Prediction of lncRNAs and their interactions with nucleic acids: benchmarking bioinformatics tools

Long noncoding RNA study: Genome-wide approaches

Long noncoding RNAs (lncRNAs) have been confirmed to play a crucial role in various biological processes across several species. Though many efforts have been devoted to the expansion of the lncRNAs landscape, much about lncRNAs is still unknown due to their great complexity. The development of high-throughput technologies and the constantly improved bioinformatic methods have resulted in a rapid expansion of lncRNA research and relevant databases. In this review, we introduced genome-wide research of lncRNAs in three parts: (i) novel lncRNA identification by high-throughput sequencing and computational pipelines; (ii) functional characterization of lncRNAs by expression atlas profiling, genome-scale screening, and the research of cancer-related lncRNAs; (iii) mechanism research by large-scale experimental technologies and computational analysis. Besides, primary experimental methods and bioinformatic pipelines related to these three parts are summarized. This review aimed to provide a comprehensive and systemic overview of lncRNA genome-wide research strategies and indicate a genome-wide lncRNA research system.

EphrinA5 regulates cell motility by modulating Snhg15/DNA triplex-dependent targeting of DNMT1 to the Ncam1 promoter

Cell-cell communication is mediated by membrane receptors and their ligands, such as the Eph/ephrin system, orchestrating cell migration during development and in diverse cancer types. Epigenetic mechanisms are key for integrating external "signals", e.g., from neighboring cells, into the transcriptome in health and disease. Previously, we reported ephrinA5 to trigger transcriptional changes of lncRNAs and protein-coding genes in cerebellar granule cells, a cell model for medulloblastoma. LncRNAs represent important adaptors for epigenetic writers through which they regulate gene expression. Here, we investigate a lncRNA-mediated targeting of DNMT1 to specific gene loci by the combined power of in silico modeling of RNA/DNA interactions and wet lab approaches, in the context of the clinically relevant use case of ephrinA5-dependent regulation of cellular motility of cerebellar granule cells. We provide evidence that Snhg15, a cancer-related lncRNA, recruits DNMT1 to the Ncam1 promoter through RNA/DNA triplex structure formation and the interaction with DNMT1. This mediates DNA methylation-dependent silencing of Ncam1, being abolished by ephrinA5 stimulation-triggered reduction of Snhg15 expression. Hence, we here propose a triple helix recognition mechanism, underlying cell motility regulation via lncRNA-targeted DNA methylation in a clinically relevant context.

Pan-cancer survey of lncRNA rewiring and functional alternation in tumor-infiltrating T cell by scLNC

Long non-coding RNAs (lncRNAs) have been reported to involve in diverse biological processes, including tumor immunity. Since lncRNAs are expressed with high cell-type specificity, investigation of lncRNAs at the single-cell level will unveil the cell-type-specific functions of lncRNAs. However, at the single-cell level, a systematic pan-cancer analysis of lncRNA functions in tumor immune microenvironments (TIMEs) remains lacking. Here, we performed pan-cancer single-cell profiling of lncRNA functions in TIMEs and developed a tool, scLNC, tailored for lncRNA functional characterization at the single-cell level. scLNC enabled the comparison of lncRNA function from the levels of lncRNA-mRNA pairs, lncRNA regulatory unit activity and unit function in a cell-type-specific manner. Applying scLNC, our analysis depicted the cross-tumor and tumor-specific lncRNA regulatory profiles in the T cell subtypes and revealed the new regulatory units that lncRNAs established in tumor-infiltrating T cells, particularly in the tumor-enriched T cells. We further characterized the activity and functional alternations of lncRNAs through their regulatory units. Overall, our findings suggested that lncRNAs played an important role in the regulation of cytokine production, cell activation and migration in tumor-enriched T cells and further in immunotherapy.

PmliHFM: Predicting Plant miRNA-lncRNA Interactions with Hybrid Feature Mining Network

Due to the crucial role of interactions between microRNAs (miRNAs) and long non-coding RNAs (lncRNAs) in biological processes, the study of their biological functions is necessary. So far, the various computational methods have been employed to make predictions of the miRNA-lncRNA interaction, which compensate for the inadequacy of biological experiments. However, the existing methods do not consider the differences between miRNA and lncRNA in feature extraction. In this paper, we propose a hybrid feature mining network, named PmliHFM, for predicting plant miRNA-lncRNA interactions. Firstly, miRNA and lncRNA with different sequence lengths are encoded by different encodings, which can reduce the loss of information caused by using the same coding approach. Then, a hybrid feature mining network is designed to adapt to different encoding methods and extract more useful feature information than a single network. Finally, an ensemble module is utilized to integrate the training results of the hybrid feature mining network, while a prediction module is employed to determine whether there are interactions. By testing on multiple test sets, PmliHFM outperforms several state-of-the-art approaches. The results show that the AUC of PmliHFM achieves 0.8[Formula: see text], 3.1[Formula: see text] and 0.4[Formula: see text] improvement respectively on three balanced datasets, and achieves 2.1[Formula: see text] and 1.8[Formula: see text] improvement respectively on two imbalanced datasets. These experiments demonstrate the feasibility of the proposed method.

