amplified patients. Analysis of the MYCN binding sites showed that MYCN principally served as an activator. Using a computational approach, we identified 32 MYCN co-regulators, and some of these findings are supported by previous studies. Moreover, we investigated the interplay between MYCN transcriptional and microRNA post-transcriptional regulations and identified several microRNAs, such as miR-124-3p and miR-93-5p, which may significantly contribute to neuroblastoma pathogenesis. We also found MYCN and its regulated microRNAs acted together to repress the tumor suppressor genes. This work provides a comprehensive view of MYCN regulations for exploring therapeutic targets in neuroblastoma, as well as insights into the mechanism of neuroblastoma tumorigenesis.

The miR-125 family is an important regulator of the expression and maintenance of maternal effect genes during preimplantational embryo development

Previously, we reported that Sebox is a new maternal effect gene (MEG) that is required for early embryo development beyond the two-cell (2C) stage because this gene orchestrates the expression of important genes for zygotic genome activation (ZGA). However, regulators of Sebox expression remain unknown. Therefore, the objectives of the present study were to use bioinformatics tools to identify such regulatory microRNAs (miRNAs) and to determine the effects of the identified miRNAs on Sebox expression. Using computational algorithms, we identified a motif within the 3′UTR of Sebox mRNA that is specific to the seed region of the miR-125 family, which includes miR-125a-5p, miR-125b-5p and miR-351-5p. During our search for miRNAs, we found that the Lin28a 3′UTR also contains the same binding motif for the seed region of the miR-125 family. In addition, we confirmed that Lin28a also plays a role as a MEG and affects ZGA at the 2C stage, without affecting oocyte maturation or fertilization. Thus, we provide the first report indicating that the miR-125 family plays a crucial role in regulating MEGs related to the 2C block and in regulating ZGA through methods such as affecting Sebox and Lin28a in oocytes and embryos.

Overexpression of Chromosome 21 miRNAs May Affect Mitochondrial Function in the Hearts of Down Syndrome Fetuses

Dosage-dependent upregulation of most of chromosome 21 (Hsa21) genes has been demonstrated in heart tissues of fetuses with Down syndrome (DS). Also miRNAs might play important roles in the cardiac phenotype as they are highly expressed in the heart and regulate cardiac development. Five Hsa21 miRNAs have been well studied in the past: miR-99a-5p, miR-125b-2-5p, let-7c-5p, miR-155-5p, and miR-802-5p but few information is available about their expression in trisomic tissues. In this study, we evaluated the expression of these miRNAs in heart tissues from DS fetuses, showing that miR-99a-5p, miR-155-5p, and let-7c-5p were overexpressed in trisomic hearts. To investigate their role, predicted targets were obtained from different databases and cross-validated using the gene expression profiling dataset we previously generated for fetal hearts. Eighty-five targets of let-7c-5p, 33 of miR-155-5p, and 10 of miR-99a-5p were expressed in fetal heart and downregulated in trisomic hearts. As nuclear encoded mitochondrial genes were found downregulated in trisomic hearts and mitochondrial dysfunction is a hallmark of DS phenotypes, we put special attention to let-7c-5p and miR-155-5p targets downregulated in DS fetal hearts and involved in mitochondrial function. The let-7c-5p predicted target SLC25A4/ANT1 was identified as a possible candidate for both mitochondrial and cardiac anomalies.

PRINS Non-Coding RNA Regulates Nucleic Acid-Induced Innate Immune Responses of Human Keratinocytes

Cytosolic DNA fragments are recognized as pathogen- and danger-associated molecular patterns that induce a cascade of innate immune responses. Moreover, excessive cytosolic DNA can enhance chronic inflammation, predominantly by activating inflammasomes, and thereby contributing to the pathogenesis of chronic diseases, such as psoriasis. Psoriasis associated non-protein coding RNA induced by stress (PRINS) is a long non-coding RNA, which has been shown to be associated with psoriasis susceptibility and cellular stress responses; however, the precise mechanism of its action has not been determined. Here, we provide evidence that, in addition to inflammasome activation, cytosolic DNA induces intracellular inflammatory reactions while decreasing PRINS gene expression. Furthermore, PRINS overexpression can ameliorate the inflammatory-mediator production of keratinocytes induced by cytosolic DNA. Overexpression of PRINS resulted in decreased interleukin-6 (IL-6) and chemokine (C–C motif) ligand 5 (CCL-5) expression and secretion. In silico analysis predicted direct binding sites between PRINS and the mRNAs, which was confirmed by an in vitro binding assay and on cellular level. Our results indicated that PRINS binds directly to the mRNAs of IL-6 and CCL-5 at specific binding sites and eventually destabilizes these mRNAs, leading to a decrease in their expression and secretion of the corresponding proteins. These results may indicate a restrictive role for PRINS in inflammatory processes.

miSTAR: miRNA target prediction through modeling quantitative and qualitative miRNA binding site information in a stacked model structure

