RMBase: a resource for decoding the landscape of RNA modifications from high-throughput sequencing data

Predicting RNA 5-Methylcytosine Sites by Using Essential Sequence Features and Distributions

Methylation is one of the most common and considerable modifications in biological systems mediated by multiple enzymes. Recent studies have shown that methylation has been widely identified in different RNA molecules. RNA methylation modifications have various kinds, such as 5-methylcytosine (m5C). However, for individual methylation sites, their functions still remain to be elucidated. Testing of all methylation sites relies heavily on high-throughput sequencing technology, which is expensive and labor consuming. Thus, computational prediction approaches could serve as a substitute. In this study, multiple machine learning models were used to predict possible RNA m5C sites on the basis of mRNA sequences in human and mouse. Each site was represented by several features derived from $k$ -mers of an RNA subsequence containing such site as center. The powerful max-relevance and min-redundancy (mRMR) feature selection method was employed to analyse these features. The outcome feature list was fed into incremental feature selection method, incorporating four classification algorithms, to build efficient models. Furthermore, the sites related to features used in the models were also investigated.

Research Progress of circRNAs in Inflammatory Mechanisms of Diabetic Retinopathy: An Emerging Star with Potential Therapeutic Targets

PURPOSE

 We summarized the existing studies to elaborate the biogenesis and function of circRNAs, the effect of aberrant circRNAs expression in the mechanism of inflammation and diabetic retinopathy (DR) respectively and further explored the vital roles of circRNAs in inflammation involved in DR. Methods: We conducted a systematical literature search of abundant electronic databases (PubMed, GeneMedical and MEDLINE) up to August 2021. Results: In this review, we exhibited the biogenesis and function of circRNAs and highlighted the components of inflammatory mediators implicated in DR. Numerous circRNAs, such as circHIPK3, circZNF609, circRNA_0084043, circ_0002570, circ_0041795, circEhmt1 and circ-ITCH were discovered to play vital roles in inflammation involved in DR, which provided new ideas for diagnosis and treatment of DR. Moreover, we proposed not only the epigenetic functions of circRNAs but also novel forms of the inflammatory response, including pyroptosis, to inspire further exploration and creative research in this field. Conclusion: CircRNAs were implicated in the progression and development of inflammation in DR via aberrant expression and modulation of gene expression, serving as an emerging star with potential therapeutic targets.

Accurate identification of RNA D modification using multiple features

As one of the common post-transcriptional modifications in tRNAs, dihydrouridine (D) has prominent effects on regulating the flexibility of tRNA as well as cancerous diseases. Facing with the expensive and time-consuming sequencing techniques to detect D modification, precise computational tools can largely promote the progress of molecular mechanisms and medical developments. We proposed a novel predictor, called iRNAD_XGBoost, to identify potential D sites using multiple RNA sequence representations. In this method, by considering the imbalance problem using hybrid sampling method SMOTEEEN, the XGBoost-selected top 30 features are applied to construct model. The optimized model showed high Sn and Sp values of 97.13% and 97.38% over jackknife test, respectively. For the independent experiment, these two metrics separately achieved 91.67% and 94.74%. Compared with iRNAD method, this model illustrated high generalizability and consistent prediction efficiencies for positive and negative samples, which yielded satisfactory MCC scores of 0.94 and 0.86, respectively. It is inferred that the chemical property and nucleotide density features (CPND), electron-ion interaction pseudopotential (EIIP and PseEIIP) as well as dinucleotide composition (DNC) are crucial to the recognition of D modification. The proposed predictor is a promising tool to help experimental biologists investigate molecular functions.

Epigenetic modifications in acute myeloid leukemia: The emerging role of circular RNAs (Review)

Canonical epigenetic modifications, which include histone modification, chromatin remodeling and DNA methylation, play key roles in numerous cellular processes. Epigenetics underlies how cells that posses DNA with similar sequences develop into different cell types with different functions in an organism. Earlier epigenetic research has primarily been focused at the chromatin level. However, the number of studies on epigenetic modifications of RNA, such as N1‑methyladenosine, 2'‑O‑ribosemethylation, inosine, 5‑methylcytidine, N6‑methyladenosine (m6A) and pseudouridine, has seen an increase. Circular RNAs (circRNAs), a type of RNA species that lacks a 5' cap or 3' poly(A) tail, are abundantly expressed in acute myeloid leukemia (AML) and may regulate disease progression. circRNAs possess various functions, including microRNA sponging, gene transcription regulation and RNA‑binding protein interaction. Furthermore, circRNAs are m6A methylated in other types of cancer, such as colorectal and hypopharyngeal squamous cell cancers. Therefore, the critical roles of circRNA epigenetic modifications, particularly m6A, and their possible involvement in AML are discussed in the present review. Epigenetic modification of circRNAs may become a diagnostic and therapeutic target for AML in the future.

Pol3Base: a resource for decoding the interactome, expression, evolution, epitranscriptome and disease variations of Pol III-transcribed ncRNAs

RNA polymerase III (Pol III) transcribes hundreds of non-coding RNA genes (ncRNAs), which involve in a variety of cellular processes. However, the expression, functions, regulatory networks and evolution of these Pol III-transcribed ncRNAs are still largely unknown. In this study, we developed a novel resource, Pol3Base (http://rna.sysu.edu.cn/pol3base/), to decode the interactome, expression, evolution, epitranscriptome and disease variations of Pol III-transcribed ncRNAs. The current release of Pol3Base includes thousands of regulatory relationships between ∼79 000 ncRNAs and transcription factors by mining 56 ChIP-seq datasets. By integrating CLIP-seq datasets, we deciphered the interactions of these ncRNAs with >240 RNA binding proteins. Moreover, Pol3Base contains ∼9700 RNA modifications located within thousands of Pol III-transcribed ncRNAs. Importantly, we characterized expression profiles of ncRNAs in >70 tissues and 28 different tumor types. In addition, by comparing these ncRNAs from human and mouse, we revealed about 4000 evolutionary conserved ncRNAs. We also identified ∼11 403 tRNA-derived small RNAs (tsRNAs) in 32 different tumor types. Finally, by analyzing somatic mutation data, we investigated the mutation map of these ncRNAs to help uncover their potential roles in diverse diseases. This resource will help expand our understanding of potential functions and regulatory networks of Pol III-transcribed ncRNAs.

The Expression Patterns and Prognostic Value of the Proteasome Activator Subunit Gene Family in Gastric Cancer Based on Integrated Analysis

Increasing evidence supports that proteasome activator subunit (PSME) genes play an indispensable role in multiple tumors. The diverse expression patterns, prognostic value, underlying mechanism, and the role in the immunotherapy of PSME genes in gastric cancer (GC) have yet to be fully elucidated. We systematically demonstrated the functions of these genes in GC using various large databases, unbiased in silico approaches, and experimental validation. We found that the median expression levels of all PSME genes were significantly higher in GC tissues than in normal tissues. Our findings showed that up-regulated PSME1 and PSME2 expression significantly correlated with favorable overall survival, post-progression survival, and first progression survival in GC patients. The expression of PSME1 and PSME2 was positively correlated with the infiltration of most immune cells and the activation of anti-cancer immunity cycle steps. Moreover, GC patients with high PSME1 and PSME2 expression have higher immunophenoscore and tumor mutational burden. In addition, a receiver operating characteristic analysis suggested that PSME3 and PSME4 had high diagnostic performance for distinguishing GC patients from healthy individuals. Moreover, our further analysis indicated that PSME genes exert an essential role in GC, and the present study indicated that PSME1 and PSME2 may be potential prognostic markers for enhancing survival and prognostic accuracy in GC patients and may even act as potential biomarkers for GC patients indicating a response to immunotherapy. PSME3 may serve as an oncogene in tumorigenesis and may be a promising therapeutic target for GC. PSME4 had excellent diagnostic performance and could serve as a good diagnostic indicator for GC.

