Computational Methods and Applications for Identifying Disease-Associated lncRNAs as Potential Biomarkers and Therapeutic Targets

GCN-Based Heterogeneous Complex Feature Learning to Enhance Predictability for LncRNA-Disease Associations

Using computational models to predict potential lncRNA-disease associations (LDAs) has emerged as an effective supplement to bioexperiments for exploring the pathogenesis of diseases. However, current computational models still face limitations in their ability to learn the complex features of bionetworks. In this study, HGCNLDA, a model which combines graph convolutional network (GCN)-based aggregation, heterogeneous information fusion, and a bilinear-decoder to infer LDAs was proposed. Recognizing the need to extract essential features during data processing, our HGCNLDA explored four key steps for uncovering interaction patterns within the bionetwork: (1) a novel type of tripartite heterogeneous network, known as the lncRNA-disease-miRNA network (LDMN), was constructed using computed similarities and known associations. (2) Homogeneous and heterogeneous features of nodes were extracted from domains within the LDMN by a GCN-based encoder. (3) Feature fusions, including bipolymerization operations and attention mechanism, were employed to capture a more accurate and comprehensive representation of nodes. (4) Bilinear-decoder was used to rebuild the edge type (or rating type) for a specific node pair, resulting in the predicted association score. Through a 5-fold cross-validation on two data sets, namely, data set1 and data set2, our HGCNLDA consistently demonstrated superior performance compared to five related models. It almost achieved the highest AUROC and AUPR values on both data sets, especially on data set2 where the results obtained were more challenging and objective. Case studies involving three real cancer scenarios further validated the practicality of HGCNLDA in identifying potential LDAs in real-world contexts. The source code and data for this study are available at https://github.com/zywait/HGCNLDA.

Multi-Omics Mining of lncRNAs with Biological and Clinical Relevance in Cancer

In this review, we provide a general overview of the current panorama of mining strategies for multi-omics data to investigate lncRNAs with an actual or potential role as biological markers in cancer. Several multi-omics studies focusing on lncRNAs have been performed in the past with varying scopes. Nevertheless, many questions remain regarding the pragmatic application of different molecular technologies and bioinformatics algorithms for mining multi-omics data. Here, we attempt to address some of the less discussed aspects of the practical applications using different study designs for incorporating bioinformatics and statistical analyses of multi-omics data. Finally, we discuss the potential improvements and new paradigms aimed at unraveling the role and utility of lncRNAs in cancer and their potential use as molecular markers for cancer diagnosis and outcome prediction.

Specific topology and topological connection sensitivity enhanced graph learning for lncRNA-disease association prediction

Predicting disease-related candidate long noncoding RNAs (lncRNAs) is beneficial for exploring disease pathogenesis due to the close relations between lncRNAs and the occurrence and development of human diseases. It is a long-term and challenging task to adequately extract specific and local topologies in individual lncRNA network and individual disease network, and integrate the information of the connection relationships. We propose a new graph learning-based prediction method to encode specific and local topologies from each individual network, neighbor topologies with different connection relationships, and pairwise attributes. We first construct a lncRNA network composed of all the lncRNA nodes and their similarities, and a single disease network that contains all the disease nodes and disease similarities. Then, a network-aware graph convolutional autoencoder is constructed to encode the specific and local topologies of each network. Secondly, a heterogeneous network is established to embed all lncRNA, disease, and miRNA nodes and their various connections. Afterwards, a connection-sensitive graph neural network is designed to deeply integrate the neighbor node attributes and connection characteristics in the heterogeneous network and learn neighbor topological representations. We also construct both connection-level and topology representation-level attention mechanisms to extract informative connections and topological representations. Finally, we build a multi-layer convolutional neural networks with weighted residuals to adaptively complement the detailed features to pairwise attribute encoding. Comprehensive experiments and comparison results demonstrated that NCPred outperforms seven state-of-the-art prediction methods. The ablation studies demonstrated the importance of local topology learning, neighbor topology learning, and pairwise attribute encoding. Case studies on prostate, lung, and breast cancers further revealed NCPred's capacity to screen potential candidate disease-related lncRNAs.