In microRNA (miRNA) target prediction, typically two levels of information need to be modeled: the number of potential miRNA binding sites present in a target mRNA and the genomic context of each individual site. Single model structures insufficiently cope with this complex training data structure, consisting of feature vectors of unequal length as a consequence of the varying number of miRNA binding sites in different mRNAs. To circumvent this problem, we developed a two-layered, stacked model, in which the influence of binding site context is separately modeled. Using logistic regression and random forests, we applied the stacked model approach to a unique data set of 7990 probed miRNA-mRNA interactions, hereby including the largest number of miRNAs in model training to date. Compared to lower-complexity models, a particular stacked model, named miSTAR (miRNA stacked model target prediction; www.mi-star.org), displays a higher general performance and precision on top scoring predictions. More importantly, our model outperforms published and widely used miRNA target prediction algorithms. Finally, we highlight flaws in cross-validation schemes for evaluation of miRNA target prediction models and adopt a more fair and stringent approach.

miRNAs target databases: developmental methods and target identification techniques with functional annotations

PURPOSE

microRNA (miRNA) regulates diverse biological mechanisms and metabolisms in plants and animals. Thus, the discoveries of miRNA has revolutionized the life sciences and medical research.The miRNA represses and cleaves the targeted mRNA by binding perfect or near perfect or imperfect complementary base pairs by RNA-induced silencing complex (RISC) formation during biogenesis process. One miRNA interacts with one or more mRNA genes and vice versa, hence takes part in causing various diseases. In this paper, the different microRNA target databases and their functional annotations developed by various researchers have been reviewed. The concurrent research review aims at comprehending the significance of miRNA and presenting the existing status of annotated miRNA target resources built by researchers henceforth discovering the knowledge for diagnosis and prognosis.

METHODS AND RESULTS

This review discusses the applications and developmental methodologies for constructing target database as well as the utility of user interface design. An integrated architecture is drawn and a graphically comparative study of present status of miRNA targets in diverse diseases and various biological processes is performed. These databases comprise of information such as miRNA target-associated disease, transcription factor binding sites (TFBSs) in miRNA genomic locations, polymorphism in miRNA target, A-to-I edited target, Gene Ontology (GO), genome annotations, KEGG (Kyoto Encyclopedia of Genes and Genomes) pathways, target expression analysis, TF-miRNA and miRNA-mRNA interaction networks, drugs-targets interactions, etc.

CONCLUSION

miRNA target databases contain diverse experimentally and computationally predicted target through various algorithms. The comparison of various miRNA target database has been performed on various parameters. The computationally predicted target databases suffer from false positive information as there is no common theory for prediction of miRNA targets. The review conclusion emphasizes the need of more intelligent computational improvement for the miRNA target identification, their functional annotations and datasbase development.

Combining Results from Distinct MicroRNA Target Prediction Tools Enhances the Performance of Analyses

Target prediction is generally the first step toward recognition of bona fide microRNA (miRNA)-target interactions in living cells. Several target prediction tools are now available, which use distinct criteria and stringency to provide the best set of candidate targets for a single miRNA or a subset of miRNAs. However, there are many false-negative predictions, and consensus about the optimum strategy to select and use the output information provided by the target prediction tools is lacking. We compared the performance of four tools cited in literature—TargetScan (TS), miRanda-mirSVR (MR), Pita, and RNA22 (R22), and we determined the most effective approach for analyzing target prediction data (individual, union, or intersection). For this purpose, we calculated the sensitivity, specificity, precision, and correlation of these approaches using 10 miRNAs (miR-1-3p, miR-17-5p, miR-21-5p, miR-24-3p, miR-29a-3p, miR-34a-5p, miR-124-3p, miR-125b-5p, miR-145-5p, and miR-155-5p) and 1,400 genes (700 validated and 700 non-validated) as targets of these miRNAs. The four tools provided a subset of high-quality predictions and returned few false-positive predictions; however, they could not identify several known true targets. We demonstrate that union of TS/MR and TS/MR/R22 enhanced the quality of in silico prediction analysis of miRNA targets. We conclude that the union rather than the intersection of the aforementioned tools is the best strategy for maximizing performance while minimizing the loss of time and resources in subsequent in vivo and in vitro experiments for functional validation of miRNA-target interactions.

Improving prediction accuracy using decision-tree-based meta-strategy and multi-threshold sequential-voting exemplified by miRNA target prediction

Lots of computational predictors have been developed for fast and large-scale analysis of biological data. However, many of them were developed long time ago when training datasets or sets of input features were rather small. Consequently, the utility of these predictors in much large datasets, which are very common in nowadays, need to be examined carefully. In addition, with the rapid development of scientific research, the expectation on the prediction accuracy of computational predictors is continuously uplifting. Therefore, developing novel strategies to improve the prediction accuracies of computational predictors becomes critical. In this study, the predictive results of existing individual miRNA target predictors were integrated into a decision-tree to make meta-prediction. When the multi-threshold sequential-voting technique was used, the prediction accuracy of the decision-tree was significantly improved by at least thirty percentage points compared to the individual predictors.

miRTar2GO: a novel rule-based model learning method for cell line specific microRNA target prediction that integrates Ago2 CLIP-Seq and validated microRNA-target interaction data

MicroRNAs (miRNAs) are ∼19-22 nucleotides (nt) long regulatory RNAs that regulate gene expression by recognizing and binding to complementary sequences on mRNAs. The key step in revealing the function of a miRNA, is the identification of miRNA target genes. Recent biochemical advances including PAR-CLIP and HITS-CLIP allow for improved miRNA target predictions and are widely used to validate miRNA targets. Here, we present miRTar2GO, which is a model, trained on the common rules of miRNA-target interactions, Argonaute (Ago) CLIP-Seq data and experimentally validated miRNA target interactions. miRTar2GO is designed to predict miRNA target sites using more relaxed miRNA-target binding characteristics. More importantly, miRTar2GO allows for the prediction of cell-type specific miRNA targets. We have evaluated miRTar2GO against other widely used miRNA target prediction algorithms and demonstrated that miRTar2GO produced significantly higher F1 and G scores. Target predictions, binding specifications, results of the pathway analysis and gene ontology enrichment of miRNA targets are freely available at http://www.mirtar2go.org.