Characterizing RNA Pseudouridylation by Convolutional Neural Networks

Pseudouridine (Ψ) is the most prevalent post-transcriptional RNA modification and is widespread in small cellular RNAs and mRNAs. However, the functions, mechanisms, and precise distribution of Ψs (especially in mRNAs) still remain largely unclear. The landscape of Ψs across the transcriptome has not yet been fully delineated. Here, we present a highly effective model based on a convolutional neural network (CNN), called PseudoUridyLation Site Estimator (PULSE), to analyze large-scale profiling data of Ψ sites and characterize the contextual sequence features of pseudouridylation. PULSE, consisting of two alternatively-stacked convolution and pooling layers followed by a fully-connected neural network, can automatically learn the hidden patterns of pseudouridylation from the local sequence information. Extensive validation tests demonstrated that PULSE can outperform other state-of-the-art prediction methods and achieve high prediction accuracy, thus enabling us to further characterize the transcriptome-wide landscape of Ψ sites. We further showed that the prediction results derived from PULSE can provide novel insights into understanding the functional roles of pseudouridylation, such as the regulations of RNA secondary structure, codon usage, translation, and RNA stability, and the connection to single nucleotide variants. The source code and final model for PULSE are available at https://github.com/mlcb-thu/PULSE.

Bioinformatical identification of key genes regulated by IGF2BP2-mediated RNA N6-methyladenosine and prediction of prognosis in hepatocellular carcinoma

Background

The treatment of hepatocellular carcinoma (HCC), a malignant cancer with global spread, remains unsatisfactory, and novel prognostic biomarkers need to be identified. N6-methyladenosine (m⁶A) has been found to regulate tumor initiation and progression through different mechanisms. As a dynamic and reversible messenger RNA (mRNA) modification, m⁶A can be read by insulin-like growth factor 2 mRNA-binding protein 2 (IGF2BP2). IGF2BP2 targets thousands of mRNA transcripts, which may be involved in HCC progression.

Methods

In this study, we integrated 4 classes of datasets including The Cancer Genome Atlas (TCGA)-LICH, m⁶A-sequencing data of HepG2 cells, and RNA-sequencing data of IGF2BP2-knockdown HepG2 cells to explore the key genes regulated by IGF2BP2-mediated m⁶A in HCC. The expression and m⁶A modification of candidates were validation in independent microarray expression profile of HCC tissue and annotated m⁶A database RMBase. The relationship of immune cell infiltration and the genes expression was estimated by CIBERSORT and TIMER.

Results

A total of 89 candidate genes were filtered. Next, cluster analysis was performed base on functions and pathways to identify the enrichment pathways. By constructing a protein-protein interaction (PPI) network, we found 54 nodes. Ten significant genes were filtered from the PPI. These genes were validated in data of an independent microarray and an m⁶A database. We found that the upregulation of these 10 genes was associated with poor prognosis. In addition, we showed the expression of these 10 genes was associated with the infiltration of variety of immune cell and tumor purity.

Conclusions

These identified genes may provide novel insights and facilitate the development of potential biomarkers for HCC diagnosis, as well as provide clues for IGF2BP2 inhibition therapy in HCC.

The functional roles, cross-talk and clinical implications of m6A modification and circRNA in hepatocellular carcinoma

Hepatocellular carcinoma (HCC) is one of the leading causes of cancer-related deaths worldwide. HCC has high rates of death and recurrence, as well as very low survival rates. N6-methyladenosine (m6A) is the most abundant modification in eukaryotic RNAs, and circRNAs are a class of circular noncoding RNAs that are generated by back-splicing and they modulate multiple functions in a variety of cellular processes. Although the carcinogenesis of HCC is complex, emerging evidence has indicated that m6A modification and circRNA play vital roles in HCC development and progression. However, the underlying mechanisms governing HCC, their cross-talk, and clinical implications have not been fully elucidated. Therefore, in this paper, we elucidated the biological functions and molecular mechanisms of m6A modification in the carcinogenesis of HCC by illustrating three different regulatory factors ("writer", "eraser", and "reader") of the m6A modification process. Additionally, we dissected the functional roles of circRNAs in various malignant behaviors of HCC, thereby contributing to HCC initiation, progression and relapse. Furthermore, we demonstrated the cross-talk and interplay between m6A modification and circRNA by revealing the effects of the collaboration of circRNA and m6A modification on HCC progression. Finally, we proposed the clinical potential and implications of m6A modifiers and circRNAs as diagnostic biomarkers and therapeutic targets for HCC diagnosis, treatment and prognosis evaluation.

Beyond small molecules: targeting G-quadruplex structures with oligonucleotides and their analogues

G-Quadruplexes (G4s) are widely studied secondary DNA/RNA structures, naturally occurring when G-rich sequences are present. The strategic localization of G4s in genome areas of crucial importance, such as proto-oncogenes and telomeres, entails fundamental implications in terms of gene expression regulation and other important biological processes. Although thousands of small molecules capable to induce G4 stabilization have been reported over the past 20 years, approaches based on the hybridization of a synthetic probe, allowing sequence-specific G4-recognition and targeting are still rather limited. In this review, after introducing important general notions about G4s, we aim to list, explain and critically analyse in more detail the principal approaches available to target G4s by using oligonucleotides and synthetic analogues such as Locked Nucleic Acids (LNAs) and Peptide Nucleic Acids (PNAs), reporting on the most relevant examples described in literature to date.

Porpoise: a new approach for accurate prediction of RNA pseudouridine sites

Pseudouridine is a ubiquitous RNA modification type present in eukaryotes and prokaryotes, which plays a vital role in various biological processes. Almost all kinds of RNAs are subject to this modification. However, it remains a great challenge to identify pseudouridine sites via experimental approaches, requiring expensive and time-consuming experimental research. Therefore, computational approaches that can be used to perform accurate in silico identification of pseudouridine sites from the large amount of RNA sequence data are highly desirable and can aid in the functional elucidation of this critical modification. Here, we propose a new computational approach, termed Porpoise, to accurately identify pseudouridine sites from RNA sequence data. Porpoise builds upon a comprehensive evaluation of 18 frequently used feature encoding schemes based on the selection of four types of features, including binary features, pseudo k-tuple composition, nucleotide chemical property and position-specific trinucleotide propensity based on single-strand (PSTNPss). The selected features are fed into the stacked ensemble learning framework to enable the construction of an effective stacked model. Both cross-validation tests on the benchmark dataset and independent tests show that Porpoise achieves superior predictive performance than several state-of-the-art approaches. The application of model interpretation tools demonstrates the importance of PSTNPs for the performance of the trained models. This new method is anticipated to facilitate community-wide efforts to identify putative pseudouridine sites and formulate novel testable biological hypothesis.

Shaping the Bacterial Epitranscriptome-5'-Terminal and Internal RNA Modifications

All domains of life utilize a diverse set of modified ribonucleotides that can impact the sequence, structure, function, stability, and the fate of RNAs, as well as their interactions with other molecules. Today, more than 160 different RNA modifications are known that decorate the RNA at the 5'-terminus or internal RNA positions. The boost of next-generation sequencing technologies sets the foundation to identify and study the functional role of RNA modifications. The recent advances in the field of RNA modifications reveal a novel regulatory layer between RNA modifications and proteins, which is central to developing a novel concept called "epitranscriptomics." The majority of RNA modifications studies focus on the eukaryotic epitranscriptome. In contrast, RNA modifications in prokaryotes are poorly characterized. This review outlines the current knowledge of the prokaryotic epitranscriptome focusing on mRNA modifications. Here, it is described that several internal and 5'-terminal RNA modifications either present or likely present in prokaryotic mRNA. Thereby, the individual techniques to identify these epitranscriptomic modifications, their writers, readers and erasers, and their proposed functions are explored. Besides that, still unanswered questions in the field of prokaryotic epitranscriptomics are pointed out, and its future perspectives in the dawn of next-generation sequencing technologies are outlined.

