scholarly journals Attention-based deep multiple instances learning for classifying circular RNA and other long non-coding RNA

2021 ◽  
Author(s):  
Yunhe Liu ◽  
Qiqing Fu ◽  
Xueqing Peng ◽  
Chaoyu Zhu ◽  
Gang Liu ◽  
...  
Keyword(s):  
Network Architecture ◽  
Enrichment Analysis ◽  
Circular Rna ◽  
Biological Processes ◽  
Non Coding Rna ◽  
Strong Representation ◽  
Handwritten Digit ◽  
Deep Learning Network ◽  
Sparse Features ◽  
Long Non Coding Rna

Circular RNA (circRNA) is a distinguishable circular formed long non-coding RNA (lncRNA), which has specific roles in transcriptional regulation, multiple biological processes. The identification of circRNA from other lncRNA is necessary for relevant research. In this study, we designed attention-based multi-instance learning (MIL) network architecture, which can be fed with raw sequence, to learn the sparse features in sequences and accomplish the identification task for circRNAs. The model outperformed previously reported models. Following the effectiveness validation of the attention score by the handwritten digit dataset, the key sequence loci underlying circRNAs recognition were obtained based on the corresponding attention score. Moreover, the motif enrichment analysis of the extracted key sequences identified some of the key motifs for circRNA formation. In conclusion, we designed a deep learning network architecture suitable for gene sequence learning with sparse features and implemented to the circRNA identification, and the network has a strong representation capability with its indication of some key loci.

scholarly journals Attention-Based Deep Multiple Instances Learning For Classifying Circular RNA and Other Long Non-Coding RNA

2021 ◽  
Author(s):  
Yunhe Liu ◽  
Qiqing Fu ◽  
Xueqing peng ◽  
Chaoyu Zhu ◽  
Gang Liu ◽  
...  
Keyword(s):  
Network Architecture ◽  
Enrichment Analysis ◽  
Circular Rna ◽  
Biological Processes ◽  
Non Coding Rna ◽  
Strong Representation ◽  
Handwritten Digit ◽  
Deep Learning Network ◽  
Sparse Features ◽  
Long Non Coding Rna

Abstract Circular RNA (circRNA) is a distinguishable circular formed long non-coding RNA (lncRNA), which has specific roles in transcriptional regulation, multiple biological processes. The identification of circRNA from other lncRNA is necessary for relevant research. In this study, we designed attention-based multi-instance learning (MIL) network architecture, which can be fed with raw sequence, to learn the sparse features in sequences and accomplish the identification task for circRNAs. The model outperformed previously reported models. Following the effectiveness validation of the attention score by the handwritten digit dataset, the key sequence loci underlying circRNAs recognition were obtained based on the corresponding attention score. Moreover, the motif enrichment analysis of the extracted key sequences identified some of the key motifs for circRNA formation. In conclusion, we designed a deep learning network architecture suitable for gene sequence learning with sparse features and implemented to the circRNA identification, and the network has a strong representation capability with its indication of some key loci.

scholarly journals Attention-Based Deep Multiple-Instance Learning for Classifying Circular RNA and Other Long Non-Coding RNA

Genes ◽  
2021 ◽  
Vol 12 (12) ◽  
pp. 2018
Author(s):  
Yunhe Liu ◽  
Qiqing Fu ◽  
Xueqing Peng ◽  
Chaoyu Zhu ◽  
Gang Liu ◽  
...  
Keyword(s):  
Network Architecture ◽  
Enrichment Analysis ◽  
Circular Rna ◽  
Multiple Instance Learning ◽  
Identification Task ◽  
Rna Sequences ◽  
Non Coding Rna ◽  
Deep Learning Network ◽  
Sparse Features ◽  
Long Non Coding Rna

Circular RNA (circRNA) is a distinguishable circular formed long non-coding RNA (lncRNA), which has specific roles in transcriptional regulation, multiple biological processes. The identification of circRNA from other lncRNA is necessary for relevant research. In this study, we designed attention-based multi-instance learning (MIL) network architecture fed with a raw sequence, to learn the sparse features of RNA sequences and to accomplish the circRNAs identification task. The model outperformed the state-of-art models. Moreover, following the validation of the attention mechanism effectiveness by the handwritten digit dataset, the key sequence loci underlying circRNA’s recognition were obtained based on the corresponding attention score. Then, motif enrichment analysis identified some of the key motifs for circRNA formation. In conclusion, we designed deep learning network architecture suitable for learning gene sequences with sparse features and implemented it for the circRNA identification task, and the model has strong representation capability in the indication of some key loci.