Novel blood-based microRNA biomarker panel for early diagnosis of chronic pancreatitis

Chronic pancreatitis (CP) is an inflammatory disease characterized by progressive fibrosis of pancreas. Early diagnosis will improve the prognosis of patients. This study aimed to obtain serum miRNA biomarkers for early diagnosis of CP. In the current study, we analyzed the differentially expressed miRNAs (DEmiRs) of CP patients from Gene Expression Omnibus (GEO), and the DEmiRs in plasma of early CP patients (n = 10) from clinic by miRNA microarrays. Expression levels of DEmiRs were further tested in clinical samples including early CP patients (n = 20), late CP patients (n = 20) and healthy controls (n = 18). The primary endpoints were area under curve (AUC) and expression levels of DEmiRs. Four DEmiRs (hsa-miR-320a-d) were obtained from GEO CP, meanwhile two (hsa-miR-221 and hsa-miR-130a) were identified as distinct biomarkers of early CP by miRNA microarrays. When applied on clinical serum samples, hsa-miR-320a-d were accurate in predicting late CP, while hsa-miR-221 and hsa-miR-130a were accurate in predicting early CP with AUC of 100.0% and 87.5%. Our study indicates that miRNA expression profile is different in early and late CP. Hsa-miR-221 and hsa-miR-130a are biomarkers of early CP, and the panel of the above 6 serum miRNAs has the potential to be applied clinically for early diagnosis of CP.

Visualization and Analysis of MiRNA-Targets Interactions Networks

MicroRNAs are a class of small, noncoding RNA molecules of 21-25 nucleotides in length that regulate the gene expression by base-pairing with the target mRNAs, mainly leading to down-regulation or repression of the target genes. MicroRNAs are involved in diverse regulatory pathways in normal and pathological conditions. In this context, it is highly important to identify the targets of specific microRNA in order to understand the mechanism of its regulation and consequently its involvement in disease. However, the microRNA target identification is experimentally laborious and time-consuming. The in silico prediction of microRNA targets is an extremely useful approach because you can identify potential mRNA targets, reduce the number of possibilities and then, validate a few microRNA-mRNA interactions in an in vitro experimental model. In this chapter, we describe, in a simple way, bioinformatics guidelines to use miRWalk database and Cytoscape software for analyzing microRNA-mRNA interactions through their visualization as a network.

Computational and Experimental Identification of Tissue-Specific MicroRNA Targets

In this chapter we discuss computational methods for the prediction of microRNA (miRNA) targets. More specifically, we consider machine learning-based approaches and explain why these methods have been relatively unsuccessful in reducing the number of false positive predictions. Further we suggest approaches designed to improve their performance by considering tissue-specific target regulation. We argue that the miRNA targetome differs depending on the tissue type and introduce a novel algorithm that predicts miRNA targets specifically for colorectal cancer. We discuss features of miRNAs and target sites that affect target recognition, and how next-generation sequencing data can support the identification of novel miRNAs, differentially expressed miRNAs and their tissue-specific mRNA targets. In addition, we introduce some experimental approaches for the validation of miRNA targets as well as web-based resources sharing predicted and validated miRNA target interactions.

Genome-Wide Analysis of MicroRNA-Regulated Transcripts

MicroRNAs (miRNAs) are small noncoding RNAs that regulate gene expression by either degrading transcripts or repressing translation . Over the past decade the significance of miRNAs has been unraveled by the characterization of their involvement in crucial cellular functions and the development of disease. However, continued progress in understanding the endogenous importance of miRNAs, as well as their potential uses as therapeutic tools, has been hindered by the difficulty of positively identifying miRNA targets. To face this challenge algorithmic approaches have primarily been utilized to date, but strictly mathematical models have thus far failed to produce a generally accurate, widely accepted methodology for accurate miRNA target determination. As such, several laboratory-based, comprehensive strategies for experimentally identifying all cellular miRNA regulations simultaneously have recently been developed. This chapter discusses the advantages and limitations of both classic and comprehensive strategies for miRNA target prediction .

Gene-microRNA network module analysis for ovarian cancer

Background

MicroRNAs (miRNAs) are involved in many biological processes by regulating post-transcriptional gene expression. The alterations of the regulatory pathways can cause different diseases including cancer. Although many works have been done to study the gene-miRNA regulatory network, the intertwined relationship is far from being fully understood. The objective of this study is to integrate both gene expression data and miRNA data so as to explore the complex relationships among them.

Methods

By integrating the networks consisting of gene coexpression, miRNA coexpression, gene-miRNA coexpression, and the known gene-miRNA interactions, we aim to find the most connected network modules so as to study their functions and properties. In this paper, we proposed an optimization model for identification of the modules in the integrated networks. This model tries to find both the modules in the gene-gene and miRNA-miRNA coexpression networks and the densely connected gene-miRNA subneworks. An approximation computational method was developed to solve the optimization problem.