Recent Development in Biomedical Applications of Oligonucleotides with Triplex-Forming Ability

A DNA structure, known as triple-stranded DNA, is made up of three oligonucleotide chains that wind around one another to form a triple helix (TFO). Hoogsteen base pairing describes how triple-stranded DNA may be built at certain conditions by the attachment of the third strand to an RNA, PNA, or DNA, which might all be employed as oligonucleotide chains. In each of these situations, the oligonucleotides can be employed as an anchor, in conjunction with a specific bioactive chemical, or as a messenger that enables switching between transcription and replication through the triplex-forming zone. These data are also considered since various illnesses have been linked to the expansion of triplex-prone sequences. In light of metabolic acidosis and associated symptoms, some consideration is given to the impact of several low-molecular-weight compounds, including pH on triplex production in vivo. The review is focused on the development of biomedical oligonucleotides with triplexes.

Fast RNA-RNA Interaction Prediction Methods for Interaction Analysis of Transcriptome-Scale Large Datasets

The computational prediction of RNA-RNA interactions has long been studied in RNA informatics. Most of the existing approaches focused on the interaction prediction of short RNAs in small datasets. However, in recent years, two fast prediction methods, RIsearch2 and RIblast, have been developed to predict transcriptome-scale interactions or long RNA interactions. The key idea of the software acceleration of these tools was the integration of a seed-and-extend method, which is used in fast sequence alignment tools, into RNA-RNA interaction prediction. As a result, the two software programs were ten to a thousand times faster than the existing tools; because of this acceleration, detection of genome-wide microRNA target sites or interaction partners of function-unknown long noncoding RNAs has become possible. In this review, we describe the basic concept of the algorithm, its applications, and the future perspectives of the fast RNA-RNA interaction prediction tools.

Computational Methods to Study DNA:DNA:RNA Triplex Formation by lncRNAs

Long non-coding RNAs (lncRNAs) impact cell function via numerous mechanisms. In the nucleus, interactions between lncRNAs and DNA and the consequent formation of non-canonical nucleic acid structures seems to be particularly relevant. Along with interactions between single-stranded RNA (ssRNA) and single-stranded DNA (ssDNA), such as R-loops, ssRNA can also interact with double-stranded DNA (dsDNA) to form DNA:DNA:RNA triplexes. A major challenge in the study of DNA:DNA:RNA triplexes is the identification of the precise RNA component interacting with specific regions of the dsDNA. As this is a crucial step towards understanding lncRNA function, there exist several computational methods designed to predict these sequences. This review summarises the recent progress in the prediction of triplex formation and highlights important DNA:DNA:RNA triplexes. In particular, different prediction tools (Triplexator, LongTarget, TRIPLEXES, Triplex Domain Finder, TriplexFFP, TriplexAligner and Fasim-LongTarget) will be discussed and their use exemplified by selected lncRNAs, whose DNA:DNA:RNA triplex forming potential was validated experimentally. Collectively, these tools revealed that DNA:DNA:RNA triplexes are likely to be numerous and make important contributions to gene expression regulation.

A universal model of RNA.DNA:DNA triplex formation accurately predicts genome-wide RNA-DNA interactions

RNA.DNA:DNA triple helix (triplex) formation is a form of RNA-DNA interaction which regulates gene expression but is difficult to study experimentally in vivo. This makes accurate computational prediction of such interactions highly important in the field of RNA research. Current predictive methods use canonical Hoogsteen base pairing rules, which whilst biophysically valid, may not reflect the plastic nature of cell biology. Here, we present the first optimization approach to learn a probabilistic model describing RNA-DNA interactions directly from motifs derived from triplex sequencing data. We find that there are several stable interaction codes, including Hoogsteen base pairing and novel RNA-DNA base pairings, which agree with in vitro measurements. We implemented these findings in TriplexAligner, a program that uses the determined interaction codes to predict triplex binding. TriplexAligner predicts RNA-DNA interactions identified in all-to-all sequencing data more accurately than all previously published tools in human and mouse and also predicts previously studied triplex interactions with known regulatory functions. We further validated a novel triplex interaction using biophysical experiments. Our work is an important step towards better understanding of triplex formation and allows genome-wide analyses of RNA-DNA interactions.

Deep learning tools are top performers in long non-coding RNA prediction

The increasing amount of transcriptomic data has brought to light vast numbers of potential novel RNA transcripts. Accurately distinguishing novel long non-coding RNAs (lncRNAs) from protein-coding messenger RNAs (mRNAs) has challenged bioinformatic tool developers. Most recently, tools implementing deep learning architectures have been developed for this task, with the potential of discovering sequence features and their interactions still not surfaced in current knowledge. We compared the performance of deep learning tools with other predictive tools that are currently used in lncRNA coding potential prediction. A total of 15 tools representing the variety of available methods were investigated. In addition to known annotated transcripts, we also evaluated the use of the tools in actual studies with real-life data. The robustness and scalability of the tools' performance was tested with varying sized test sets and test sets with different proportions of lncRNAs and mRNAs. In addition, the ease-of-use for each tested tool was scored. Deep learning tools were top performers in most metrics and labelled transcripts similarly with each other in the real-life dataset. However, the proportion of lncRNAs and mRNAs in the test sets affected the performance of all tools. Computational resources were utilized differently between the top-ranking tools, thus the nature of the study may affect the decision of choosing one well-performing tool over another. Nonetheless, the results suggest favouring the novel deep learning tools over other tools currently in broad use.