CASC15：A tumor-associated long non-coding RNA

2020 ◽  
Vol 26 ◽  
Author(s):  
Bei Wang ◽  
Wen Xu ◽  
Yuxuan Cai ◽  
Chong Guo ◽  
Gang Zhou ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Therapeutic Target ◽  
Cancer Colorectal ◽  
Tumor Development ◽  
Pathophysiological Mechanism ◽  
Biological Processes ◽  
Non Coding Rna ◽  
Cancer Côlon ◽  
The Relationship ◽  
Long Non Coding Rna

Background: CASC15, one of long non-coding RNA, is involved in the regulation of many tumor biological processes, and is expected to become a new biological therapeutic target. This paper aims to elucidate the pathophysiological function of CASC15 in various tumors. Methods: The relationship between CASC15 and tumors was analyzed by searching references, and summarizes the specific pathophysiological mechanism of CASC15. Results: LncRNA CASC15 is closely related to tumor development, and has been shown to be abnormally high expressed in all kinds of tumors, including breast cancer, cervical cancer, lung cancer, hepatocellular carcinoma, gastric cancer, bladder cancer, colon cancer, colorectal cancer, cardiac hypertrophy, intrahepatic cholangiocarcinoma, leukemia, melanoma, tongue squamous cell carcinoma, nasopharyngeal carcinoma. However, CASC15 has been found to be downexpressed abnormally in ovarian cancer, glioma and neuroblastoma. Besides, it is identified that CASC15 can affect the proliferation, invasion and apoptosis of tumors. Conclusion: LncRNA CASC15 has the potential to become a new therapeutic target or marker for a variety of tumors.

scholarly journals Dynamic characteristics and functional analysis provide new insights into long non-coding RNA responsive to Verticillium dahliae infection in Gossypium hirsutum

2021 ◽  
Vol 21 (1) ◽  
Author(s):  
Guoning Wang ◽  
Xingfen Wang ◽  
Yan Zhang ◽  
Jun Yang ◽  
Zhikun Li ◽  
...  
Keyword(s):  
Disease Resistance ◽  
Verticillium Wilt ◽  
Dynamic Characteristics ◽  
Target Genes ◽  
Acid Metabolism ◽  
Expression Profiles ◽  
Enrichment Analysis ◽  
Alpha Linolenic Acid ◽  
Non Coding Rna ◽  
Long Non Coding Rna

Abstract Background Verticillium wilt is a widespread and destructive disease, which causes serious loss of cotton yield and quality. Long non-coding RNA (lncRNA) is involved in many biological processes, such as plant disease resistance response, through a variety of regulatory mechanisms, but their possible roles in cotton against Verticillium dahliae infection remain largely unclear. Results Here, we measured the transcriptome of resistant G. hirsutum following infection by V. dahliae and 4277 differentially expressed lncRNAs (delncRNAs) were identified. Localization and abundance analysis revealed that delncRNAs were biased distribution on chromosomes. We explored the dynamic characteristics of disease resistance related lncRNAs in chromosome distribution, induced expression profiles, biological function, and these lncRNAs were divided into three categories according to their induced expression profiles. For the delncRNAs, 687 cis-acting pairs and 14,600 trans-acting pairs of lncRNA-mRNA were identified, which indicated that trans-acting was the main way of Verticillium wilt resistance-associated lncRNAs regulating target mRNAs in cotton. Analyzing the regulation pattern of delncRNAs revealed that cis-acting and trans-acting lncRNAs had different ways to influence target genes. Gene Ontology (GO) enrichment analysis revealed that the regulatory function of delncRNAs participated significantly in stimulus response process, kinase activity and plasma membrane components. Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis indicated that delncRNAs participated in some important disease resistance pathways, such as plant-pathogen interaction, alpha-linolenic acid metabolism and plant hormone signal transduction. Additionally, 21 delncRNAs and 10 target genes were identified as being involved in alpha-linolenic acid metabolism associated with the biosynthesis of jasmonic acid (JA). Subsequently, we found that GhlncLOX3 might regulate resistance to V. dahliae through modulating the expression of GhLOX3 implicated in JA biosynthesis. Further functional analysis showed that GhlncLOX3-silenced seedlings displayed a reduced resistance to V. dahliae, with down-regulated expression of GhLOX3 and decreased content of JA. Conclusion This study shows the dynamic characteristics of delncRNAs in multiaspect, and suggests that GhlncLOX3-GhLOX3-JA network participates in response to V. dahliae invasion. Our results provide novel insights for genetic improvement of Verticillium wilt resistance in cotton using lncRNAs.