RFhy-m2G: Identification of RNA N2-methylguanosine modification sites based on random forest and hybrid features

N2-methylguanosine is a post-transcriptional modification of RNA that is found in eukaryotes and archaea. The biological function of m2G modification discovered so far is to control and stabilize the three-dimensional structure of tRNA and the dynamic barrier of reverse transcription. To discover additional biological functions of m2G, it is necessary to develop time-saving and labor-saving calculation tools to identify m2G. In this paper, based on hybrid features and a random forest, a novel predictor, RFhy-m2G, was developed to identify the m2G modification sites for three species. The hybrid feature used by the predictor is used to fuse the three features of ENAC, PseDNC, and NPPS. These three features include primary sequence derivation properties, physicochemical properties, and position-specific properties. Since there are redundant features in hybrid features, MRMD2.0 is used for optimal feature selection. Through feature analysis, it is found that the optimal hybrid features obtained still contain three kinds of properties, and the hybrid features can more accurately identify m2G modification sites and improve prediction performance. Based on five-fold cross-validation and independent testing to evaluate the prediction model, the accuracies obtained were 0.9982 and 0.9417, respectively. The robustness of the predictor is demonstrated by comparisons with other predictors.

Elucidating the Functions of Non-Coding RNAs from the Perspective of RNA Modifications

It is now commonly accepted that most of the mammalian genome is transcribed as RNA, yet less than 2% of such RNA encode for proteins. A majority of transcribed RNA exists as non-protein-coding RNAs (ncRNAs) with various functions. Because of the lack of sequence homologies among most ncRNAs species, it is difficult to infer the potential functions of ncRNAs by examining sequence patterns, such as catalytic domains, as in the case of proteins. Added to the existing complexity of predicting the functions of the ever-growing number of ncRNAs, increasing evidence suggests that various enzymes modify ncRNAs (e.g., ADARs, METTL3, and METTL14), which has opened up a new field of study called epitranscriptomics. Here, we examine the current status of ncRNA research from the perspective of epitranscriptomics.

Novel insights into the m6A-RNA methyltransferase METTL3 in cancer

N6-methyladenosine (m⁶A) is a prevalent internal RNA modification in higher eukaryotic cells. As the pivotal m⁶A regulator, RNA methyltransferase-like 3 (METTL3) is responsible for methyl group transfer in the progression of m⁶A modification. This epigenetic regulation contributes to the structure and functional regulation of RNA and further promotes tumorigenesis and tumor progression. Accumulating evidence has illustrated the pivotal roles of METTL3 in a variety of human cancers. Here, we systemically summarize the interaction between METTL3 and RNAs, and illustrate the multiple functions of METTL3 in human cancer. METLL3 is aberrantly expressed in a variety of tumors. Elevation of METTL3 is usually associated with rapid progression and poor prognosis of tumors. On the other hand, METTL3 may also function as a tumor suppressor in several cancers. Based on the tumor-promoting effect of METTL3, the possibility of applying METTL3 inhibitors is further discussed, which is expected to provide novel insights into antitumor therapy.

A brief review of RNA modification related database resources

To date, over 150 different RNA modifications have been identified, playing crucial roles in biological processes and disease pathogenesis. Thanks to the advancement of high-throughput sequencing technologies employed for transcriptome-wide mapping, a bunch of RNA modification databases have emerged as an exciting area, which promotes further investigation of the mechanisms and functions of these modified ribonucleotides. This article introduces the high-throughput sequencing technique developed for transcriptome-wide mapping of RNA modifications, as well as the procedures and main techniques of building these databases from the developers' perspective. It also reviews existing RNA modification databases in terms of their main functions, species, the number of sites they collected, the annotations, and the tools they provided. From the view of users, we further analyze and compare these databases in terms of their functions. For instance, these databases can be applied to record chemical structures and biosynthetic pathways, or unravel the epi-transcriptome comprehensively, or only investigate specific features of RNA modifications. Additionally, the limitations of the existing approaches are discussed, and some future suggestions are offered.

Identification of RNA pseudouridine sites using deep learning approaches

Pseudouridine(Ψ) is widely popular among various RNA modifications which have been confirmed to occur in rRNA, mRNA, tRNA, and nuclear/nucleolar RNA. Hence, identifying them has vital significance in academic research, drug development and gene therapies. Several laboratory techniques for Ψ identification have been introduced over the years. Although these techniques produce satisfactory results, they are costly, time-consuming and requires skilled experience. As the lengths of RNA sequences are getting longer day by day, an efficient method for identifying pseudouridine sites using computational approaches is very important. In this paper, we proposed a multi-channel convolution neural network using binary encoding. We employed k-fold cross-validation and grid search to tune the hyperparameters. We evaluated its performance in the independent datasets and found promising results. The results proved that our method can be used to identify pseudouridine sites for associated purposes. We have also implemented an easily accessible web server at http://103.99.176.239/ipseumulticnn/.

YTH Domain Proteins: A Family of m6A Readers in Cancer Progression

N⁶-methyladenosine (m⁶A) is the most abundant internal modification in eukaryotic messenger RNAs (mRNAs). m⁶A RNA methylation is involved in all stages of RNA life cycle, from RNA processing, nuclear output, translation regulation to RNA degradation, indicating that m⁶A has various functions affecting RNA metabolism positively or negatively. Reading proteins are vital in regulating the translation and stability of m⁶A mRNAs positively or negatively. Recent studies have enhanced the understanding of the molecular mechanism of the YT521-B homology (YTH) domain family and the modification of m⁶A. This study aimed to review the specific mechanisms, functions, and interactions of the YTH domain protein family. It also discussed future research directions, thus providing new ideas for the clinical diagnosis and targeted therapy of cancer.

N6-methyladenosine modification of MALAT1 promotes metastasis via reshaping nuclear speckles

N6-methyladenosine (m⁶A), one of the most prevalent RNA post-transcriptional modifications, is involved in numerous biological processes. In previous studies, the functions of m⁶A were typically identified by perturbing the activity of the methyltransferase complex. Here, we dissect the contribution of m⁶A to an individual-long noncoding RNA-metastasis-associated lung adenocarcinoma transcript 1 (MALAT1). The mutant MALAT1 lacking m⁶A-motifs significantly suppressed the metastatic potential of cancer cells both in vitro and in vivo in mouse. Super-resolution imaging showed that the concatenated m⁶A residues on MALAT1 acted as a scaffold for recruiting YTH-domain-containing protein 1 (YTHDC1) to nuclear speckles. We further reveal that the recognition of MALAT1-m⁶A by YTHDC1 played a critical role in maintaining the composition and genomic binding sites of nuclear speckles, which regulate the expression of several key oncogenes. Furthermore, artificially tethering YTHDC1 onto m⁶A-deficient MALAT1 largely rescues the metastatic potential of cancer cells.

Posttranscriptional modifications at the 37th position in the anticodon stem-loop of tRNA: structural insights from MD simulations

Transfer RNA (tRNA) is the most diversely modified RNA. Although the strictly conserved purine position 37 in the anticodon stem-loop undergoes modifications that are phylogenetically distributed, we do not yet fully understand the roles of these modifications. Therefore, molecular dynamics simulations are used to provide molecular-level details for how such modifications impact the structure and function of tRNA. A focus is placed on three hypermodified base families that include the parent i6A, t6A, and yW modifications, as well as derivatives. Our data reveal that the hypermodifications exhibit significant conformational flexibility in tRNA, which can be modulated by additional chemical functionalization. Although the overall structure of the tRNA anticodon stem remains intact regardless of the modification considered, the anticodon loop must rearrange to accommodate the bulky, dynamic hypermodifications, which includes changes in the nucleotide glycosidic and backbone conformations, and enhanced or completely new nucleobase-nucleobase interactions compared to unmodified tRNA or tRNA containing smaller (m1G) modifications at the 37th position. Importantly, the extent of the changes in the anticodon loop is influenced by the addition of small functional groups to parent modifications, implying each substituent can further fine-tune tRNA structure. Although the dominant conformation of the ASL is achieved in different ways for each modification, the molecular features of all modified tRNA drive the ASL domain to adopt the functional open-loop conformation. Importantly, the impact of the hypermodifications is preserved in different sequence contexts. These findings highlight the likely role of regulating mRNA structure and translation.