Identification of Long Noncoding RNAs Involved in Eyelid Pigmentation of Hereford Cattle

Several ocular pathologies in cattle, such as ocular squamous cell carcinoma and infectious keratoconjunctivitis, have been associated with low pigmentation of the eyelids. The main objective of this study was to analyze the transcriptome of eyelid skin in Hereford cattle using strand-specific RNA sequencing technology to characterize and identify long noncoding RNAs (lncRNAs). We compared the expression of lncRNAs between pigmented and unpigmented eyelids and analyzed the interaction of lncRNAs and putative target genes to reveal the genetic basis underlying eyelid pigmentation in cattle. We predicted 4,937 putative lncRNAs mapped to the bovine reference genome, enriching the catalog of lncRNAs in Bos taurus. We found 27 differentially expressed lncRNAs between pigmented and unpigmented eyelids, suggesting their involvement in eyelid pigmentation. In addition, we revealed potential links between some significant differentially expressed lncRNAs and target mRNAs involved in the immune response and pigmentation. Overall, this study expands the catalog of lncRNAs in cattle and contributes to a better understanding of the biology of eyelid pigmentation.

RNAI-FRID: novel feature representation method with information enhancement and dimension reduction for RNA-RNA interaction

Different ribonucleic acids (RNAs) can interact to form regulatory networks that play important role in many life activities. Molecular biology experiments can confirm RNA-RNA interactions to facilitate the exploration of their biological functions, but they are expensive and time-consuming. Machine learning models can predict potential RNA-RNA interactions, which provide candidates for molecular biology experiments to save a lot of time and cost. Using a set of suitable features to represent the sample is crucial for training powerful models, but there is a lack of effective feature representation for RNA-RNA interaction. This study proposes a novel feature representation method with information enhancement and dimension reduction for RNA-RNA interaction (named RNAI-FRID). Diverse base features are first extracted from RNA data to contain more sample information. Then, the extracted base features are used to construct the complex features through an arithmetic-level method. It greatly reduces the feature dimension while keeping the relationship between molecule features. Since the dimension reduction may cause information loss, in the process of complex feature construction, the arithmetic mean strategy is adopted to enhance the sample information further. Finally, three feature ranking methods are integrated for feature selection on constructed complex features. It can adaptively retain important features and remove redundant ones. Extensive experiment results show that RNAI-FRID can provide reliable feature representation for RNA-RNA interaction with higher efficiency and the model trained with generated features obtain better performance than other deep neural network predictors.

Genome-wide regulation of CpG methylation by ecCEBPα in acute myeloid leukemia

Background: Acute myeloid leukemia (AML) is a hematopoietic malignancy characterized by genetic and epigenetic aberrations that alter the differentiation capacity of myeloid progenitor cells. The transcription factor CEBPα is frequently mutated in AML patients leading to an increase in DNA methylation in many genomic locations. Previously, it has been shown that ecCEBPα (extra coding CEBP α) - a lncRNA transcribed in the same direction as CEBPα gene - regulates DNA methylation of CEBPα promoter in cis. Here, we hypothesize that ecCEBPα could participate in the regulation of DNA methylation in trans.

Method: First, we retrieved the methylation profile of AML patients with mutated CEBPα locus from The Cancer Genome Atlas (TCGA). We then predicted the ecCEBPα secondary structure in order to check the potential of ecCEBPα to form triplexes around CpG loci and checked if triplex formation influenced CpG methylation, genome-wide.

Results: Using DNA methylation profiles of AML patients with a mutated CEBPα locus, we show that ecCEBPα could interact with DNA by forming DNA:RNA triple helices and protect regions near its binding sites from global DNA methylation. Further analysis revealed that triplex-forming oligonucleotides in ecCEBPα are structurally unpaired supporting the DNA-binding potential of these regions. ecCEBPα triplexes supported with the RNA-chromatin co-localization data are located in the promoters of leukemia-linked transcriptional factors such as MLF2.

Discussion: Overall, these results suggest a novel regulatory mechanism for ecCEBPα as a genome-wide epigenetic modulator through triple-helix formation which may provide a foundation for sequence-specific engineering of RNA for regulating methylation of specific genes.

A comprehensive method protocol for annotation and integrated functional understanding of lncRNAs

Long non-coding RNAs (lncRNAs) are of fundamental biological importance; however, their functional role is often unclear or loosely defined as experimental characterization is challenging and bioinformatic methods are limited. We developed a novel integrated method protocol for the annotation and detailed functional characterization of lncRNAs within the genome. It combines annotation, normalization and gene expression with sequence-structure conservation, functional interactome and promoter analysis. Our protocol allows an analysis based on the tissue and biological context, and is powerful in functional characterization of experimental and clinical RNA-Seq datasets including existing lncRNAs. This is demonstrated on the uncharacterized lncRNA GATA6-AS1 in dilated cardiomyopathy.