scholarly journals Therapies Targeted at Non-Coding RNAs in Prevention and Limitation of Myocardial Infarction and Subsequent Cardiac Remodeling—Current Experience and Perspectives

2021 ◽  
Vol 22 (11) ◽  
pp. 5718
Author(s):  
Michal Kowara ◽  
Sonia Borodzicz-Jazdzyk ◽  
Karolina Rybak ◽  
Maciej Kubik ◽  
Agnieszka Cudnoch-Jedrzejewska
Keyword(s):  
Myocardial Infarction ◽  
Atherosclerotic Plaque ◽  
Cardiac Remodeling ◽  
Therapeutic Option ◽  
Circular Rna ◽  
Plaque Progression ◽  
Cardiac Damage ◽  
Non Coding Rna ◽  
Causes Of Mortality ◽  
Long Non Coding Rna

Myocardial infarction is one of the major causes of mortality worldwide and is a main cause of heart failure. This disease appears as a final point of atherosclerotic plaque progression, destabilization, and rupture. As a consequence of cardiomyocytes death during the infarction, the heart undergoes unfavorable cardiac remodeling, which results in its failure. Therefore, therapies aimed to limit the processes of atherosclerotic plaque progression, cardiac damage during the infarction, and subsequent remodeling are urgently warranted. A hopeful therapeutic option for the future medicine is targeting and regulating non-coding RNA (ncRNA), like microRNA, circular RNA (circRNA), or long non-coding RNA (lncRNA). In this review, the approaches targeted at ncRNAs participating in the aforementioned pathophysiological processes involved in myocardial infarction and their outcomes in preclinical studies have been concisely presented.

scholarly journals Prognostic value of an immune long non-coding RNA signature in liver hepatocellular carcinoma

2019 ◽  
Author(s):  
rui kong ◽  
Nan Wang ◽  
Wei Han ◽  
Yuejuan Zheng ◽  
Jie Lu
Keyword(s):  
Hepatocellular Carcinoma ◽  
Principal Component ◽  
Enrichment Analysis ◽  
Risk Groups ◽  
Gene Set Enrichment Analysis ◽  
Kaplan Meier ◽  
Non Coding Rna ◽  
Liver Hepatocellular Carcinoma ◽  
Long Non Coding Rna

Abstract Background: In recent years, long non-coding RNAs (lncRNAs) are emerging as crucial regulators in the immunological process of liver hepatocellular carcinoma (LIHC). Increasing studies have found that some lncRNAs could be used as a diagnostic or therapeutic target for clinical management, but little research has investigated the role of immune-related lncRNA in tumor prognosis. In this study, we aimed to develop an immune lncRNA signature for the precise diagnosis and prognosis of liver hepatocellular carcinoma. Methods: Gene expression profiles of LIHC samples obtained from TCGA were screened for immune-related genes using two reference gene sets. The optimal immune-related lncRNA signature was built via correlational analysis, univariate and multivariate cox analysis. Then the Kaplan-Meier plot, ROC curve, clinical analysis, gene set enrichment analysis, and principal component analysis were carried out to evaluate the capability of immune lncRNA signature as a prognostic indicator. Results: Six long non-coding RNA MSC−AS1, AC009005.1, AL117336.3, AL031985.3, AL365203.2, AC099850.3 were identified via correlation analysis and cox regression analysis considering their interactions with immune genes. Next, tumor samples were separated into two risk groups by the signature with different clinical outcomes. Stratification analysis showed the prognostic ability of this signature acted as an independent factor. The AUC value of ROC curve was 0.779. The Kaplan-Meier method was used in survival analysis and results showed a statistical difference between the two risk groups. The predictive performance of this signature was validated by principal component analysis (PCA). Data from gene set enrichment analysis (GSEA) further unveiled several potential biological processes of these biomarkers may involve in. Conclusion: In summary, the study demonstrated the potential role of the six-lncRNA signature served as an independent prognostic factor for LIHC patients.