PseUdeep: RNA Pseudouridine Site Identification with Deep Learning Algorithm

Background: Pseudouridine (Ψ) is a common ribonucleotide modification that plays a significant role in many biological processes. The identification of Ψ modification sites is of great significance for disease mechanism and biological processes research in which machine learning algorithms are desirable as the lab exploratory techniques are expensive and time-consuming. Results: In this work, we propose a deep learning framework, called PseUdeep, to identify Ψ sites of three species: H. sapiens, S. cerevisiae, and M. musculus. In this method, three encoding methods are used to extract the features of RNA sequences, that is, one-hot encoding, K-tuple nucleotide frequency pattern, and position-specific nucleotide composition. The three feature matrices are convoluted twice and fed into the capsule neural network and bidirectional gated recurrent unit network with a self-attention mechanism for classification. Conclusion: Compared with other state-of-the-art methods, our model gets the highest accuracy of the prediction on the independent testing data set S-200; the accuracy improves 12.38%, and on the independent testing data set H-200, the accuracy improves 0.68%. Moreover, the dimensions of the features we derive from the RNA sequences are only 109,109, and 119 in H. sapiens, M. musculus, and S. cerevisiae, which is much smaller than those used in the traditional algorithms. On evaluation via tenfold cross-validation and two independent testing data sets, PseUdeep outperforms the best traditional machine learning model available. PseUdeep source code and data sets are available at https://github.com/dan111262/PseUdeep.

Advances in the profiling of N6-methyladenosine (m6A) modifications

Over 160 RNA modifications have been identified, including N⁷-methylguanine (m⁷G), N⁶-methyladenosine (m⁶A), and 5-methylcytosine (m⁵C). These modifications play key roles in regulating the fate of RNA. In eukaryotes, m⁶A is the most abundant mRNA modification, accounting for over 80% of all RNA methylation modifications. Highly dynamic m⁶A modification may exert important effects on organismal reproduction and development. Significant advances in understanding the mechanism of m⁶A modification have been made using immunoprecipitation, chemical labeling, and site-directed mutagenesis, combined with next-generation sequencing. Single-molecule real-time and nanopore direct RNA sequencing (DRS) approaches provide additional ways to study RNA modifications at the cellular level. In this review, we explore the technical history of identifying m⁶A RNA modifications, emphasizing technological advances in detecting m⁶A modification. In particular, we discuss the challenge of generating accurate dynamic single-base resolution m⁶A maps and also strategies for improving detection specificity. Finally, we outline a roadmap for future research in this area, focusing on the application of RNA epigenetic modification, represented by m⁶A modification.

The Distinct Function and Localization of METTL3/METTL14 and METTL16 Enzymes in Cardiomyocytes

It has become evident that epitranscriptome events, mediated by specific enzymes, regulate gene expression and, subsequently, cell differentiation processes. We show that methyltransferase-like proteins METTL3/METTL14 and N⁶-adenosine methylation (m6A) in RNAs are homogeneously distributed in embryonic hearts, and histone deacetylase (HDAC) inhibitors valproic acid and Trichostatin A (TSA) up-regulate METTL3/METTL14 proteins. The levels of METTL3 in mouse adult hearts, isolated from male and female animals, were lower in the aorta and pulmonary trunks when compared with atria, but METT14 was up-regulated in the aorta and pulmonary trunk, in comparison with ventriculi. Aging caused METTL3 down-regulation in aorta and atria in male animals. Western blot analysis in differentiated mouse embryonic stem cells (mESCs), containing 10-30 percent of cardiomyocytes, showed METTL3/METTL14 down-regulation, while the differentiation-induced increased level of METTL16 was observed in both wild type (wt) and HDAC1 depleted (dn) cells. In parallel, experimental differentiation in especially HDAC1 wild type cells was accompanied by depletion of m6A in RNA. Immunofluorescence analysis of individual cells revealed the highest density of METTL3/METTL14 in α-actinin positive cardiomyocytes when compared with the other cells in the culture undergoing differentiation. In both wt and HDAC1 dn cells, the amount of METTL16 was also up-regulated in cardiomyocytes when compared to co-cultivated cells. Together, we showed that distinct anatomical regions of the mouse adult hearts are characterized by different levels of METTL3 and METTL14 proteins, which are changed during aging. Experimental cell differentiation was also accompanied by changes in METTL-like proteins and m6A in RNA; in particular, levels and distribution patterns of METTL3/METTL14 proteins were different from the same parameters studied in the case of the METTL16 protein.

REW-ISA: unveiling local functional blocks in epi-transcriptome profiling data via an RNA expression-weighted iterative signature algorithm

Background

Recent studies have shown that N⁶-methyladenosine (m⁶A) plays a critical role in numbers of biological processes and complex human diseases. However, the regulatory mechanisms of most methylation sites remain uncharted. Thus, in-depth study of the epi-transcriptomic patterns of m⁶A may provide insights into its complex functional and regulatory mechanisms.

Results

Due to the high economic and time cost of wet experimental methods, revealing methylation patterns through computational models has become a more preferable way, and drawn more and more attention. Considering the theoretical basics and applications of conventional clustering methods, an RNA Expression Weighted Iterative Signature Algorithm (REW-ISA) is proposed to find potential local functional blocks (LFBs) based on MeRIP-Seq data, where sites are hyper-methylated or hypo-methylated simultaneously across the specific conditions. REW-ISA adopts RNA expression levels of each site as weights to make sites of lower expression level less significant. It starts from random sets of sites, then follows iterative search strategies by thresholds of rows and columns to find the LFBs in m⁶A methylation profile. Its application on MeRIP-Seq data of 69,446 methylation sites under 32 experimental conditions unveiled 6 LFBs, which achieve higher enrichment scores than ISA. Pathway analysis and enzyme specificity test showed that sites remained in LFBs are highly relevant to the m⁶A methyltransferase, such as METTL3, METTL14, WTAP and KIAA1429. Further detailed analyses for each LFB even showed that some LFBs are condition-specific, indicating that methylation profiles of some specific sites may be condition relevant.

Conclusions

REW-ISA finds potential local functional patterns presented in m⁶A profiles, where sites are co-methylated under specific conditions.

Dynamic landscape and evolution of m6A methylation in human

m⁶A is a prevalent internal modification in mRNAs and has been linked to the diverse effects on mRNA fate. To explore the landscape and evolution of human m⁶A, we generated 27 m⁶A methylomes across major adult tissues. These data reveal dynamic m⁶A methylation across tissue types, uncover both broadly or tissue-specifically methylated sites, and identify an unexpected enrichment of m⁶A methylation at non-canonical cleavage sites. A comparison of fetal and adult m⁶A methylomes reveals that m⁶A preferentially occupies CDS regions in fetal tissues. Moreover, the m⁶A sub-motifs vary between fetal and adult tissues or across tissue types. From the evolutionary perspective, we uncover that the selection pressure on m⁶A sites varies and depends on their genic locations. Unexpectedly, we found that ∼40% of the 3′UTR m⁶A sites are under negative selection, which is higher than the evolutionary constraint on miRNA binding sites, and much higher than that on A-to-I RNA modification. Moreover, the recently gained m⁶A sites in human populations are clearly under positive selection and associated with traits or diseases. Our work provides a resource of human m⁶A profile for future studies of m⁶A functions, and suggests a role of m⁶A modification in human evolutionary adaptation and disease susceptibility.

Prognostic Value of an m6A RNA Methylation Regulator-Based Signature in Patients with Hepatocellular Carcinoma

Purposes

Hepatocellular carcinoma (HCC) is one of the most common malignant tumors in the world. Recent researches have demonstrated that m6A methylation regulators play a key role in various cancers, such as gastric cancer and colon adenocarcinoma. Several m6A methylation regulators are reported to predict the prognosis of HCC. Therefore, there is a need to further identify the predictive value of m6A methylation regulators in HCC.

Methods

We utilized The Cancer Genome Atlas (TCGA) database to obtain the gene expression profile of m6A RNA methylation regulators and clinical information for patients with HCC. Besides, we identified two clusters of HCC with various clinical factors by consensus clustering analysis. Then the least absolute shrinkage and selection operator (LASSO) and the Cox regression analysis were applied to construct a prognostic signature.