Computational analysis of sense-antisense chimeric transcripts reveals their potential regulatory features and the landscape of expression in human cells

Many human genes are transcribed from both strands and produce sense-antisense gene pairs. Sense-antisense (SAS) chimeric transcripts are produced upon the coalescing of exons/introns from both sense and antisense transcripts of the same gene. SAS chimera was first reported in prostate cancer cells. Subsequently, numerous SAS chimeras have been reported in the ChiTaRS-2.1 database. However, the landscape of their expression in human cells and functional aspects are still unknown. We found that longer palindromic sequences are a unique feature of SAS chimeras. Structural analysis indicates that a long hairpin-like structure formed by many consecutive Watson-Crick base pairs appears because of these long palindromic sequences, which possibly play a similar role as double-stranded RNA (dsRNA), interfering with gene expression. RNA-RNA interaction analysis suggested that SAS chimeras could significantly interact with their parental mRNAs, indicating their potential regulatory features. Here, 267 SAS chimeras were mapped in RNA-seq data from 16 healthy human tissues, revealing their expression in normal cells. Evolutionary analysis suggested the positive selection favoring sense-antisense fusions that significantly impacted the evolution of their function and structure. Overall, our study provides detailed insight into the expression landscape of SAS chimeras in human cells and identifies potential regulatory features.

LncRNAs as Therapeutic Targets and Potential Biomarkers for Lipid-Related Diseases

Lipid metabolism is an essential biological process involved in nutrient adjustment, hormone regulation, and lipid homeostasis. An irregular lifestyle and long-term nutrient overload can cause lipid-related diseases, including atherosclerosis, myocardial infarction (MI), obesity, and fatty liver diseases. Thus, novel tools for efficient diagnosis and treatment of dysfunctional lipid metabolism are urgently required. Furthermore, it is known that lncRNAs based regulation like sponging microRNAs (miRNAs) or serving as a reservoir for microRNAs play an essential role in the progression of lipid-related diseases. Accordingly, a better understanding of the regulatory roles of lncRNAs in lipid-related diseases would provide the basis for identifying potential biomarkers and therapeutic targets for lipid-related diseases. This review highlighted the latest advances on the potential biomarkers of lncRNAs in lipid-related diseases and summarised current knowledge on dysregulated lncRNAs and their potential molecular mechanisms. We have also provided novel insights into the underlying mechanisms of lncRNAs which might serve as potential biomarkers and therapeutic targets for lipid-related diseases. The information presented here may be useful for designing future studies and advancing investigations of lncRNAs as biomarkers for diagnosis, prognosis, and therapy of lipid-related diseases.

From Micro to Long: Non-Coding RNAs in Tamoxifen Resistance of Breast Cancer Cells

Breast cancer is the most commonly diagnosed cancer and the leading cause of cancer mortality among women. Two thirds of patients are classified as hormone receptor positive, based on expression of estrogen receptor alpha (ERα), the main driver of breast cancer cell proliferation, and/or progesterone receptor, which is regulated by ERα. Despite presenting the best prognosis, these tumors can recur when patients acquire resistance to treatment by aromatase inhibitors or antiestrogen such as tamoxifen (Tam). The mechanisms that are involved in Tam resistance are complex and involve multiple signaling pathways. Recently, roles for microRNAs and lncRNAs in controlling ER expression and/or tamoxifen action have been described, but the underlying mechanisms are still little explored. In this review, we will discuss the current state of knowledge on the roles of microRNAs and lncRNAs in the main mechanisms of tamoxifen resistance in hormone receptor positive breast cancer. In the future, this knowledge can be used to identify patients at a greater risk of relapse due to the expression patterns of ncRNAs that impact response to Tam, in order to guide their treatment more efficiently and possibly to design therapeutic strategies to bypass mechanisms of resistance.

Ensemble Deep Learning Based on Multi-level Information Enhancement and Greedy Fuzzy Decision for Plant miRNA-lncRNA Interaction Prediction

MicroRNAs (miRNAs) and long non-coding RNAs (lncRNAs) are both non-coding RNAs (ncRNAs) and their interactions play important roles in biological processes. Computational methods, such as machine learning and various bioinformatics tools, can predict potential miRNA-lncRNA interactions, which is significant for studying their mechanisms and biological functions. A growing number of RNA interaction predictors for animal have been reported, but they are unreliable for plant due to the differences of ncRNAs in animal and plant. It is urgent to build a reliable plant predictor, especially for cross-species. This paper proposes an ensemble deep learning model based on multi-level information enhancement and greedy fuzzy decision (PmliPEMG) for plant miRNA-lncRNA interaction prediction. The fusion complex features, multi-scale convolutional long short-term memory networks, and attention mechanism are adopted to enhance the sample information at the feature, scale, and model levels, respectively. An ensemble deep learning model is built based on a novel method (greedy fuzzy decision) which greatly improves the efficiency. The multi-level information enhancement and greedy fuzzy decision are verified to have the positive effects on prediction performance. PmliPEMG can be applied to the cross-species prediction. It shows better performance and stronger generalization ability than state-of-the-art predictors and may provide valuable references for related research.