scholarly journals LINC01018 and SMIM25 sponged miR-182-5p in endometriosis revealed by the ceRNA network construction

2020 ◽  
Vol 34 ◽  
pp. 205873842097630
Author(s):  
Li Jiang ◽  
Mengmeng Zhang ◽  
Sixue Wang ◽  
Yuzhen Xiao ◽  
Jingni Wu ◽  
...  
Keyword(s):  
Messenger Rna ◽  
Kegg Pathway ◽  
Enrichment Analysis ◽  
Gene Set Enrichment Analysis ◽  
Differentially Expressed ◽  
Protein Protein Interaction ◽  
Cerna Network ◽  
Non Coding Rna ◽  
Differentially Expressed Mirnas ◽  
Long Non Coding Rna

The current study intended to explore the interaction of the long non-coding RNA (lncRNA), microRNA (miRNA), and messenger RNA (mRNA) under the background of competitive endogenous RNA (ceRNA) network in endometriosis (EMs). The differentially expressed miRNAs (DEmiRs), differentially expressed lncRNA (DELs), and differentially expressed genes (DEGs) between EMs ectopic (EC) and eutopic (EU) endometrium based on three RNA-sequencing datasets (GSE105765, GSE121406, and GSE105764) were identified, which were used for the construction of ceRNA network. Then, DEGs in the ceRNA network were performed with Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway, and protein-protein interaction (PPI) analysis. Besides, the DEmiRs in the ceRNA network were validated in GSE124010. And the target DELs and DEGs of verified DEmiRs were validated in GSE86534. The correlation of verified DEmiRs, DEGs, and DELs was explored. Moreover, gene set enrichment analysis (GSEA) was applied to investigate the function of verified DEmiRs, DEGs, and DELs. Overall, 1352 DEGs and 595 DELs from GSE105764, along with 27 overlapped DEmiRs between GSE105765 and GSE121406, were obtained. Subsequently, a ceRNA network, including 11 upregulated and 16 downregulated DEmiRs, 7 upregulated and 13 downregulated DELs, 48 upregulated and 46 downregulated DEGs, was constructed. The GO and KEGG pathway analysis showed that this ceRNA network probably was associated with inflammation-related pathways. Furthermore, hsa-miR-182-5p and its target DELs (LINC01018 and SMIM25) and DEGs (BNC2, CHL1, HMCN1, PRDM16) were successfully verified in the validation analysis. Besides, hsa-miR-182-5p was significantly negatively correlated with these target DELs and DEGs. The GSEA analysis implied that high expression of LINC01018, SMIM25, and CHL1, and low expression of hsa-miR-182-5p would activate inflammation-related pathways in endometriosis EU samples. LINC01018 and SMIM25 might sponge hsa-miR-182-5p to upregulate downstream genes such as CHL1 to promote the development of endometriosis.

scholarly journals Long Non-Coding RNAs in Multiple Myeloma

2019 ◽  
Vol 5 (1) ◽  
pp. 13 ◽  
Author(s):  
Romana Butova ◽  
Petra Vychytilova-Faltejskova ◽  
Adela Souckova ◽  
Sabina Sevcikova ◽  
Roman Hajek
Keyword(s):  
Multiple Myeloma ◽  
Plasma Cells ◽  
Current Knowledge ◽  
Biological Processes ◽  
Cell Distribution ◽  
Rna Molecules ◽  
Non Coding Rna ◽  
Non Coding Rnas ◽  
New Treatment ◽  
Long Non Coding Rna