Results

Except for ZC3H13 and METTL14, a majority of the thirteen m6A RNA methylation regulators were significantly overexpressed in HCC specimens. HCC patients were classified into two groups (cluster 1 and cluster 2). The cluster 1 was with a significantly worse prognosis than cluster 2, and most of the 13 known m6A RNA methylation regulators were upregulated in cluster 1. Besides, we developed a prognostic signature consisting of YTHDF2, YTHDF1, METTL3, KIAA1429, and ZC3H13, which could successfully differentiate high-risk patients. More importantly, univariate and multivariate Cox regression analysis indicated that the signature-based risk score was an independent prognostic factor for patients with HCC.

Conclusions

Our study showed these five m6A RNA methylation regulators can be used as practical and reliable prognostic tools of HCC, which might have potential value for therapeutic strategies.

Bioinformatics approaches for deciphering the epitranscriptome: Recent progress and emerging topics

Post-transcriptional RNA modification occurs on all types of RNA and plays a vital role in regulating every aspect of RNA function. Thanks to the development of high-throughput sequencing technologies, transcriptome-wide profiling of RNA modifications has been made possible. With the accumulation of a large number of high-throughput datasets, bioinformatics approaches have become increasing critical for unraveling the epitranscriptome. We review here the recent progress in bioinformatics approaches for deciphering the epitranscriptomes, including epitranscriptome data analysis techniques, RNA modification databases, disease-association inference, general functional annotation, and studies on RNA modification site prediction. We also discuss the limitations of existing approaches and offer some future perspectives.

Prediction of m5C Modifications in RNA Sequences by Combining Multiple Sequence Features

5-Methylcytosine (m5C) is a well-known post-transcriptional modification that plays significant roles in biological processes, such as RNA metabolism, tRNA recognition, and stress responses. Traditional high-throughput techniques on identification of m5C sites are usually time consuming and expensive. In addition, the number of RNA sequences shows explosive growth in the post-genomic era. Thus, machine-learning-based methods are urgently requested to quickly predict RNA m5C modifications with high accuracy. Here, we propose a noval support-vector-machine (SVM)-based tool, called iRNA-m5C_SVM, by combining multiple sequence features to identify m5C sites in Arabidopsis thaliana. Eight kinds of popular feature-extraction methods were first investigated systematically. Then, four well-performing features were incorporated to construct a comprehensive model, including position-specific propensity (PSP) (PSNP, PSDP, and PSTP, associated with frequencies of nucleotides, dinucleotides, and trinucleotides, respectively), nucleotide composition (nucleic acid, di-nucleotide, and tri-nucleotide compositions; NAC, DNC, and TNC, respectively), electron-ion interaction pseudopotentials of trinucleotide (PseEIIPs), and general parallel correlation pseudo-dinucleotide composition (PC-PseDNC-general). Evaluated accuracies over 10-fold cross-validation and independent tests achieved 73.06% and 80.15%, respectively, which showed the best predictive performances in A. thaliana among existing models. It is believed that the proposed model in this work can be a promising alternative for further research on m5C modification sites in plant.

Classification and function of RNA-protein interactions

Almost all RNAs need to interact with proteins to fully exert their functions, and proteins also bind to RNAs to act as regulators. It has now become clear that RNA-protein interactions play important roles in many biological processes among organisms. Despite the great progress that has been made in the field, there is still no precise classification system for RNA-protein interactions, which makes it challenging to further decipher the functions and mechanisms of these interactions. In this review, we propose four different categories of RNA-protein interactions according to their basic characteristics: RNA motif-dependent RNA-protein interactions, RNA structure-dependent RNA-protein interactions, RNA modification-dependent RNA-protein interactions, and RNA guide-based RNA-protein interactions. Moreover, the integration of different types of RNA-protein interactions and the regulatory factors implicated in these interactions are discussed. Furthermore, we emphasize the functional diversity of these four types of interactions in biological processes and disease development and assess emerging trends in this exciting research field. This article is categorized under: RNA Interactions with Proteins and Other Molecules > Protein-RNA Interactions: Functional Implications RNA Interactions with Proteins and Other Molecules > Protein-RNA Recognition RNA Processing > RNA Editing and Modification.

Predicting sites of epitranscriptome modifications using unsupervised representation learning based on generative adversarial networks

Epitranscriptome is an exciting area that studies different types of modifications in transcripts and the prediction of such modification sites from the transcript sequence is of significant interest. However, the scarcity of positive sites for most modifications imposes critical challenges for training robust algorithms. To circumvent this problem, we propose MR-GAN, a generative adversarial network (GAN) based model, which is trained in an unsupervised fashion on the entire pre-mRNA sequences to learn a low dimensional embedding of transcriptomic sequences. MR-GAN was then applied to extract embeddings of the sequences in a training dataset we created for eight epitranscriptome modifications, including m⁶A, m¹A, m¹G, m²G, m⁵C, m⁵U, 2'-O-Me, Pseudouridine (Ψ) and Dihydrouridine (D), of which the positive samples are very limited. Prediction models were trained based on the embeddings extracted by MR-GAN. We compared the prediction performance with the one-hot encoding of the training sequences and SRAMP, a state-of-the-art m⁶A site prediction algorithm and demonstrated that the learned embeddings outperform one-hot encoding by a significant margin for up to 15% improvement. Using MR-GAN, we also investigated the sequence motifs for each modification type and uncovered known motifs as well as new motifs not possible with sequences directly. The results demonstrated that transcriptome features extracted using unsupervised learning could lead to high precision for predicting multiple types of epitranscriptome modifications, even when the data size is small and extremely imbalanced.

A Mini-review of the Computational Methods Used in Identifying RNA 5-Methylcytosine Sites

RNA 5-methylcytosine (m⁵C) is one of the pillars of post-transcriptional modification (PTCM). A growing body of evidence suggests that m⁵C plays a vital role in RNA metabolism. Accurate localization of RNA m⁵C sites in tissue cells is the premise and basis for the in-depth understanding of the functions of m⁵C. However, the main experimental methods of detecting m⁵C sites are limited to varying degrees. Establishing a computational model to predict modification sites is an excellent complement to wet experiments for identifying m⁵C sites. In this review, we summarized some available m⁵C predictors and discussed the characteristics of these methods.

iPseU-Layer: Identifying RNA Pseudouridine Sites Using Layered Ensemble Model

Pseudouridine represents one of the most prevalent post-transcriptional RNA modifications. The identification of pseudouridine sites is an essential step toward understanding RNA functions, RNA structure stabilization, translation process, and RNA stability; however, high-throughput experimental techniques remain expensive and time-consuming in lab explorations and biochemical processes. Thus, how to develop an efficient pseudouridine site identification method based on machine learning is very important both in academic research and drug development. Motived by this, we present an effective layered ensemble model designated as iPseU-Layer for identification of RNA pseudouridine sites. The proposed iPseU-Layer approach is essentially based on three different machine learning layers including: feature selection layer, feature extraction and fusion layer, and prediction layer. The feature selection layer reduces the dimensionality, which can be regarded as a data pre-processing stage. The feature extraction and fusion layer utilizes an ensemble method which is implemented through various machine learning algorithms to generate some outputs. The prediction layer applies classic random forest to identify the final results. Furthermore, we systematically conduct the validation experiments using cross-validation tests and independent test with the current state-of-the-art models. The proposed iPseU-Layer provides a promising predictive performance in terms of sensitivity, specificity, accuracy and Matthews correlation coefficient. Collectively, these findings indicate that the framework of iPseU-Layer is a feasible and effective strategy for the prediction of RNA pseudouridine sites.

Is There Any Sequence Feature in the RNA Pseudouridine Modification Prediction Problem?