Results

We applied the method to 556 human ovarian cancer samples with both gene expression data and miRNA expression data. The identified modules are significantly enriched by miRNA clusters, GO-BPs, and KEGG pathways. We compared our method with some existing methods and showed the better performance of our method. We also showed that the miRNAs and genes in our identified modules are associated with cancers, especially ovarian cancer.

Conclusions

This study provides strong support that the subnetworks consisting of genes and miRNAs with close interactions contribute the cancers. The proposed computational method can be applied to other studies that are related to different types of networks.

Electronic supplementary material

The online version of this article (doi:10.1186/s12918-016-0357-1) contains supplementary material, which is available to authorized users.

Prediction of human miRNA target genes using computationally reconstructed ancestral mammalian sequences

MicroRNAs (miRNA) are short single-stranded RNA molecules derived from hairpin-forming precursors that play a crucial role as post-transcriptional regulators in eukaryotes and viruses. In the past years, many microRNA target genes (MTGs) have been identified experimentally. However, because of the high costs of experimental approaches, target genes databases remain incomplete. Although several target prediction programs have been developed in the recent years to identify MTGs in silico, their specificity and sensitivity remain low. Here, we propose a new approach called MirAncesTar, which uses ancestral genome reconstruction to boost the accuracy of existing MTGs prediction tools for human miRNAs. For each miRNA and each putative human target UTR, our algorithm makes uses of existing prediction tools to identify putative target sites in the human UTR, as well as in its mammalian orthologs and inferred ancestral sequences. It then evaluates evidence in support of selective pressure to maintain target site counts (rather than sequences), accounting for the possibility of target site turnover. It finally integrates this measure with several simpler ones using a logistic regression predictor. MirAncesTar improves the accuracy of existing MTG predictors by 26% to 157%. Source code and prediction results for human miRNAs, as well as supporting evolutionary data are available at http://cs.mcgill.ca/∼blanchem/mirancestar.

MiRTDL: A Deep Learning Approach for miRNA Target Prediction

MicroRNAs (miRNAs) regulate genes that are associated with various diseases. To better understand miRNAs, the miRNA regulatory mechanism needs to be investigated and the real targets identified. Here, we present miRTDL, a new miRNA target prediction algorithm based on convolutional neural network (CNN). The CNN automatically extracts essential information from the input data rather than completely relying on the input dataset generated artificially when the precise miRNA target mechanisms are poorly known. In this work, the constraint relaxing method is first used to construct a balanced training dataset to avoid inaccurate predictions caused by the existing unbalanced dataset. The miRTDL is then applied to 1,606 experimentally validated miRNA target pairs. Finally, the results show that our miRTDL outperforms the existing target prediction algorithms and achieves significantly higher sensitivity, specificity and accuracy of 88.43, 96.44, and 89.98 percent, respectively. We also investigate the miRNA target mechanism, and the results show that the complementation features are more important than the others.

TarPmiR: a new approach for microRNA target site prediction

MOTIVATION

The identification of microRNA (miRNA) target sites is fundamentally important for studying gene regulation. There are dozens of computational methods available for miRNA target site prediction. Despite their existence, we still cannot reliably identify miRNA target sites, partially due to our limited understanding of the characteristics of miRNA target sites. The recently published CLASH (crosslinking ligation and sequencing of hybrids) data provide an unprecedented opportunity to study the characteristics of miRNA target sites and improve miRNA target site prediction methods.

RESULTS

Applying four different machine learning approaches to the CLASH data, we identified seven new features of miRNA target sites. Combining these new features with those commonly used by existing miRNA target prediction algorithms, we developed an approach called TarPmiR for miRNA target site prediction. Testing on two human and one mouse non-CLASH datasets, we showed that TarPmiR predicted more than 74.2% of true miRNA target sites in each dataset. Compared with three existing approaches, we demonstrated that TarPmiR is superior to these existing approaches in terms of better recall and better precision.

AVAILABILITY AND IMPLEMENTATION

The TarPmiR software is freely available at http://hulab.ucf.edu/research/projects/miRNA/TarPmiR/ CONTACTS: haihu@cs.ucf.edu or xiaoman@mail.ucf.edu

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Tiny giants of gene regulation: experimental strategies for microRNA functional studies

The discovery over two decades ago of short regulatory microRNAs (miRNAs) has led to the inception of a vast biomedical research field dedicated to understanding these powerful orchestrators of gene expression. Here we aim to provide a comprehensive overview of the methods and techniques underpinning the experimental pipeline employed for exploratory miRNA studies in animals. Some of the greatest challenges in this field have been uncovering the identity of miRNA-target interactions and deciphering their significance with regard to particular physiological or pathological processes. These endeavors relied almost exclusively on the development of powerful research tools encompassing novel bioinformatics pipelines, high-throughput target identification platforms, and functional target validation methodologies. Thus, in an unparalleled manner, the biomedical technology revolution unceasingly enhanced and refined our ability to dissect miRNA regulatory networks and understand their roles in vivo in the context of cells and organisms. Recurring motifs of target recognition have led to the creation of a large number of multifactorial bioinformatics analysis platforms, which have proved instrumental in guiding experimental miRNA studies. Subsequently, the need for discovery of miRNA-target binding events in vivo drove the emergence of a slew of high-throughput multiplex strategies, which now provide a viable prospect for elucidating genome-wide miRNA-target binding maps in a variety of cell types and tissues. Finally, deciphering the functional relevance of miRNA post-transcriptional gene silencing under physiological conditions, prompted the evolution of a host of technologies enabling systemic manipulation of miRNA homeostasis as well as high-precision interference with their direct, endogenous targets. For further resources related to this article, please visit the WIREs website.