XGBoost framework with feature selection for the prediction of RNA N5-methylcytosine sites

5-methylcytosine (m5C) is indeed a critical post-transcriptional alteration that is widely present in various kinds of RNAs and is crucial to the fundamental biological processes. By correctly identifying the m5C-methylation sites on RNA, clinicians can more clearly comprehend the precise function of these m5C-sites in different biological processes. Due to their effectiveness and affordability, computational methods have received greater attention over the last few years for the identification of methylation sites in various species. To precisely identify RNA m5C locations in five different species including Homo sapiens, Arabidopsis thaliana, Mus musculus, Drosophila melanogaster, and Danio rerio, we proposed a more effective and accurate model named m5C-pred. To create m5C-pred, five distinct feature encoding techniques were combined to extract features from the RNA sequence, and then we used SHapley Additive exPlanations to choose the best features among them, followed by XGBoost as a classifier. We applied the novel optimization method called Optuna to quickly and efficiently determine the best hyperparameters. Finally, the proposed model was evaluated using independent test datasets, and we compared the results with the previous methods. Our approach, m5C- pred, is anticipated to be useful for accurately identifying m5C sites, outperforming the currently available state-of-the-art techniques.

Recent advances in predicting lncRNA-disease associations based on computational methods

Mutations in and dysregulation of long non-coding RNAs (lncRNAs) are closely associated with the development of various human complex diseases, but only a few lncRNAs have been experimentally confirmed to be associated with human diseases. Predicting new potential lncRNA-disease associations (LDAs) will help us to understand the pathogenesis of human diseases and to detect disease markers, as well as in disease diagnosis, prevention and treatment. Computational methods can effectively narrow down the screening scope of biological experiments, thereby reducing the duration and cost of such experiments. In this review, we outline recent advances in computational methods for predicting LDAs, focusing on LDA databases, lncRNA/disease similarity calculations, and advanced computational models. In addition, we analyze the limitations of various computational models and discuss future challenges and directions for development.

Long non-coding ribonucleic acid urothelial carcinoma-associated 1 promotes high glucose-induced human retinal endothelial cells angiogenesis through regulating micro-ribonucleic acid-624-3p/vascular endothelial growth factor C

AIMS/INTRODUCTION

Emerging evidence has indicated that long non-coding ribonucleic acids play important roles in the development and progression of diabetic retinopathy (DR). It is reported that urothelial carcinoma-associated 1 (UCA1) is highly expressed in diabetic lymphoendothelial cells and influences glucose metabolism in rats with DR. The aim of the present study was to explore the role of UCA1 in the mechanism of DR.

MATERIALS AND METHODS

Gene expression analyses in fibrovascular membranes excised from patients with DR using public microarray datasets (GSE60436). Reverse transcription polymerase chain reaction was carried out to detect UCA1, micro-ribonucleic acid (miR)-624-3p and vascular endothelial growth factor C (VEGF-C) expressions in the blood of patients and human retinal endothelial cells (HRECs). Furthermore, Cell Counting kit-8, Transwell assay, and tube formation assay were used to identify biological effects of UCA1 on HRECs proliferation, migration ability and angiogenesis in vitro.

RESULTS

UCA1 and VEGF-C were elevated in DR patients and high glucose-induced HRECs cell lines, whereas miR-624-3p was decreased. UCA1 inhibition inhibited proliferation, angiogenesis and migration of HRECs cells under high-glucose condition. Luciferase reporter assay showed that UCA1 could sponge with miR-624-3p, which could directly target VEGF-C. Finally, we proved a pathway that UCA1 promoted cell proliferation, migration and angiogenesis through sponging with miR-624-3p, thereby upregulating VEGF-C in high-glucose-induced HRECs.

CONCLUSIONS

We identified UCA1 as an important factor associated with DR, which could regulate the expression of VEGF-C by sponging miR-624-3p in human retinal endothelial cells. Our results pave the way for further studies on diagnostic and therapeutic studies related to UCA1 in DR patients.

Long non-coding RNAs in metabolic diseases: from bench to bedside

Long non-coding RNAs (lncRNAs) are being widely studied for their implications in physiological and diseases contexts but their functions and therapeutic potentials in metabolic diseases are far from clarified. Here, we report a summary of current advances in lncRNAs identification, functions, role as biomarkers and therapeutic prospects in metabolic diseases.