Computational Analysis Predicts Hundreds of Coding lncRNAs in Zebrafish

Simple Summary

Noncoding RNAs (ncRNAs) regulate a variety of fundamental life processes such as development, physiology, metabolism and circadian rhythmicity. RNA-sequencing (RNA-seq) technology has facilitated the sequencing of the whole transcriptome, thereby capturing and quantifying the dynamism of transcriptome-wide RNA expression profiles. However, much remains unrevealed in the huge noncoding RNA datasets that require further bioinformatic analysis. In this study, we applied six bioinformatic tools to investigate coding potentials of approximately 21,000 lncRNAs. A total of 313 lncRNAs are predicted to be coded by all the six tools. Our findings provide insights into the regulatory roles of lncRNAs and set the stage for the functional investigation of these lncRNAs and their encoded micropeptides.

Abstract

Recent studies have demonstrated that numerous long noncoding RNAs (ncRNAs having more than 200 nucleotide base pairs (lncRNAs)) actually encode functional micropeptides, which likely represents the next regulatory biology frontier. Thus, identification of coding lncRNAs from ever-increasing lncRNA databases would be a bioinformatic challenge. Here we employed the Coding Potential Alignment Tool (CPAT), Coding Potential Calculator 2 (CPC2), LGC web server, Coding-Non-Coding Identifying Tool (CNIT), RNAsamba, and MicroPeptide identification tool (MiPepid) to analyze approximately 21,000 zebrafish lncRNAs and computationally to identify 2730–6676 zebrafish lncRNAs with high coding potentials, including 313 coding lncRNAs predicted by all the six bioinformatic tools. We also compared the sensitivity and specificity of these six bioinformatic tools for identifying lncRNAs with coding potentials and summarized their strengths and weaknesses. These predicted zebrafish coding lncRNAs set the stage for further experimental studies.

Approaches to Identify and Characterise the Post-Transcriptional Roles of lncRNAs in Cancer

It is becoming increasingly evident that the non-coding genome and transcriptome exert great influence over their coding counterparts through complex molecular interactions. Among non-coding RNAs (ncRNA), long non-coding RNAs (lncRNAs) in particular present increased potential to participate in dysregulation of post-transcriptional processes through both RNA and protein interactions. Since such processes can play key roles in contributing to cancer progression, it is desirable to continue expanding the search for lncRNAs impacting cancer through post-transcriptional mechanisms. The sheer diversity of mechanisms requires diverse resources and methods that have been developed and refined over the past decade. We provide an overview of computational resources as well as proven low-to-high throughput techniques to enable identification and characterisation of lncRNAs in their complex interactive contexts. As more cancer research strategies evolve to explore the non-coding genome and transcriptome, we anticipate this will provide a valuable primer and perspective of how these technologies have matured and will continue to evolve to assist researchers in elucidating post-transcriptional roles of lncRNAs in cancer.

The lncRNA Toolkit: Databases and In Silico Tools for lncRNA Analysis

Long non-coding RNAs (lncRNAs) are a rapidly expanding field of research, with many new transcripts identified each year. However, only a small subset of lncRNAs has been characterized functionally thus far. To aid investigating the mechanisms of action by which new lncRNAs act, bioinformatic tools and databases are invaluable. Here, we review a selection of computational tools and databases for the in silico analysis of lncRNAs, including tissue-specific expression, protein coding potential, subcellular localization, structural conformation, and interaction partners. The assembled lncRNA toolkit is aimed primarily at experimental researchers as a useful starting point to guide wet-lab experiments, mainly containing multi-functional, user-friendly interfaces. With more and more new lncRNA analysis tools available, it will be essential to provide continuous updates and maintain the availability of key software in the future.

Direct Interactions with Nascent Transcripts Is Potentially a Common Targeting Mechanism of Long Non-Coding RNAs

Although thousands of mammalian long non-coding RNAs (lncRNAs) have been reported in the last decade, their functional annotation remains limited. A wet-lab approach to detect functions of a novel lncRNA usually includes its knockdown followed by RNA sequencing and identification of the deferentially expressed genes. However, identification of the molecular mechanism(s) used by the lncRNA to regulate its targets frequently becomes a challenge. Previously, we developed the ASSA algorithm that detects statistically significant inter-molecular RNA-RNA interactions. Here we designed a workflow that uses ASSA predictions to estimate the ability of an lncRNA to function via direct base pairing with the target transcripts (co- or post-transcriptionally). The workflow was applied to 300+ lncRNA knockdown experiments from the FANTOM6 pilot project producing statistically significant predictions for 71 unique lncRNAs (104 knockdowns). Surprisingly, the majority of these lncRNAs were likely to function co-transcriptionally, i.e., hybridize with the nascent transcripts of the target genes. Moreover, a number of the obtained predictions were supported by independent iMARGI experimental data on co-localization of lncRNA and chromatin. We detected an evolutionarily conserved lncRNA CHASERR (AC013394.2 or LINC01578) that could regulate target genes co-transcriptionally via interaction with a nascent transcript by directing CHD2 helicase. The obtained results suggested that this nuclear lncRNA may be able to activate expression of the target genes in trans by base-pairing with the nascent transcripts and directing the CHD2 helicase to the regulated promoters leading to open the chromatin and active transcription. Our study highlights the possible importance of base-pairing between nuclear lncRNAs and nascent transcripts for the regulation of gene expression.