Pseudouridine (Ψ) is the most abundant RNA modification and has been found in many kinds of RNAs, including snRNA, rRNA, tRNA, mRNA, and snoRNA. Thus, Ψ sites play a significant role in basic research and drug development. Although some experimental techniques have been developed to identify Ψ sites, they are expensive and time consuming, especially in the post-genomic era with the explosive growth of known RNA sequences. Thus, highly accurate computational methods are urgently required to quickly detect the Ψ sites on uncharacterized RNA sequences. Several predictors have been proposed using multifarious features, but their evaluated performances are still unsatisfactory. In this study, we first identified Ψ sites for H. sapiens, S. cerevisiae, and M. musculus using the sequence features from the bi-profile Bayes (BPB) method based on the random forest (RF) and support vector machine (SVM) algorithms, where the performances were evaluated using 5-fold cross-validation and independent tests. It was found that the SVM-based accuracies were 3.55% and 5.09% lower than the iPseU-CUU predictor for the H_990 and S_628 datasets, respectively. Almost the same-level results were obtained for M_994 and an independent H_200 dataset, even showing a 5.0% improvement for S_200. Then, three different kinds of features, including basic Kmer, general parallel correlation pseudo-dinucleotide composition (PC-PseDNC-General), and nucleotide chemical property (NCP) and nucleotide density (ND) from the iRNA-PseU method, were combined with BPB to show their comprehensive performances, where the effective features are selected by the max-relevance-max-distance (MRMD) method. The best evaluated accuracies of the combined features for the S_628 and M_994 datasets were achieved at 70.54% and 72.45%, which were 2.39% and 0.65% higher than iPseU-CUU. For the S_200 dataset, it was also improved 8% from 69% to 77%. However, there was no obvious improvement for H. sapiens, which was evaluated as approximately 63.23% and 72.0% for the H_990 and H_200 datasets, respectively. The overall performances for Ψ identification using BPB features as well as the combined features were not obviously improved. Although some kinds of feature extraction methods based on the RNA sequence information have been applied to construct the predictors in previous studies, the corresponding accuracies are generally in the range of 60%-70%. Thus, researchers need to reconsider whether there is any sequence feature in the RNA Ψ modification prediction problem.

Plant organellar RNA editing: what 30 years of research has revealed

The central dogma in biology defines the flow of genetic information from DNA to RNA to protein. Accordingly, RNA molecules generally accurately follow the sequences of the genes from which they are transcribed. This rule is transgressed by RNA editing, which creates RNA products that differ from their DNA templates. Analyses of the RNA landscapes of terrestrial plants have indicated that RNA editing (in the form of C-U base transitions) is highly prevalent within organelles (that is, mitochondria and chloroplasts). Numerous C→U conversions (and in some plants also U→C) alter the coding sequences of many of the organellar transcripts and can also produce translatable mRNAs by creating AUG start sites or eliminating premature stop codons, or affect the RNA structure, influence splicing and alter the stability of RNAs. RNA-binding proteins are at the heart of post-transcriptional RNA expression. The C-to-U RNA editing process in plant mitochondria involves numerous nuclear-encoded factors, many of which have been identified as pentatricopeptide repeat (PPR) proteins that target editing sites in a sequence-specific manner. In this review we report on major discoveries on RNA editing in plant organelles, since it was first documented 30 years ago.

iPseU-NCP: Identifying RNA pseudouridine sites using random forest and NCP-encoded features

Background

Pseudouridine modification is most commonly found among various kinds of RNA modification occurred in both prokaryotes and eukaryotes. This biochemical event has been proved to occur in multiple types of RNAs, including rRNA, mRNA, tRNA, and nuclear/nucleolar RNA. Hence, gaining a holistic understanding of pseudouridine modification can contribute to the development of drug discovery and gene therapies. Although some laboratory techniques have come up with moderately good outcomes in pseudouridine identification, they are costly and required skilled work experience. We propose iPseU-NCP – an efficient computational framework to predict pseudouridine sites using the Random Forest (RF) algorithm combined with nucleotide chemical properties (NCP) generated from RNA sequences. The benchmark dataset collected from Chen et al. (2016) was used to develop iPseU-NCP and fairly compare its performances with other methods.

Results

Under the same experimental settings, comparing with three state-of-the-art methods including iPseU-CNN, PseUI, and iRNA-PseU, the Matthew’s correlation coefficient (MCC) of our model increased by about 20.0%, 55.0%, and 109.0% when tested on the H. sapiens (H_200) dataset and by about 6.5%, 35.0%, and 150.0% when tested on the S. cerevisiae (S_200) dataset, respectively. This significant growth in MCC is very important since it ensures the stability and performance of our model. With those two independent test datasets, our model also presented higher accuracy with a success rate boosted by 7.0%, 13.0%, and 20.0% and 2.0%, 9.5%, and 25.0% when compared to iPseU-CNN, PseUI, and iRNA-PseU, respectively. For majority of other evaluation metrics, iPseU-NCP demonstrated superior performance as well.

Conclusions

iPseU-NCP combining the RF and NPC-encoded features showed better performances than other existing state-of-the-art methods in the identification of pseudouridine sites. This also shows an optimistic view in addressing biological issues related to human diseases.

METTL3 regulates inflammatory pain by modulating m6A-dependent pri-miR-365-3p processing

N⁶-methyladenosine (m⁶A) modification in RNA has been implicated in diverse biological processes. However, very little is currently known about its role in nociceptive modulation. Here, we found that the level of spinal m⁶A modification was significantly increased in a mouse model of Complete Freund's Adjuvant (CFA)-induced chronic inflammatory pain, which was accompanied with the augmentation of methyltransferase-like 3 (METTL3) expression in the spinal cord. Knockdown of spinal METTL3 prevented and reversed CFA-induced pain behaviors and spinal neuronal sensitization. In contrast, overexpression of spinal METTL3 produced pain behaviors and neuronal sensitization in naive mice. Moreover, we found that METTL3 positively modulated the pri-miR-65-3p processing in a microprocessor protein DiGeorge critical region 8-dependent manner. Collectively, our findings reveal an important role of METTL3-mediated m⁶A modification in nociceptive sensitization and provide a novel perspective on m⁶A modification in the development of pathological pain.

Comprehensive review and assessment of computational methods for predicting RNA post-transcriptional modification sites from RNA sequences

RNA post-transcriptional modifications play a crucial role in a myriad of biological processes and cellular functions. To date, more than 160 RNA modifications have been discovered; therefore, accurate identification of RNA-modification sites is fundamental for a better understanding of RNA-mediated biological functions and mechanisms. However, due to limitations in experimental methods, systematic identification of different types of RNA-modification sites remains a major challenge. Recently, more than 20 computational methods have been developed to identify RNA-modification sites in tandem with high-throughput experimental methods, with most of these capable of predicting only single types of RNA-modification sites. These methods show high diversity in their dataset size, data quality, core algorithms, features extracted and feature selection techniques and evaluation strategies. Therefore, there is an urgent need to revisit these methods and summarize their methodologies, in order to improve and further develop computational techniques to identify and characterize RNA-modification sites from the large amounts of sequence data. With this goal in mind, first, we provide a comprehensive survey on a large collection of 27 state-of-the-art approaches for predicting N1-methyladenosine and N6-methyladenosine sites. We cover a variety of important aspects that are crucial for the development of successful predictors, including the dataset quality, operating algorithms, sequence and genomic features, feature selection, model performance evaluation and software utility. In addition, we also provide our thoughts on potential strategies to improve the model performance. Second, we propose a computational approach called DeepPromise based on deep learning techniques for simultaneous prediction of N1-methyladenosine and N6-methyladenosine. To extract the sequence context surrounding the modification sites, three feature encodings, including enhanced nucleic acid composition, one-hot encoding, and RNA embedding, were used as the input to seven consecutive layers of convolutional neural networks (CNNs), respectively. Moreover, DeepPromise further combined the prediction score of the CNN-based models and achieved around 43% higher area under receiver-operating curve (AUROC) for m1A site prediction and 2–6% higher AUROC for m6A site prediction, respectively, when compared with several existing state-of-the-art approaches on the independent test. In-depth analyses of characteristic sequence motifs identified from the convolution-layer filters indicated that nucleotide presentation at proximal positions surrounding the modification sites contributed most to the classification, whereas those at distal positions also affected classification but to different extents. To maximize user convenience, a web server was developed as an implementation of DeepPromise and made publicly available at http://DeepPromise.erc.monash.edu/, with the server accepting both RNA sequences and genomic sequences to allow prediction of two types of putative RNA-modification sites.