Predicting LncRNA-Disease Association by a Random Walk With Restart on Multiplex and Heterogeneous Networks

Studies have found that long non-coding RNAs (lncRNAs) play important roles in many human biological processes, and it is critical to explore potential lncRNA–disease associations, especially cancer-associated lncRNAs. However, traditional biological experiments are costly and time-consuming, so it is of great significance to develop effective computational models. We developed a random walk algorithm with restart on multiplex and heterogeneous networks of lncRNAs and diseases to predict lncRNA–disease associations (MHRWRLDA). First, multiple disease similarity networks are constructed by using different approaches to calculate similarity scores between diseases, and multiple lncRNA similarity networks are also constructed by using different approaches to calculate similarity scores between lncRNAs. Then, a multiplex and heterogeneous network was constructed by integrating multiple disease similarity networks and multiple lncRNA similarity networks with the lncRNA–disease associations, and a random walk with restart on the multiplex and heterogeneous network was performed to predict lncRNA–disease associations. The results of Leave-One-Out cross-validation (LOOCV) showed that the value of Area under the curve (AUC) was 0.68736, which was improved compared with the classical algorithm in recent years. Finally, we confirmed a few novel predicted lncRNAs associated with specific diseases like colon cancer by literature mining. In summary, MHRWRLDA contributes to predict lncRNA–disease associations.

Predicting LncRNA–Disease Association by a Random Walk With Restart on Multiplex and Heterogeneous Networks

Studies have found that long non-coding RNAs (lncRNAs) play important roles in many human biological processes, and it is critical to explore potential lncRNA–disease associations, especially cancer-associated lncRNAs. However, traditional biological experiments are costly and time-consuming, so it is of great significance to develop effective computational models. We developed a random walk algorithm with restart on multiplex and heterogeneous networks of lncRNAs and diseases to predict lncRNA–disease associations (MHRWRLDA). First, multiple disease similarity networks are constructed by using different approaches to calculate similarity scores between diseases, and multiple lncRNA similarity networks are also constructed by using different approaches to calculate similarity scores between lncRNAs. Then, a multiplex and heterogeneous network was constructed by integrating multiple disease similarity networks and multiple lncRNA similarity networks with the lncRNA–disease associations, and a random walk with restart on the multiplex and heterogeneous network was performed to predict lncRNA–disease associations. The results of Leave-One-Out cross-validation (LOOCV) showed that the value of Area under the curve (AUC) was 0.68736, which was improved compared with the classical algorithm in recent years. Finally, we confirmed a few novel predicted lncRNAs associated with specific diseases like colon cancer by literature mining. In summary, MHRWRLDA contributes to predict lncRNA–disease associations.

A systematic evaluation of the computational tools for lncRNA identification

The computational identification of long non-coding RNAs (lncRNAs) is important to study lncRNAs and their functions. Despite the existence of many computation tools for lncRNA identification, to our knowledge, there is no systematic evaluation of these tools on common datasets and no consensus regarding their performance and the importance of the features used. To fill this gap, in this study, we assessed the performance of 17 tools on several common datasets. We also investigated the importance of the features used by the tools. We found that the deep learning-based tools have the best performance in terms of identifying lncRNAs, and the peptide features do not contribute much to the tool accuracy. Moreover, when the transcripts in a cell type were considered, the performance of all tools significantly dropped, and the deep learning-based tools were no longer as good as other tools. Our study will serve as an excellent starting point for selecting tools and features for lncRNA identification.