Web-based tools for microRNAs involved in human cancer

MicroRNAs (miRNAs/miRs) are a family of small, endogenous and evolutionarily-conserved non-coding RNAs that are involved in the regulation of several cellular and functional processes. miRNAs can act as oncogenes or tumor suppressors in all types of cancer, and could be used as prognostic and diagnostic biomarkers. Databases and computational algorithms are behind the majority of the research performed on miRNAs. These tools assemble and curate the relevant information on miRNAs and present it in a user-friendly manner. The current review presents 14 online databases that address every aspect of miRNA cancer research. Certain databases focus on miRNAs and a particular type of cancer, while others analyze the behavior of miRNAs in different malignancies at the same time. Additional databases allow researchers to search for mutations in miRNAs or their targets, and to review the naming history of a particular miRNA. All these databases are open-access, and are a valuable tool for those researchers working with these molecules, particularly those who lack access to an advanced computational infrastructure.

Naïve Bayes classifier predicts functional microRNA target interactions in colorectal cancer

Alterations in the expression of miRNAs have been extensively characterized in several cancers, including human colorectal cancer (CRC). Recent publications provide evidence for tissue-specific miRNA target recognition. Several computational methods have been developed to predict miRNA targets; however, all of these methods assume a general pattern underlying these interactions and therefore tolerate reduced prediction accuracy and a significant number of false predictions. The motivation underlying the presented work was to unravel the relationship between miRNAs and their target mRNAs in CRC. We developed a novel computational algorithm for miRNA-target prediction in CRC using a Naïve Bayes classifier. The algorithm, which is referred to as CRCmiRTar, was trained with data from validated miRNA target interactions in CRC and other cancer entities. Furthermore, we identified a set of position-based, sequence, structural, and thermodynamic features that identify CRC-specific miRNA target interactions. Evaluation of the algorithm showed a significant improvement of performance with respect to AUC, and sensitivity, compared to other widely used algorithms based on machine learning. Based on miRNA and gene expression profiles in CRC tissues with similar clinical and pathological features, our classifier predicted 204 functional interactions, which involve 11 miRNAs and 41 mRNAs in this cancer entity. While the approach is here validated for CRC, the implementation of disease-specific miRNA target prediction algorithms can be easily adopted for other applications too. The identification of disease-specific miRNA target interactions may also facilitate the identification of potential drug targets.

MicroRNAs: exploring new horizons in osteoarthritis

INTRODUCTION

Osteoarthritis (OA) is a common disease worldwide leading to significant morbidity. The underlying disease process is multifactorial however there is increasing focus on molecular mechanisms. MicroRNAs are small non-coding segments of RNA that have important regulatory functions at a cellular level. These molecules are readily detectable in human tissues and circulation. They are increasingly recognised as having a major role in many disease processes - including malignancy and inflammatory processes.

OBJECTIVE

This review paper aims to provide a comprehensive update on the evidence for miRNA roles in OA.

DESIGN

A comprehensive literature search was performed using key medical subject headings (MeSH) terms 'microRNA' and 'osteoarthritis'.

RESULTS

Several miRNAs have been identified as having aberrant expression levels in OA. Some of these include miR-9, miR-27, miR-34a, miR-140, miR-146a, miR-558 and miR-602. Many of the dysregulated miRNAs have been shown to regulate expression of inflammatory pathways such as interleukin-mediated or matrix metalloproteinase-13 (MMP-13)-mediated degradation of the articular cartilage extracellular matrix (ECM). MiRNAs may also play a role in pain pathways and hence expression of clinical symptoms.

CONCLUSIONS

Recent evidence has shown that miRNAs in the circulation may reflect underlying disease states and hence serve as potential markers for disease activity. These findings may represent possible future therapeutic applications in the management of OA.

Analyzing the Role of MicroRNAs in Schizophrenia in the Context of Common Genetic Risk Variants

IMPORTANCE

The recent implication of 108 genomic loci in schizophrenia marked a great advancement in our understanding of the disease. Against the background of its polygenic nature there is a necessity to identify how schizophrenia risk genes interplay. As regulators of gene expression, microRNAs (miRNAs) have repeatedly been implicated in schizophrenia etiology. It is therefore of interest to establish their role in the regulation of schizophrenia risk genes in disease-relevant biological processes.

OBJECTIVE

To examine the role of miRNAs in schizophrenia in the context of disease-associated genetic variation.

DESIGN, SETTING, AND PARTICIPANTS

The basis of this study was summary statistics from the largest schizophrenia genome-wide association study meta-analysis to date (83 550 individuals in a meta-analysis of 52 genome-wide association studies) completed in 2014 along with publicly available data for predicted miRNA targets. We examined whether schizophrenia risk genes were more likely to be regulated by miRNA. Further, we used gene set analyses to identify miRNAs that are regulators of schizophrenia risk genes.

MAIN OUTCOMES AND MEASURES

Results from association tests for miRNA targetomes and related analyses.