The Tudor SND1 protein is an m6A RNA reader essential for replication of Kaposi's sarcoma-associated herpesvirus

N6-methyladenosine (m6A) is the most abundant internal RNA modification of cellular mRNAs. m6A is recognised by YTH domain-containing proteins, which selectively bind to m6A-decorated RNAs regulating their turnover and translation. Using an m6A-modified hairpin present in the Kaposi's sarcoma associated herpesvirus (KSHV) ORF50 RNA, we identified seven members from the 'Royal family' as putative m6A readers, including SND1. RIP-seq and eCLIP analysis characterised the SND1 binding profile transcriptome-wide, revealing SND1 as an m6A reader. We further demonstrate that the m6A modification of the ORF50 RNA is critical for SND1 binding, which in turn stabilises the ORF50 transcript. Importantly, SND1 depletion leads to inhibition of KSHV early gene expression showing that SND1 is essential for KSHV lytic replication. This work demonstrates that members of the 'Royal family' have m6A-reading ability, greatly increasing their epigenetic functions beyond protein methylation.

RNAm5CPred: Prediction of RNA 5-Methylcytosine Sites Based on Three Different Kinds of Nucleotide Composition

5-methylcytosine (m5C) is one of the most common and abundant post-transcriptional modifications (PTCMs) in RNA. Recent studies showed that m5C plays important roles in many biological functions such as RNA metabolism and cell fate decision. Because most experimental methods that determine m5C sites across the transcriptome are time-consuming and expensive, it is urgent to develop accurate computational methods to identify m5C sites effectively. A benchmark dataset is important for developing and evaluating computational methods. In this work, we constructed four different datasets according to the data redundancy and imbalance. Based on these datasets, we generated three different kinds of features, i.e., KNFs (K-nucleotide frequencies), KSNPFs (K-spaced nucleotide pair frequencies), and pseDNC (pseudo-dinucleotide composition), and then used a support vector machine (SVM) to build our models. Based on the imbalanced and nonredundant dataset, Met935, we extensively studied the three kinds of features and determined an optimal combination of the features. Based on the feature combination, we built models on the three different datasets and compared them with state-of-the-art models. According to the predictive results of the stringent jackknife test, the models based on the three features, 4NF, 1SNPF, and pseDNC, are superior or comparable to other methods. To determine the best model between the models based on the imbalanced dataset Met935 and the balanced dataset Met240, we further evaluated the two models on an independent test set Test1157. Our results demonstrate that the model based on the balanced dataset Met240 achieved the highest recall (68.79%) and the highest Matthews correlation coefficient (MCC) (0.154). In addition, the model is also superior to other state-of-the-art methods according to the integrated parameter MCC on the independent test set. Thus, we selected the model based on Met240 as our final model, which was named RNAm5CPred. In addition, a web server for RNAm5CPred (http://zhulab.ahu.edu.cn/RNAm5CPred/) has been provided to facilitate experimental research.

Effect of methylation of adenine N6 on kink turn structure depends on location

N6-methyladenine is the most common covalent modification in cellular RNA species, with demonstrated functional consequences. At the molecular level this methylation could alter local RNA structure, and/or modulate the binding of specific proteins. We have previously shown that trans-Hoogsteen-sugar (sheared) A:G base pairs can be completely disrupted by methylation, and that this occurs in a sub-set ofD/D k-turn structures. In this work we have investigated to what extent sequence context affects the severity with which inclusion of N6-methyladenine into different A:G base pairs of a standard k-turn affects RNA folding and L7Ae protein binding. We find that local sequence has a major influence, ranging from complete absence of folding and protein binding to a relatively mild effect. We have determined the crystal structure of one of these species both free and protein-bound, showing the environment of the methyl group and the way the modification is accommodated into the k-turn structure.

Efficient Epidermal Growth Factor Receptor Targeting Oligonucleotide as a Potential Molecule for Targeted Cancer Therapy

Epidermal growth factor receptor (EGFR) is associated with the progression of a wide range of cancers including breast, glioma, lung, and liver cancer. The observation that EGFR inhibition can limit the growth of EGFR positive cancers has led to the development of various EGFR inhibitors including monoclonal antibodies and small-molecule inhibitors. However, the reported toxicity and drug resistance greatly compromised the clinical outcome of such inhibitors. As a type of chemical antibodies, nucleic acid aptamer provides an opportunity to overcome the obstacles faced by current EGFR inhibitors. In this study, we have developed and investigated the therapeutic potential of a 27mer aptamer CL-4RNV616 containing 2′-O-Methyl RNA and DNA nucleotides. Our results showed that CL-4RNV616 not only displayed enhanced stability in human serum, but also effectively recognized and inhibited the proliferation of EGFR positive Huh-7 liver cancer, MDA-MB-231 breast cancer, and U87MG glioblastoma cells, with an IC50 value of 258.9 nM, 413.7 nM, and 567.9 nM, respectively. Furthermore, TUNEL apoptosis assay revealed that CL-4RNV616 efficiently induced apoptosis of cancer cells. In addition, clinical breast cancer biopsy-based immunostaining assay demonstrated that CL-4RNV616 had a comparable detection efficacy for EGFR positive breast cancer with commonly used commercial antibodies. Based on the results, we firmly believe that CL-4RNV616 could be useful in the development of targeted cancer therapeutics and diagnostics.

iRNA-m2G: Identifying N2-methylguanosine Sites Based on Sequence-Derived Information

RNA N²-methylguanosine (m2G) is one kind of posttranscriptional modification and plays crucial roles in the control and stabilization of tRNA. However, our knowledge about the biological functions of m2G is still limited. The key step of revealing its new function is to recognize the m2G sites in the transcriptome. Since there is no effective method for detecting m2G sites, it is desirable to develop new methods to identify m2G sites. In this study, a computational predictor called iRNA-m2G was proposed to identify m2G sites in eukaryotic transcriptomes. In iRNA-m2G, the RNA sequences were encoded by using nucleotide chemical property and accumulated nucleotide frequency. iRNA-m2G was not only validated by the rigorous jackknife test on the benchmark dataset but also examined by performing cross-species validations. In addition, iRNA-m2G was also tested on an independent dataset. It was found that the accuracies obtained by iRNA-m2G were all quite promising in these tests, indicating that the proposed method could become a powerful tool for identifying m2G sites. Finally, a user-friendly web server for iRNA-m2G is freely accessible at http://lin-group.cn/server/iRNA-m2G.php.

m6AVar: a database of functional variants involved in m6A modification

Identifying disease-causing variants among a large number of single nucleotide variants (SNVs) is still a major challenge. Recently, N⁶-methyladenosine (m⁶A) has become a research hotspot because of its critical roles in many fundamental biological processes and a variety of diseases. Therefore, it is important to evaluate the effect of variants on m⁶A modification, in order to gain a better understanding of them. Here, we report m6AVar (http://m6avar.renlab.org), a comprehensive database of m⁶A-associated variants that potentially influence m⁶A modification, which will help to interpret variants by m⁶A function. The m⁶A-associated variants were derived from three different m⁶A sources including miCLIP/PA-m⁶A-seq experiments (high confidence), MeRIP-Seq experiments (medium confidence) and transcriptome-wide predictions (low confidence). Currently, m6AVar contains 16 132 high, 71 321 medium and 326 915 low confidence level m⁶A-associated variants. We also integrated the RBP-binding regions, miRNA-targets and splicing sites associated with variants to help users investigate the effect of m⁶A-associated variants on post-transcriptional regulation. Because it integrates the data from genome-wide association studies (GWAS) and ClinVar, m6AVar is also a useful resource for investigating the relationship between the m⁶A-associated variants and disease. Overall, m6AVar will serve as a useful resource for annotating variants and identifying disease-causing variants.

dreamBase: DNA modification, RNA regulation and protein binding of expressed pseudogenes in human health and disease

Although thousands of pseudogenes have been annotated in the human genome, their transcriptional regulation, expression profiles and functional mechanisms are largely unknown. In this study, we developed dreamBase (http://rna.sysu.edu.cn/dreamBase) to facilitate the investigation of DNA modification, RNA regulation and protein binding of potential expressed pseudogenes from multidimensional high-throughput sequencing data. Based on ∼5500 ChIP-seq and DNase-seq datasets, we identified genome-wide binding profiles of various transcription-associated factors around pseudogene loci. By integrating ∼18 000 RNA-seq data, we analysed the expression profiles of pseudogenes and explored their co-expression patterns with their parent genes in 32 cancers and 31 normal tissues. By combining microRNA binding sites, we demonstrated complex post-transcriptional regulation networks involving 275 microRNAs and 1201 pseudogenes. We generated ceRNA networks to illustrate the crosstalk between pseudogenes and their parent genes through competitive binding of microRNAs. In addition, we studied transcriptome-wide interactions between RNA binding proteins (RBPs) and pseudogenes based on 458 CLIP-seq datasets. In conjunction with epitranscriptome sequencing data, we also mapped 1039 RNA modification sites onto 635 pseudogenes. This database will provide insights into the transcriptional regulation, expression, functions and mechanisms of pseudogenes as well as their roles in biological processes and diseases.