Identification of Rhythmically Expressed LncRNAs in the Zebrafish Pineal Gland and Testis

Noncoding RNAs have been known to contribute to a variety of fundamental life processes, such as development, metabolism, and circadian rhythms. However, much remains unrevealed in the huge noncoding RNA datasets, which require further bioinformatic analysis and experimental investigation-and in particular, the coding potential of lncRNAs and the functions of lncRNA-encoded peptides have not been comprehensively studied to date. Through integrating the time-course experimentation with state-of-the-art computational techniques, we studied tens of thousands of zebrafish lncRNAs from our own experiments and from a published study including time-series transcriptome analyses of the testis and the pineal gland. Rhythmicity analysis of these data revealed approximately 700 rhythmically expressed lncRNAs from the pineal gland and the testis, and their GO, COG, and KEGG pathway functions were analyzed. Comparative and conservative analyses determined 14 rhythmically expressed lncRNAs shared between both the pineal gland and the testis, and 15 pineal gland lncRNAs as well as 3 testis lncRNAs conserved among zebrafish, mice, and humans. Further, we computationally analyzed the conserved lncRNA-encoded peptides, and revealed three pineal gland and one testis lncRNA-encoded peptides conserved among these three species, which were further investigated for their three-dimensional (3D) structures and potential functions. Our computational findings provided novel annotations and regulatory mechanisms for hundreds of rhythmically expressed pineal gland and testis lncRNAs in zebrafish, and set the stage for their experimental studies in the near future.

Circulating exosomal miR-363-5p inhibits lymph node metastasis by downregulating PDGFB and serves as a potential noninvasive biomarker for breast cancer

Sentinel lymph node (LN) biopsy is currently the standard procedure for clinical LN-negative breast cancer (BC) patients but it is prone to false-negative results and complications. Thus, an accurate noninvasive approach for LN staging is urgently needed in clinical practice. Here, circulating exosomal microRNA (miRNA) expression profiles in peripheral blood from BC patients and age-matched healthy women were obtained and analyzed. We identified an exosomal miRNA, miR-363-5p, that was significantly downregulated in exosomes from plasma of BC patients with LN metastasis which exhibited a consistent decreasing trend in tissue samples from multiple independent datasets. Plasma exosomal miR-363-5p achieved high diagnostic performance in distinguishing LN-positive patients from LN-negative patients. The high miR-363-5p expression level was significantly correlated with improved overall survival. Functional assays demonstrated that exosomal miR-363-5p modulates platelet-derived growth factor (PDGF) signaling activity by targeting PDGFB to inhibit cell proliferation and migration. Our study revealed, for the first time, plasma exosomal miR-363-5p plays a tumor suppressor role in BC and has the potential for noninvasive LN staging and prognosis prediction of BC.

Long noncoding RNAs in triple-negative breast cancer: A new frontier in the regulation of tumorigenesis

In recent years, triple-negative breast cancer (TNBC) has emerged as the most aggressive subtype of breast cancer and is usually associated with increased mortality worldwide. The severity of TNBC is primarily observed in younger women, with cases ranging from approximately 12%-24% of all breast cancer cases. The existing hormonal therapies offer limited clinical solutions in completely circumventing the TNBC, with chemoresistance and tumor recurrences being the common hurdles in the path of TNBC treatment. Accumulating evidence has correlated the dysregulation of long noncoding RNAs (lncRNAs) with increased cell proliferation, invasion, migration, tumor growth, chemoresistance, and decreased apoptosis in TNBC. Various clinical studies have revealed that aberrant expression of lncRNAs in TNBC tissues is associated with poor prognosis, lower overall survival, and disease-free survival. Due to these specific characteristics, lncRNAs have emerged as novel diagnostic and prognostic biomarkers for TNBC treatment. However, the underlying mechanism through which lncRNAs perform their actions remains unclear, and extensive research is being carried out to reveal it. Therefore, understanding of mechanisms regulating the modulation of lncRNAs will be a substantial breakthrough in effective treatment therapies for TNBC. This review highlights the association of several lncRNAs in TNBC progression and treatment, along with their possible functions and mechanisms.

Multi-Omics Data Analyses Construct a Six Immune-Related Genes Prognostic Model for Cervical Cancer in Tumor Microenvironment