RESULTS

In line with previous studies, we found that similar to other complex traits, schizophrenia risk genes were more likely to be regulated by miRNAs (P < 2 × 10-16). Further, the gene set analyses revealed several miRNAs regulating schizophrenia risk genes, with the strongest enrichment for targets of miR-9-5p (P = .0056 for enrichment among the top 1% most-associated single-nucleotide polymorphisms, corrected for multiple testing). It is further of note that MIR9-2 is located in a genomic region showing strong evidence for association with schizophrenia (P = 7.1 × 10-8). The second and third strongest gene set signals were seen for the targets of miR-485-5p and miR-137, respectively.

CONCLUSIONS AND RELEVANCE

This study provides evidence for a role of miR-9-5p in the etiology of schizophrenia. Its implication is of particular interest as the functions of this neurodevelopmental miRNA tie in with established disease biology: it has a regulatory loop with the fragile X mental retardation homologue FXR1 and regulates dopamine D2 receptor density.

Maximum margin classifier working in a set of strings

Numbers and numerical vectors account for a large portion of data. However, recently, the amount of string data generated has increased dramatically. Consequently, classifying string data is a common problem in many fields. The most widely used approach to this problem is to convert strings into numerical vectors using string kernels and subsequently apply a support vector machine that works in a numerical vector space. However, this non-one-to-one conversion involves a loss of information and makes it impossible to evaluate, using probability theory, the generalization error of a learning machine, considering that the given data to train and test the machine are strings generated according to probability laws. In this study, we approach this classification problem by constructing a classifier that works in a set of strings. To evaluate the generalization error of such a classifier theoretically, probability theory for strings is required. Therefore, we first extend a limit theorem for a consensus sequence of strings demonstrated by one of the authors and co-workers in a previous study. Using the obtained result, we then demonstrate that our learning machine classifies strings in an asymptotically optimal manner. Furthermore, we demonstrate the usefulness of our machine in practical data analysis by applying it to predicting protein-protein interactions using amino acid sequences and classifying RNAs by the secondary structure using nucleotide sequences.

Integration of Multiple Genomic and Phenotype Data to Infer Novel miRNA-Disease Associations

MicroRNAs (miRNAs) play an important role in the development and progression of human diseases. The identification of disease-associated miRNAs will be helpful for understanding the molecular mechanisms of diseases at the post-transcriptional level. Based on different types of genomic data sources, computational methods for miRNA-disease association prediction have been proposed. However, individual source of genomic data tends to be incomplete and noisy; therefore, the integration of various types of genomic data for inferring reliable miRNA-disease associations is urgently needed. In this study, we present a computational framework, CHNmiRD, for identifying miRNA-disease associations by integrating multiple genomic and phenotype data, including protein-protein interaction data, gene ontology data, experimentally verified miRNA-target relationships, disease phenotype information and known miRNA-disease connections. The performance of CHNmiRD was evaluated by experimentally verified miRNA-disease associations, which achieved an area under the ROC curve (AUC) of 0.834 for 5-fold cross-validation. In particular, CHNmiRD displayed excellent performance for diseases without any known related miRNAs. The results of case studies for three human diseases (glioblastoma, myocardial infarction and type 1 diabetes) showed that all of the top 10 ranked miRNAs having no known associations with these three diseases in existing miRNA-disease databases were directly or indirectly confirmed by our latest literature mining. All these results demonstrated the reliability and efficiency of CHNmiRD, and it is anticipated that CHNmiRD will serve as a powerful bioinformatics method for mining novel disease-related miRNAs and providing a new perspective into molecular mechanisms underlying human diseases at the post-transcriptional level. CHNmiRD is freely available at http://www.bio-bigdata.com/CHNmiRD.

A new approach to human microRNA target prediction using ensemble pruning and rotation forest

MicroRNAs (miRNAs) are small non-coding RNAs that have important functions in gene regulation. Since finding miRNA target experimentally is costly and needs spending much time, the use of machine learning methods is a growing research area for miRNA target prediction. In this paper, a new approach is proposed by using two popular ensemble strategies, i.e. Ensemble Pruning and Rotation Forest (EP-RTF), to predict human miRNA target. For EP, the approach utilizes Genetic Algorithm (GA). In other words, a subset of classifiers from the heterogeneous ensemble is first selected by GA. Next, the selected classifiers are trained based on the RTF method and then are combined using weighted majority voting. In addition to seeking a better subset of classifiers, the parameter of RTF is also optimized by GA. Findings of the present study confirm that the newly developed EP-RTF outperforms (in terms of classification accuracy, sensitivity, and specificity) the previously applied methods over four datasets in the field of human miRNA target. Diversity-error diagrams reveal that the proposed ensemble approach constructs individual classifiers which are more accurate and usually diverse than the other ensemble approaches. Given these experimental results, we highly recommend EP-RTF for improving the performance of miRNA target prediction.

DIANA-miRGen v3.0: accurate characterization of microRNA promoters and their regulators

microRNAs (miRNAs) are small non-coding RNAs that actively fine-tune gene expression. The accurate characterization of the mechanisms underlying miRNA transcription regulation will further expand our knowledge regarding their implication in homeostatic and pathobiological networks. Aim of DIANA-miRGen v3.0 (http://www.microrna.gr/mirgen) is to provide for the first time accurate cell-line-specific miRNA gene transcription start sites (TSSs), coupled with genome-wide maps of transcription factor (TF) binding sites in order to unveil the mechanisms of miRNA transcription regulation. To this end, more than 7.3 billion RNA-, ChIP- and DNase-Seq next generation sequencing reads were analyzed/assembled and combined with state-of-the-art miRNA TSS prediction and TF binding site identification algorithms. The new database schema and web interface facilitates user interaction, provides advanced queries and innate connection with other DIANA resources for miRNA target identification and pathway analysis. The database currently supports 276 miRNA TSSs that correspond to 428 precursors and >19M binding sites of 202 TFs on a genome-wide scale in nine cell-lines and six tissues of Homo sapiens and Mus musculus.