MODOMICS: a database of RNA modification pathways. 2017 update

MODOMICS is a database of RNA modifications that provides comprehensive information concerning the chemical structures of modified ribonucleosides, their biosynthetic pathways, the location of modified residues in RNA sequences, and RNA-modifying enzymes. In the current database version, we included the following new features and data: extended mass spectrometry and liquid chromatography data for modified nucleosides; links between human tRNA sequences and MINTbase - a framework for the interactive exploration of mitochondrial and nuclear tRNA fragments; new, machine-friendly system of unified abbreviations for modified nucleoside names; sets of modified tRNA sequences for two bacterial species, updated collection of mammalian tRNA modifications, 19 newly identified modified ribonucleosides and 66 functionally characterized proteins involved in RNA modification. Data from MODOMICS have been linked to the RNAcentral database of RNA sequences. MODOMICS is available at http://modomics.genesilico.pl.

RMBase v2.0: deciphering the map of RNA modifications from epitranscriptome sequencing data

More than 100 distinct chemical modifications to RNA have been characterized so far. However, the prevalence, mechanisms and functions of various RNA modifications remain largely unknown. To provide transcriptome-wide landscapes of RNA modifications, we developed the RMBase v2.0 (http://rna.sysu.edu.cn/rmbase/), which is a comprehensive database that integrates epitranscriptome sequencing data for the exploration of post-transcriptional modifications of RNAs and their relationships with miRNA binding events, disease-related single-nucleotide polymorphisms (SNPs) and RNA-binding proteins (RBPs). RMBase v2.0 was expanded with ∼600 datasets and ∼1 397 000 modification sites from 47 studies among 13 species, which represents an approximately 10-fold expansion when compared with the previous release. It contains ∼1 373 000 N6-methyladenosines (m⁶A), ∼5400 N1-methyladenosines (m¹A), ∼9600 pseudouridine (Ψ) modifications, ∼1000 5-methylcytosine (m⁵C) modifications, ∼5100 2′-O-methylations (2′-O-Me), and ∼2800 modifications of other modification types. Moreover, we built a new module called ‘Motif’ that provides the visualized logos and position weight matrices (PWMs) of the modification motifs. We also constructed a novel module termed ‘modRBP’ to study the relationships between RNA modifications and RBPs. Additionally, we developed a novel web-based tool named ‘modMetagene’ to plot the metagenes of RNA modification along a transcript model. This database will help researchers investigate the potential functions and mechanisms of RNA modifications.

Epigenetics in Neurodevelopment: Emerging Role of Circular RNA

Canonical epigenetic modifications, including DNA methylation, histone modification and chromatin remodeling, play a role in numerous life processes, particularly neurodevelopment. Epigenetics explains the development of cells in an organism with the same DNA sequence into different cell types with various functions. However, previous studies on epigenetics have only focused on the chromatin level. Recently, epigenetic modifications of RNA, which mainly include 6-methyladenosine (m⁶A), pseudouridine, 5-methylcytidine (m⁵C), inosine (I), 2′-O-ribosemethylation, and 1-methyladenosine (m¹A), have gained increasing attention. Circular RNAs (circRNAs), which are a type of non-coding RNA without a 5′ cap or 3′ poly (A) tail, are abundantly found in the brain and might respond to and regulate synaptic function. Also, circRNAs have various functions, such as microRNA sponge, regulation of gene transcription and interaction with RNA binding protein. In addition, circRNAs are methylated by N⁶-methyladenosine (m⁶A). In this review, we discuss the crucial roles of epigenetic modifications of circRNAs, such as m⁶A, in the genesis and development of neurons and in synaptic function and plasticity. Thus, this type of changes in circRNAs might be a therapeutic target in central nervous system (CNS) disorders and could aid the diagnosis and treatment of these disorders.

Exploring the selective vulnerability in Alzheimer disease using tissue specific variant analysis

The selective vulnerability of distinct regions of the brain is a critical factor in neurodegenerative disorders. In Alzheimer's disease (AD), neurons in hippocampus situated in medial temporal lobe are immensely damaged. Identifying tissue-specific variants is essential in order to perceive the selective vulnerability in AD. In current work, we aligned mRNA-seq data with HG19/HG38 genomic assembly and identified specific variations present in temporal, frontal and other lobes of the AD using sequence alignment map tools. We compared the results with the genome-wide association and gene expression quantitative trait loci studies of the various neurological disorders. We also distinguished variants and epitranscriptomic modifications through the RNA-modification database and evaluated the variant effect in the coding/UTR regions. In addition, we developed genetic and functional interaction networks to understand the relationship between predicted vulnerable variations and differentially expressed genes. We found that genes involved in gliogenesis, intermediate filament organization are altered in the temporal lobe. Oxidative phosphorylation, and calcium ion homeostasis are modified in the frontal lobe, and protein degradation, apoptotic signaling are altered in other lobes. From this study, we propose that disruption of glial cell structural integrity, defective gliogenesis, and failure in glia-neuron communication are the primary factors for selective vulnerability.

Evaluation of different computational methods on 5-methylcytosine sites identification

5-Methylcytosine (m5C) plays an extremely important role in the basic biochemical process. With the great increase of identified m5C sites in a wide variety of organisms, their epigenetic roles become largely unknown. Hence, accurate identification of m5C site is a key step in understanding its biological functions. Over the past several years, more attentions have been paid on the identification of m5C sites in multiple species. In this work, we firstly summarized the current progresses in computational prediction of m5C sites and then constructed a more powerful and reliable model for identifying m5C sites. To train the model, we collected experimentally confirmed m5C data from Homo sapiens, Mus musculus, Saccharomyces cerevisiae and Arabidopsis thaliana, and compared the performances of different feature extraction methods and classification algorithms for optimizing prediction model. Based on the optimal model, a novel predictor called iRNA-m5C was developed for the recognition of m5C sites. Finally, we critically evaluated the performance of iRNA-m5C and compared it with existing methods. The result showed that iRNA-m5C could produce the best prediction performance. We hope that this paper could provide a guide on the computational identification of m5C site and also anticipate that the proposed iRNA-m5C will become a powerful tool for large scale identification of m5C sites.

Molecular windows into the human brain for psychiatric disorders

Delineating the pathophysiology of psychiatric disorders has been extremely challenging but technological advances in recent decades have facilitated a deeper interrogation of molecular processes in the human brain. Initial candidate gene expression studies of the postmortem brain have evolved into genome wide profiling of the transcriptome and the epigenome, a critical regulator of gene expression. Here, we review the potential and challenges of direct molecular characterization of the postmortem human brain, and provide a brief overview of recent transcriptional and epigenetic studies with respect to neuropsychiatric disorders. Such information can now be leveraged and integrated with the growing number of genome-wide association databases to provide a functional context of trait-associated genetic variants linked to psychiatric illnesses and related phenotypes. While it is clear that the field is still developing and challenges remain to be surmounted, these recent advances nevertheless hold tremendous promise for delineating the neurobiological underpinnings of mental diseases and accelerating the development of novel medication strategies.