The cross-talk between tumor cells and the tumor microenvironment (TME) is an important factor in determining the tumorigenesis and progression of cervical cancer (CC). However, clarifying the potential mechanisms which trigger the above biological processes remains a challenge. The present study focused on immune-relevant differences at the transcriptome and somatic mutation levels through an integrative multi-omics analysis based on The Cancer Genome Atlas database. The objective of the study was to recognize the specific immune-related prognostic factors predicting the survival and response to immunotherapy of patients with CC. Firstly, eight hub immune-related prognostic genes were ultimately identified through construction of a protein–protein interaction network and Cox regression analysis. Secondly, 32 differentially mutated genes were simultaneously identified based on the different levels of immune infiltration. As a result, an immune gene-related prognostic model (IGRPM), including six factors (chemokine receptor 7 [CCR7], CD3d molecule [CD3D], CD3e molecule [CD3E], and integrin subunit beta 2 [ITGB2], family with sequence similarity 133 member A [FAM133A], and tumor protein p53 [TP53]), was finally constructed to forecast clinical outcomes of CC. Its predictive capability was further assessed and validated using the Gene Expression Omnibus validation set. In conclusion, IGRPM may be a promising prognostic signature to predict the prognoses and responses to immunotherapy of patients with CC. Moreover, the multi-omics study showed that IGRPM could be a novel therapeutic target for CC, which is a promising biomarker for indicating the immune-dominant status of the TME and revealing the potential mechanisms responsible for the tumorigenesis and progression of CC.

Knockdown of lncRNA DLEU1 inhibits the tumorigenesis of oral squamous cell carcinoma via regulation of miR‑149‑5p/CDK6 axis

Oral squamous cell carcinoma (OSCC) is a frequent malignant tumor worldwide. Long non-coding RNAs (lncRNAs) are known to play key roles in different types of cancer, including OSCC. It was previously reported that lncRNA deleted in lymphocytic leukemia 1 (DLEU1) is notably upregulated in OSCC; however, the role of DLEU1 in OSCC remains unclear. Gene and protein expression levels in OSCC cells were detected by reverse transcription-quantitative PCR and western blotting, respectively, in the present study. A Transwell assay was performed to measure cell migration and invasion. Flow cytometry was used to detect cell apoptosis, and the dual-luciferase reporter assay was applied to confirm the interaction between DLEU1, microRNA (miR)-149-5p and CDK6 in OSCC cells. DLEU1 expression was negatively associated with the survival rate of patients with OSCC. In addition, silencing of DLEU1 notably inhibited the proliferation of OSCC cells by inducing apoptosis. Meanwhile, DLEU1 directly bound to miR-149-5p, and CDK6 was found to be the direct target of miR-149-5p. Furthermore, DLEU1 knockdown-induced inhibition of OSCC cell proliferation was significantly reversed by the miR-149-5p antagomir. Knockdown of lncRNA DLEU1 reversed the proliferation of OSCC cells via regulation of the miR-149-5p/CDK6 axis. Thus, DLEU1 may serve as a novel target for treating OSCC.

Integrative Analysis of DNA Methylation Data and Transcriptome Data Identified a DNA Methylation-Dysregulated Four-LncRNA Signature for Predicting Prognosis in Head and Neck Squamous Cell Carcinoma

Increasing evidence has demonstrated the crosstalk between DNA epigenetic alterations and aberrant expression of long non-coding RNAs (lncRNAs) during carcinogenesis. However, epigenetically dysregulated lncRNAs and their functional and clinical roles in Head and Neck Squamous Cell Carcinoma (HNSCC) are still not explored. In this study, we performed an integrative analysis of DNA methylation data and transcriptome data and identified a DNA methylation-dysregulated four-lncRNA signature (DNAMeFourLncSig) from 596 DNA methylation-dysregulated lncRNAs using a machine-learning-based feature selection method, which classified the patients of the discovery cohort into two risk groups with significantly different survival including overall survival, disease-specific survival, and progression-free survival. Then the DNAMeFourLncSig was implemented to another two HNSCC patient cohorts and showed similar prognostic values in both. Results from multivariable Cox regression analysis revealed that the DNAMeFourLncSig might be an independent prognostic factor. Furthermore, the DNAMeFourLncSig was substantially correlated with the complete response rate of chemotherapy and may predict chemotherapy response. Functional in silico analysis found that DNAMeFourLncSig-related mRNAs were mainly enriched in cell differentiation, tissue development and immune-related pathways. Overall, our study will improve our understanding of underlying transcriptional and epigenetic mechanisms in HNSCC carcinogenesis and provided a new potential biomarker for the prognosis of patients with HNSCC.