Deep Learning in LncRNAome: Contribution, Challenges, and Perspectives

Long non-coding RNAs (lncRNA), the pervasively transcribed part of the mammalian genome, have played a significant role in changing our protein-centric view of genomes. The abundance of lncRNAs and their diverse roles across cell types have opened numerous avenues for the research community regarding lncRNAome. To discover and understand lncRNAome, many sophisticated computational techniques have been leveraged. Recently, deep learning (DL)-based modeling techniques have been successfully used in genomics due to their capacity to handle large amounts of data and produce relatively better results than traditional machine learning (ML) models. DL-based modeling techniques have now become a choice for many modeling tasks in the field of lncRNAome as well. In this review article, we summarized the contribution of DL-based methods in nine different lncRNAome research areas. We also outlined DL-based techniques leveraged in lncRNAome, highlighting the challenges computational scientists face while developing DL-based models for lncRNAome. To the best of our knowledge, this is the first review article that summarizes the role of DL-based techniques in multiple areas of lncRNAome.

Spectrochemical Probing of MicroRNA Duplex Using Spontaneous Raman Spectroscopy for Biosensing Applications

MicroRNAs are emerging as both diagnostic and therapeutic targets in different human pathologies. An accurate understanding of the structural dependency of microRNAs for their biological functions is essential for designing synthetic oligos with various base and linkage modifications that can transform into highly sensitive diagnostic devices and therapeutic molecules. In this proof-of-principle study, we have utilized label-free spontaneous Raman spectroscopy to understand the structural differences in sense and antisense microRNA-21 by hybridizing them with complementary RNA and DNA oligos. Overall, the results suggest that the changes in the Raman band at 785 cm^-1 originating from the phosphodiester bond of the nucleic acid backbone, linking 5' phosphate of the nucleic acid with 3' OH of the other nucleotide, can serve as a marker to identify these structural variations. Our results support the application of Raman spectroscopy in discerning intramolecular (ssRNA and ssDNA) and intermolecular (RNA-RNA, RNA-DNA, and DNA-DNA hybrids) interactions of nucleic acids. This is potentially useful for developing biosensors to quantify microRNAs in clinical samples and to design therapeutic microRNAs with robust functionality.

PmliPred: a method based on hybrid model and fuzzy decision for plant miRNA-lncRNA interaction prediction

MOTIVATION

The studies have indicated that not only microRNAs (miRNAs) or long non-coding RNAs (lncRNAs) play important roles in biological activities, but also their interactions affect the biological process. A growing number of studies focus on the miRNA-lncRNA interactions, while few of them are proposed for plant. The prediction of interactions is significant for understanding the mechanism of interaction between miRNA and lncRNA in plant.

RESULTS

This article proposes a new method for fulfilling plant miRNA-lncRNA interaction prediction (PmliPred). The deep learning model and shallow machine learning model are trained using raw sequence and manually extracted features, respectively. Then they are hybridized based on fuzzy decision for prediction. PmliPred shows better performance and generalization ability compared with the existing methods. Several new miRNA-lncRNA interactions in Solanum lycopersicum are successfully identified using quantitative real time-polymerase chain reaction from the candidates predicted by PmliPred, which further verifies its effectiveness.

AVAILABILITY AND IMPLEMENTATION

The source code of PmliPred is freely available at http://bis.zju.edu.cn/PmliPred/.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

A Spring Search Algorithm Applied to Engineering Optimization Problems

At present, optimization algorithms are used extensively. One particular type of such algorithms includes random-based heuristic population optimization algorithms, which may be created by modeling scientific phenomena, like, for example, physical processes. The present article proposes a novel optimization algorithm based on Hooke’s law, called the spring search algorithm (SSA), which aims to solve single-objective constrained optimization problems. In the SSA, search agents are weights joined through springs, which, as Hooke’s law states, possess a force that corresponds to its length. The mathematics behind the algorithm are presented in the text. In order to test its functionality, it is executed on 38 established benchmark test functions and weighed against eight other optimization algorithms: a genetic algorithm (GA), a gravitational search algorithm (GSA), a grasshopper optimization algorithm (GOA), particle swarm optimization (PSO), teaching–learning-based optimization (TLBO), a grey wolf optimizer (GWO), a spotted hyena optimizer (SHO), as well as an emperor penguin optimizer (EPO). To test the SSA’s usability, it is employed on five engineering optimization problems. The SSA delivered better fitting results than the other algorithms in unimodal objective function, multimodal objective functions, CEC 2015, in addition to the optimization problems in engineering.