Human MicroRNA miR-532-5p Exhibits Antiviral Activity against West Nile Virus via Suppression of Host Genes SESTD1 and TAB3 Required for Virus Replication

West Nile virus (WNV) is a mosquito-transmitted flavivirus that naturally circulates between mosquitos and birds but can also infect humans, causing severe neurological disease. The early host response to WNV infection in vertebrates primarily relies on the type I interferon pathway; however, recent studies suggest that microRNAs (miRNAs) may also play a notable role. In this study, we assessed the role of host miRNAs in response to WNV infection in human cells. We employed small RNA sequencing (RNA-seq) analysis to determine changes in the expression of host miRNAs in HEK293 cells infected with an Australian strain of WNV, Kunjin (WNVKUN), and identified a number of host miRNAs differentially expressed in response to infection. Three of these miRNAs were confirmed to be significantly upregulated in infected cells by quantitative reverse transcription (qRT)-PCR and Northern blot analyses, and one of them, miR-532-5p, exhibited a significant antiviral effect against WNVKUN infection. We have demonstrated that miR-532-5p targets and downregulates expression of the host genes SESTD1 and TAB3 in human cells. Small interfering RNA (siRNA) depletion studies showed that both SESTD1 and TAB3 were required for efficient WNVKUN replication. We also demonstrated upregulation of mir-532-5p expression and a corresponding decrease in the expression of its targets, SESTD1 and TAB3, in the brains of WNVKUN -infected mice. Our results show that upregulation of miR-532-5p and subsequent suppression of the SESTD1 and TAB3 genes represent a host antiviral response aimed at limiting WNVKUN infection and highlight the important role of miRNAs in controlling RNA virus infections in mammalian hosts.

IMPORTANCE

West Nile virus (WNV) is a significant viral pathogen that poses a considerable threat to human health across the globe. There is no specific treatment or licensed vaccine available for WNV, and deeper insight into how the virus interacts with the host is required to facilitate their development. In this study, we addressed the role of host microRNAs (miRNAs) in antiviral response to WNV in human cells. We identified miR-532-5p as a novel antiviral miRNA and showed that it is upregulated in response to WNV infection and suppresses the expression of the host genes TAB3 and SESTD1 required for WNV replication. Our results show that upregulation of miR-532-5p and subsequent suppression of the SESTD1 and TAB3 genes represent an antiviral response aimed at limiting WNV infection and highlight the important role of miRNAs in controlling virus infections in mammalian hosts.

Human lung microRNA profiling in pulmonary arterial hypertension secondary to congenital heart defect

OBJECTIVE

Although several microRNAs were reported to play essential roles in pulmonary artery hypertension due to hypoxia or monocrotaline, their potential role in pulmonary arterial hypertension secondary to congenital heart disease is largely unknown. This study aimed to indentify microRNAs implicated in pulmonary arterial hypertension secondary to congenital heart disease in children.

METHODS

Using microRNAs microarray, we profiled the microRNAs in the lung specimen from 12 congenital heart disease patients, (6 with pulmonary arterial hypertension and the others without). We validated the microRNAs expression using RT-PCR experiments. Then, we predicted the target genes of the promising microRNAs by bioinformatical analysis and verified its regulating role by luciferase assay and western blot experiments.

RESULTS

All the 12 patients were uneventfully recovered from cardiac surgery. Comparing to the non-pulmonary arterial hypertension lung tissue, 62 microRNAs were significantly up-regulated and 12 were significantly de-regulated in the pulmonary arterial hypertension lung tissue. Among them 27 microRNAs reached P values ≤ 0.05, we validated the up-regulation of microRNA-27b by RT-PCR experiments and found the expression level of microRNA-27b was correlated with preoperative mean pulmonary arterial pressure. In vitro, overexpression of microRNA-27b decreased the protein expression of NOTCH1 and significantly reduced luciferase activity.

CONCLUSIONS

The current study revealed for the first time that microRNAs may be important regulators in pulmonary arterial hypertension secondary to congenital heart disease, and demonstrated the correlation between microRNA-27b and pulmonary arterial hypertension with the implication of NOTCH1.

MicroRNA target prediction using thermodynamic and sequence curves

BACKGROUND

MicroRNAs (miRNAs) are small regulatory RNA that mediate RNA interference by binding to various mRNA target regions. There have been several computational methods for the identification of target mRNAs for miRNAs. However, these have considered all contributory features as scalar representations, primarily, as thermodynamic or sequence-based features. Further, a majority of these methods solely target canonical sites, which are sites with "seed" complementarity. Here, we present a machine-learning classification scheme, titled Avishkar, which captures the spatial profile of miRNA-mRNA interactions via smooth B-spline curves, separately for various input features, such as thermodynamic and sequence features. Further, we use a principled approach to uniformly model canonical and non-canonical seed matches, using a novel seed enrichment metric.