Novel lncRNAs Co-Expression Networks Identifies LINC00504 with Oncogenic Role in Luminal A Breast Cancer Cells

Long non-coding RNAs (lncRNAs) are functional transcripts with more than 200 nucleotides. These molecules exhibit great regulatory capacity and may act at different levels of gene expression regulation. Despite this regulatory versatility, the biology of these molecules is still poorly understood. Computational approaches are being increasingly used to elucidate biological mechanisms in which these lncRNAs may be involved. Co-expression networks can serve as great allies in elucidating the possible regulatory contexts in which these molecules are involved. Herein, we propose the use of the pipeline deposited in the RTN package to build lncRNAs co-expression networks using TCGA breast cancer (BC) cohort data. Worldwide, BC is the most common cancer in women and has great molecular heterogeneity. We identified an enriched co-expression network for the validation of relevant cell processes in the context of BC, including LINC00504. This lncRNA has increased expression in luminal subtype A samples, and is associated with prognosis in basal-like subtype. Silencing this lncRNA in luminal A cell lines resulted in decreased cell viability and colony formation. These results highlight the relevance of the proposed method for the identification of lncRNAs in specific biological contexts.

6mA-Pred: identifying DNA N6-methyladenine sites based on deep learning

With the accumulation of data on 6mA modification sites, an increasing number of scholars have begun to focus on the identification of 6mA sites. Despite the recognized importance of 6mA sites, methods for their identification remain lacking, with most existing methods being aimed at their identification in individual species. In the present study, we aimed to develop an identification method suitable for multiple species. Based on previous research, we propose a method for 6mA site recognition. Our experiments prove that the proposed 6mA-Pred method is effective for identifying 6mA sites in genes from taxa such as rice, Mus musculus, and human. A series of experimental results show that 6mA-Pred is an excellent method. We provide the source code used in the study, which can be obtained from http://39.100.246.211:5004/6mA_Pred/.

A prognosis-related based method for miRNA selection on liver hepatocellular carcinoma prediction

Hepatocellular carcinoma (HCC) is considered as the sixth most common cancer in the world, and it is also considered as one of the causes of death. Moreover, the poor prognosis of recurrence of HCC after surgery and metastasis is also a big problem for human health. If the disease can be diagnosed earlier, the survival rate of the patients will be improved significantly. In the early stage of hepatocellular carcinoma, the expression of miRNAs is likely to become abnormal. In our work, the expression profile of miRNAs of human HCC in cancer tissue is compared with their adjacent tissue samples collected from tumor cancer genomic Atlas (TCGA) platform, then the genes with significant difference are selected by Limma test. Selected genes are referred to predict miRNAs related to the prognosis of HCC patients. Finally, miRNAs regulated by target genes are selected by our method, and the experimental results demonstrated that our method is more efficient than biology wet experimental method with lower cost.

Machine learning-based integrative analysis of methylome and transcriptome identifies novel prognostic DNA methylation signature in uveal melanoma

Uveal melanoma (UVM) is the most common primary intraocular human malignancy with a high mortality rate. Aberrant DNA methylation has rapidly emerged as a diagnostic and prognostic signature in many cancers. However, such DNA methylation signature available in UVM remains limited. In this study, we performed a genome-wide integrative analysis of methylome and transcriptome and identified 40 methylation-driven prognostic genes (MDPGs) associated with the tumorigenesis and progression of UVM. Then, we proposed a machine-learning-based discovery and validation strategy to identify a DNA methylation-driven signature (10MeSig) composing of 10 MDPGs (AZGP1, BAI1, CCDC74A, FUT3, PLCD1, S100A4, SCN8A, SEMA3B, SLC25A38 and SLC44A3), which stratified 80 patients of the discovery cohort into two risk subtypes with significantly different overall survival (HR = 29, 95% CI: 6.7-126, P < 0.001). The 10MeSig was validated subsequently in an independent cohort with 57 patients and yielded a similar prognostic value (HR = 2.1, 95% CI: 1.2-3.7, P = 0.006). Multivariable Cox regression analysis showed that the 10MeSig is an independent predictive factor for the survival of patients with UVM. With a prospective validation study, this 10MeSig will improve clinical decisions and provide new insights into the pathogenesis of UVM.