Practical Guidance in Genome-Wide RNA:DNA Triple Helix Prediction

Long noncoding RNAs (lncRNAs) play a key role in many cellular processes including chromatin regulation. To modify chromatin, lncRNAs often interact with DNA in a sequence-specific manner forming RNA:DNA triple helices. Computational tools for triple helix search do not always provide genome-wide predictions of sufficient quality. Here, we used four human lncRNAs (MEG3, DACOR1, TERC and HOTAIR) and their experimentally determined binding regions for evaluating triplex parameters that provide the highest prediction accuracy. Additionally, we combined triplex prediction with the lncRNA secondary structure and demonstrated that considering only single-stranded fragments of lncRNA can further improve DNA-RNA triplexes prediction.

Detection of RNA-DNA binding sites in long noncoding RNAs

Long non-coding RNAs (lncRNAs) can act as scaffolds that promote the interaction of proteins, RNA, and DNA. There is increasing evidence of sequence-specific interactions of lncRNAs with DNA via triple-helix (triplex) formation. This process allows lncRNAs to recruit protein complexes to specific genomic regions and regulate gene expression. Here we propose a computational method called Triplex Domain Finder (TDF) to detect triplexes and characterize DNA-binding domains and DNA targets statistically. Case studies showed that this approach can detect the known domains of lncRNAs Fendrr, HOTAIR and MEG3. Moreover, we validated a novel DNA-binding domain in MEG3 by a genome-wide sequencing method. We used TDF to perform a systematic analysis of the triplex-forming potential of lncRNAs relevant to human cardiac differentiation. We demonstrated that the lncRNA with the highest triplex-forming potential, GATA6-AS, forms triple helices in the promoter of genes relevant to cardiac development. Moreover, down-regulation of GATA6-AS impairs GATA6 expression and cardiac development. These data indicate the unique ability of our computational tool to identify novel triplex-forming lncRNAs and their target genes.

Long non-coding RNAs identify a subset of luminal muscle-invasive bladder cancer patients with favorable prognosis

BACKGROUND

Muscle-invasive bladder cancer (MIBC) is a heterogeneous disease, and gene expression profiling has identified several molecular subtypes with distinct biological and clinicopathological characteristics. While MIBC subtyping has primarily been based on messenger RNA (mRNA), long non-coding RNAs (lncRNAs) may provide additional resolution.

METHODS

LncRNA expression was quantified from microarray data of a MIBC cohort treated with neoadjuvant chemotherapy (NAC) and radical cystectomy (RC) (n = 223). Unsupervised consensus clustering of highly variant lncRNAs identified a four-cluster solution, which was characterized using a panel of MIBC biomarkers, regulon activity profiles, gene signatures, and survival analysis. The four-cluster solution was confirmed in The Cancer Genome Atlas (TCGA) cohort (n = 405). A single-sample genomic classifier (GC) was trained using ridge-penalized logistic regression and validated in two independent cohorts (n = 255 and n = 94).

RESULTS

NAC and TCGA cohorts both contained an lncRNA cluster (LC3) with favorable prognosis that was enriched with tumors of the luminal-papillary (LP) subtype. In both cohorts, patients with LP tumors in LC3 (LPL-C3) were younger and had organ-confined, node-negative disease. The LPL-C3 tumors had enhanced FGFR3, SHH, and wild-type p53 pathway activity. In the TCGA cohort, LPL-C3 tumors were enriched for FGFR3 mutations and depleted for TP53 and RB1 mutations. A GC trained to identify these LPL-C3 patients showed robust performance in two validation cohorts.

CONCLUSIONS

Using lncRNA expression profiles, we identified a biologically distinct subgroup of luminal-papillary MIBC with a favorable prognosis. These data suggest that lncRNAs provide additional information for higher-resolution subtyping, potentially improving precision patient management.

Gene set analysis methods for the functional interpretation of non-mRNA data—Genomic range and ncRNA data

Gene set analysis (GSA) is one of the methods of choice for analyzing the results of current omics studies; however, it has been mainly developed to analyze mRNA (microarray, RNA-Seq) data. The following review includes an update regarding general methods and resources for GSA and then emphasizes GSA methods and tools for non-mRNA omics datasets, specifically genomic range data (ChIP-Seq, SNP and methylation) and ncRNA data (miRNAs, lncRNAs and others). In the end, the state of the GSA field for non-mRNA datasets is discussed, and some current challenges and trends are highlighted, especially the use of network approaches to face complexity issues.