Multiple myeloma (MM) is the second most common hematooncological disease of malignant plasma cells in the bone marrow. While new treatment brought unprecedented increase of survival of patients, MM pathogenesis is yet to be clarified. Increasing evidence of expression of long non-coding RNA molecules (lncRNA) linked to development and progression of many tumors suggested their important role in tumorigenesis. To date, over 15,000 lncRNA molecules characterized by diversity of function and specificity of cell distribution were identified in the human genome. Due to their involvement in proliferation, apoptosis, metabolism, and differentiation, they have a key role in the biological processes and pathogenesis of many diseases, including MM. This review summarizes current knowledge of non-coding RNAs (ncRNA), especially lncRNAs, and their role in MM pathogenesis. Undeniable involvement of lncRNAs in MM development suggests their potential as biomarkers.

scholarly journals LncSEA: a platform for long non-coding RNA related sets and enrichment analysis

2020 ◽  
Vol 49 (D1) ◽  
pp. D969-D980 ◽  
Author(s):  
Jiaxin Chen ◽  
Jian Zhang ◽  
Yu Gao ◽  
Yanyu Li ◽  
Chenchen Feng ◽  
...  
Keyword(s):  
Enrichment Analysis ◽  
Regulatory Elements ◽  
Biological Functions ◽  
Comprehensive Collection ◽  
Non Coding Rna ◽  
Regulatory Data ◽  
Non Coding Rnas ◽  
Dna Regulatory Elements ◽  
Long Non Coding Rna ◽  
Upstream Regulators

Abstract Long non-coding RNAs (lncRNAs) have been proven to play important roles in transcriptional processes and various biological functions. Establishing a comprehensive collection of human lncRNA sets is urgent work at present. Using reference lncRNA sets, enrichment analyses will be useful for analyzing lncRNA lists of interest submitted by users. Therefore, we developed a human lncRNA sets database, called LncSEA, which aimed to document a large number of available resources for human lncRNA sets and provide annotation and enrichment analyses for lncRNAs. LncSEA supports >40 000 lncRNA reference sets across 18 categories and 66 sub-categories, and covers over 50 000 lncRNAs. We not only collected lncRNA sets based on downstream regulatory data sources, but also identified a large number of lncRNA sets regulated by upstream transcription factors (TFs) and DNA regulatory elements by integrating TF ChIP-seq, DNase-seq, ATAC-seq and H3K27ac ChIP-seq data. Importantly, LncSEA provides annotation and enrichment analyses of lncRNA sets associated with upstream regulators and downstream targets. In summary, LncSEA is a powerful platform that provides a variety of types of lncRNA sets for users, and supports lncRNA annotations and enrichment analyses. The LncSEA database is freely accessible at http://bio.liclab.net/LncSEA/index.php.

scholarly journals Genome-wide screening of circadian and non-circadian impact of Neat1 genetic deletion

2021 ◽  
Author(s):  
Audrey Jacq ◽  
Denis Becquet ◽  
Maria-Montserrat Bello-Goutierrez ◽  
Bénédicte Boyer ◽  
Séverine Guillen ◽  
...  
Keyword(s):  
Biological Processes ◽  
Circadian Expression ◽  
Non Coding Rna ◽  
Genome Wide ◽  
Independent Functions ◽  
Rat Pituitary ◽  
Genetic Deletion ◽  
The Impact ◽  
Long Non Coding Rna ◽  
Abundant Transcript

AbstractThe functions of the long non-coding RNA, Nuclear enriched abundant transcript 1 (Neat1), are poorly understood. Neat1 is required for the formation of paraspeckles, but its respective paraspeckle-dependent or independent functions are unknown. Several studies including ours reported that Neat1 is involved in the regulation of circadian rhythms. We characterized the impact of Neat1 genetic deletion in a rat pituitary cell line. The mRNAs whose circadian expression pattern or expression level is regulated by Neat1 were identified after high-throughput RNA sequencing of the circadian transcriptome of wild-type cells compared to cells in which Neat1 was deleted by CRISPR/Cas9. The numerous RNAs affected by Neat1 deletion were found to be circadian or non-circadian, targets or non-targets of paraspeckles, and to be associated with many key biological processes showing that Neat1, interacting or independently of the circadian system, could play crucial roles in key physiological functions through diverse mechanisms.

Sign in / Sign up

Export Citation Format

Share Document