Transcriptomic Data Analyses Reveal a Reprogramed Lipid Metabolism in HCV-Derived Hepatocellular Cancer

Reprograming lipid metabolism, one of the major metabolic alterations in cancer, is believed to play an essential role in cancer development, but the exact molecular mechanism remains elusive. Here, we present a computational study of transcriptomic data of HCC with HCV etiology to investigate how lipid metabolism alters during HCC progression. Our analyses reveal that: (1) cancer tissue cells tend to synthesize fatty acids de novo and its phospholipid derivatives; (2) lipid catabolism and fatty acid oxidation are remarkably down-regulated in HCC; (3) the lipid metabolism in HCC is largely independent of lipids in blood circulation; (4) stage-specific co-expression networks for lipid metabolic genes were identified during HCC progression; and (5) the expression levels of several lipid metabolic genes that are differentially expressed or co-expressed specifically at the HCC stage have a strong correlation with cancer survival. Overall, the results provide detailed information about the reprogramed lipid metabolism in HCV-derived HCC.

LDNFSGB: prediction of long non-coding rna and disease association using network feature similarity and gradient boosting

BACKGROUND

A large number of experimental studies show that the mutation and regulation of long non-coding RNAs (lncRNAs) are associated with various human diseases. Accurate prediction of lncRNA-disease associations can provide a new perspective for the diagnosis and treatment of diseases. The main function of many lncRNAs is still unclear and using traditional experiments to detect lncRNA-disease associations is time-consuming.

RESULTS

In this paper, we develop a novel and effective method for the prediction of lncRNA-disease associations using network feature similarity and gradient boosting (LDNFSGB). In LDNFSGB, we first construct a comprehensive feature vector to effectively extract the global and local information of lncRNAs and diseases through considering the disease semantic similarity (DISSS), the lncRNA function similarity (LNCFS), the lncRNA Gaussian interaction profile kernel similarity (LNCGS), the disease Gaussian interaction profile kernel similarity (DISGS), and the lncRNA-disease interaction (LNCDIS). Particularly, two methods are used to calculate the DISSS (LNCFS) for considering the local and global information of disease semantics (lncRNA functions) respectively. An autoencoder is then used to reduce the dimensionality of the feature vector to obtain the optimal feature parameter from the original feature set. Furthermore, we employ the gradient boosting algorithm to obtain the lncRNA-disease association prediction.

CONCLUSIONS

In this study, hold-out, leave-one-out cross-validation, and ten-fold cross-validation methods are implemented on three publicly available datasets to evaluate the performance of LDNFSGB. Extensive experiments show that LDNFSGB dramatically outperforms other state-of-the-art methods. The case studies on six diseases, including cancers and non-cancers, further demonstrate the effectiveness of our method in real-world applications.

Characterization of DNA Methylation Patterns and Mining of Epigenetic Markers During Genomic Reprogramming in SCNT Embryos

Somatic cell nuclear transfer (SCNT), also known as somatic cell cloning, is a commonly used technique to study epigenetic reprogramming. Although SCNT has the advantages of being safe and able to obtain pluripotent cells, early developmental arrest happens in most SCNT embryos. Overcoming epigenetic barriers is currently the primary strategy for improving reprogramming efficiency and improving developmental rate in SCNT embryos. In this study, we analyzed DNA methylation profiles of in vivo fertilized embryos and SCNT embryos with different developmental fates. Overall DNA methylation level was higher in SCNT embryos during global de-methylation process compared to in vivo fertilized embryos. In addition, promoter region, first intron and 3′UTR were found to be the major genomic regions that were hyper-methylated in SCNT embryos. Surprisingly, we found the length of re-methylated region was directly related to the change of methylation level. Furthermore, a number of genes including Dppa2 and Dppa4 which are important for early zygotic genome activation (ZGA) were not properly activated in SCNT embryos. This study comprehensively analyzed genome-wide DNA methylation patterns in SCNT embryos and provided candidate target genes for improving efficiency of genomic reprogramming in SCNT embryos. Since SCNT technology has been widely used in agricultural and pastoral production, protection of endangered animals, and therapeutic cloning, the findings of this study have significant importance for all these fields.