Long Non-Coding RNA in the Pathogenesis of Cancers

The incidence and mortality rate of cancer has been quickly increasing in the past decades. At present, cancer has become the leading cause of death worldwide. Most of the cancers cannot be effectively diagnosed at the early stage. Although there are multiple therapeutic treatments, including surgery, radiotherapy, chemotherapy, and targeted drugs, their effectiveness is still limited. The overall survival rate of malignant cancers is still low. It is necessary to further study the mechanisms for malignant cancers, and explore new biomarkers and targets that are more sensitive and effective for early diagnosis, treatment, and prognosis of cancers than traditional biomarkers and methods. Long non-coding RNAs (lncRNAs) are a class of RNA transcripts with a length greater than 200 nucleotides. Generally, lncRNAs are not capable of encoding proteins or peptides. LncRNAs exert diverse biological functions by regulating gene expressions and functions at transcriptional, translational, and post-translational levels. In the past decade, it has been demonstrated that the dysregulated lncRNA profile is widely involved in the pathogenesis of many diseases, including cancer, metabolic disorders, and cardiovascular diseases. In particular, lncRNAs have been revealed to play an important role in tumor growth and metastasis. Many lncRNAs have been shown to be potential biomarkers and targets for the diagnosis and treatment of cancers. This review aims to briefly discuss the latest findings regarding the roles and mechanisms of some important lncRNAs in the pathogenesis of certain malignant cancers, including lung, breast, liver, and colorectal cancers, as well as hematological malignancies and neuroblastoma.

Interactions among lncRNAs, miRNAs and mRNA in colorectal cancer

Long non-coding RNAs (lncRNAs) are longer than 200 nts non-coding transcripts and have recently emerged as one of the largest and significantly diverse RNA families whereas microRNAs (miRNAs) are highly conserved short single-stranded ncRNAs (∼18-22 nucleotides). As families of small and long evolutionarily conserved ncRNAs, lncRNAs activate and repress genes via a variety of mechanisms at both transcriptional and translational levels, while miRNAs regulate protein-coding gene expression mainly through mRNA degradation or silencing, These ncRNAs have been proved to be involved in multiple biological functions, such as proliferation, differentiation, migration, angiogenesis and apoptosis. Today, while majority of studies have focused on defining the regulatory functions of lncRNAs and miRNAs, limited information have now available for the mutual regulations of lncRNAs, miRNAs and mRNA. Thus, the underlying molecular mechanisms, in particularly the interactions among lncRNAs, miRNAs and mRNA in development, growth, metastasis and therapeutic potential of cancer still remain obscure. Colorectal cancer (CRC) is known as the third most common and fourth leading cancer death worldwide. Increasing evidence showed the close correlations among aberrant expressions of lncRNAs, miRNAs and the occurrence, development of CRC. This review summarize the potential links among these RNAs in following three areas: 1, The biogenesis and roles of miRNAs in CRC; 2, The biogenesis and functions of lncRNAs in CRC; 3, The interactions among lncRNAs, miRNAs and mRNA in tumorigensis, growth, progression, EMT formation, chemoradiotherapy resistance, and therapeutic potential in CRC. We believe that identifying diverging lncRNAs, miRNAs and relevant genes, their interactions and complex molecular regulatory networks will provide important clues for understanding the mechanism and developing novel diagnostic and therapeutic strategies for CRC. Further efforts are warranted to bring the promise of regulating their activities into clinical utilities.

PLAIDOH: a novel method for functional prediction of long non-coding RNAs identifies cancer-specific LncRNA activities

BACKGROUND

Long non-coding RNAs (lncRNAs) exhibit remarkable cell-type specificity and disease association. LncRNA's functional versatility includes epigenetic modification, nuclear domain organization, transcriptional control, regulation of RNA splicing and translation, and modulation of protein activity. However, most lncRNAs remain uncharacterized due to a shortage of predictive tools available to guide functional experiments.

RESULTS

To address this gap for lymphoma-associated lncRNAs identified in our studies, we developed a new computational method, Predicting LncRNA Activity through Integrative Data-driven 'Omics and Heuristics (PLAIDOH), which has several unique features not found in other methods. PLAIDOH integrates transcriptome, subcellular localization, enhancer landscape, genome architecture, chromatin interaction, and RNA-binding (eCLIP) data and generates statistically defined output scores. PLAIDOH's approach identifies and ranks functional connections between individual lncRNA, coding gene, and protein pairs using enhancer, transcript cis-regulatory, and RNA-binding protein interactome scores that predict the relative likelihood of these different lncRNA functions. When applied to 'omics datasets that we collected from lymphoma patients, or to publicly available cancer (TCGA) or ENCODE datasets, PLAIDOH identified and prioritized well-known lncRNA-target gene regulatory pairs (e.g., HOTAIR and HOX genes, PVT1 and MYC), validated hits in multiple lncRNA-targeted CRISPR screens, and lncRNA-protein binding partners (e.g., NEAT1 and NONO). Importantly, PLAIDOH also identified novel putative functional interactions, including one lymphoma-associated lncRNA based on analysis of data from our human lymphoma study. We validated PLAIDOH's predictions for this lncRNA using knock-down and knock-out experiments in lymphoma cell models.

CONCLUSIONS

Our study demonstrates that we have developed a new method for the prediction and ranking of functional connections between individual lncRNA, coding gene, and protein pairs, which were validated by genetic experiments and comparison to published CRISPR screens. PLAIDOH expedites validation and follow-on mechanistic studies of lncRNAs in any biological system. It is available at https://github.com/sarahpyfrom/PLAIDOH .