RESULTS

We demonstrate that large number of seed-match patterns have high enrichment values, conserved across species, and that majority of miRNA binding sites involve non-canonical matches, corroborating recent findings. Using spatial curves and popular categorical features, such as target site length and location, we train a linear SVM model, utilizing experimental CLIP-seq data. Our model significantly outperforms all established methods, for both canonical and non-canonical sites. We achieve this while using a much larger candidate miRNA-mRNA interaction set than prior work.

CONCLUSIONS

We have developed an efficient SVM-based model for miRNA target prediction using recent CLIP-seq data, demonstrating superior performance, evaluated using ROC curves, specifically about 20% better than the state-of-the-art, for different species (human or mouse), or different target types (canonical or non-canonical). To the best of our knowledge we provide the first distributed framework for microRNA target prediction based on Apache Hadoop and Spark.

AVAILABILITY

All source code and data is publicly available at https://bitbucket.org/cellsandmachines/avishkar.

Genome-wide analysis of differentially expressed miRNA in PLX4720-resistant and parental human thyroid cancer cell lines

BACKGROUND

Investigating BRAF((V600E)) inhibitors (BRAFi) as a strategy to treat patients with aggressive thyroid tumors harboring the BRAF((V600E)) mutant currently is in progress, and drug resistance is expected to pose a challenge. MicroRNAs (miRNAs) are involved in development of resistance to a variety of drugs in different malignancies.

METHODS

miRNA expression profiles in the human anaplastic thyroid cancer cell line (8505c) were compared with its PLX4720-resistant counterpart (8505c-R) by the use of Illumina deep sequencing. We conducted a functional annotation and pathway analysis of the putative and experimentally validated target genes of the significantly altered miRNAs.

RESULTS

We identified 61 known and 2 novel miRNAs whose expression was altered greatly in 8505c-R. Quantitative reverse-transcription polymerase chain reaction validated altered expression of 7 selected miRNAs in 8505c-R and BCPAP-R (PLX4720-resistant papillary thyroid cancer cell line). We found 14 and 25 miRNAs whose expression levels changed substantially in 8505c and 8505c-R, respectively, after treatment with BRAFi. The mitogen-activated protein kinase and phosphatidylinositol 3-kinase-AKT pathways were among the prominent targets of many of the deregulated miRNAs.

CONCLUSION

We have identified a number of miRNAs that could be used as biomarkers of resistance to BRAFi in patients with thyroid cancer. In addition, these miRNAs can be explored as potential therapeutic targets in combination with BRAFi to overcome resistance.

MicroRNA and transcription factor mediated regulatory network analysis reveals critical regulators and regulatory modules in myocardial infarction

Myocardial infarction (MI) is a severe coronary artery disease and a leading cause of mortality and morbidity worldwide. However, the molecular mechanisms of MI have yet to be fully elucidated. In this study, we compiled MI-related genes, MI-related microRNAs (miRNAs) and known human transcription factors (TFs), and we then identified 1,232 feed-forward loops (FFLs) among these miRNAs, TFs and their co-regulated target genes through integrating target prediction. By merging these FFLs, the first miRNA and TF mediated regulatory network for MI was constructed, from which four regulators (SP1, ESR1, miR-21-5p and miR-155-5p) and three regulatory modules that might play crucial roles in MI were then identified. Furthermore, based on the miRNA and TF mediated regulatory network and literature survey, we proposed a pathway model for miR-21-5p, the miR-29 family and SP1 to demonstrate their potential co-regulatory mechanisms in cardiac fibrosis, apoptosis and angiogenesis. The majority of the regulatory relations in the model were confirmed by previous studies, which demonstrated the reliability and validity of this miRNA and TF mediated regulatory network. Our study will aid in deciphering the complex regulatory mechanisms involved in MI and provide putative therapeutic targets for MI.

Concordant dysregulation of miR-5p and miR-3p arms of the same precursor microRNA may be a mechanism in inducing cell proliferation and tumorigenesis: a lung cancer study

A precursor microRNA (miRNA) has two arms: miR-5p and miR-3p (miR-5p/-3p). Depending on the tissue or cell types, both arms can become functional. However, little is known about their coregulatory mechanisms during the tumorigenic process. Here, by using the large-scale miRNA expression profiles of five cancer types, we revealed that several of miR-5p/-3p arms were concordantly dysregulated in each cancer. To explore possible coregulatory mechanisms of concordantly dysregulated miR-5p/-3p pairs, we developed a robust computational framework and applied it to lung cancer data. The framework deciphers miR-5p/-3p coregulated protein interaction networks critical to lung cancer development. As a novel part in the method, we uniquely applied the second-order partial correlation to minimize false-positive regulations. Using 279 matched miRNA and mRNA expression profiles extracted from tumor and normal lung tissue samples, we identified 17 aberrantly expressed miR-5p/-3p pairs that potentially modulate the gene expression of 35 protein complexes. Functional analyses revealed that these complexes are associated with cancer-related biological processes, suggesting the oncogenic potential of the reported miR-5p/-3p pairs. Specifically, we revealed that the reduced expression of miR-145-5p/-3p pair potentially contributes to elevated expression of genes in the "FOXM1 transcription factor network" pathway, which may consequently lead to uncontrolled cell proliferation. Subsequently, the regulation of miR-145-5p/-3p in the FOXM1signaling pathway was validated by a cohort of 104 matched miRNA and protein (reverse-phase protein array) expression profiles in lung cancer. In summary, our computational framework provides a novel tool to study miR-5p/-3p coregulatory mechanisms in cancer and other diseases.