Recent advances in machine learning methods for predicting LncRNA and disease associations

LncRNAs act as modulators of macrophages within the tumor microenvironment

Long non-coding RNAs (lncRNAs) have been established as pivotal players in various cellular processes, encompassing the regulation of transcription, translation and post-translational modulation of proteins, thereby influencing cellular functions. Notably, lncRNAs exert a regulatory influence on diverse biological processes, particularly in the context of tumor development. Tumor-associated macrophages (TAMs) exhibit the M2 phenotype, exerting significant impact on crucial processes such as tumor initiation, angiogenesis, metastasis and immune evasion. Elevated infiltration of TAMs into the tumor microenvironment (TME) is closely associated with a poor prognosis in various cancers. LncRNAs within TAMs play a direct role in regulating cellular processes. Functioning as integral components of tumor-derived exosomes, lncRNAs prompt the M2-like polarization of macrophages. Concurrently, reports indicate that lncRNAs in tumor cells contribute to the expression and release of molecules that modulate TAMs within the TME. These actions of lncRNAs induce the recruitment, infiltration and M2 polarization of TAMs, thereby providing critical support for tumor development. In this review, we survey recent studies elucidating the impact of lncRNAs on macrophage recruitment, polarization and function across different types of cancers.

Identification and coregulation pattern analysis of long noncoding RNAs in the mouse brain after Angiostrongylus cantonensis infection

BACKGROUND

Angiostrongyliasis is a highly dangerous infectious disease. Angiostrongylus cantonensis larvae migrate to the mouse brain and cause symptoms, such as brain swelling and bleeding. Noncoding RNAs (ncRNAs) are novel targets for the control of parasitic infections. However, the role of these molecules in A. cantonensis infection has not been fully clarified.

METHODS

In total, 32 BALB/c mice were randomly divided into four groups, and the infection groups were inoculated with 40 A. cantonensis larvae by gavage. Hematoxylin and eosin (H&E) staining and RNA library construction were performed on brain tissues from infected mice. Differential expression of long noncoding RNAs (lncRNAs) and mRNAs in brain tissues was identified by high-throughput sequencing. The pathways and functions of the differentially expressed lncRNAs were determined by Kyoto Encyclopedia of Genes and Genomes (KEGG) and Gene Ontology (GO) analyses. The functions of the differentially expressed lncRNAs were further characterized by lncRNA‒microRNA (miRNA) target interactions. The potential host lncRNAs involved in larval infection of the brain were validated by quantitative real-time polymerase chain reaction (qRT‒PCR).

RESULTS

The pathological results showed that the degree of brain tissue damage increased with the duration of infection. The transcriptome results showed that 859 lncRNAs and 1895 mRNAs were differentially expressed compared with those in the control group, and several lncRNAs were highly expressed in the middle-late stages of mouse infection. GO and KEGG pathway analyses revealed that the differentially expressed target genes were enriched mainly in immune system processes and inflammatory response, among others, and several potential regulatory networks were constructed.

CONCLUSIONS

This study revealed the expression profiles of lncRNAs in the brains of mice after infection with A. cantonensis. The lncRNAs H19, F630028O10Rik, Lockd, AI662270, AU020206, and Mexis were shown to play important roles in the infection of mice with A. cantonensis infection.

RPI-EDLCN: An Ensemble Deep Learning Framework Based on Capsule Network for ncRNA-Protein Interaction Prediction

Noncoding RNAs (ncRNAs) play crucial roles in many cellular life activities by interacting with proteins. Identification of ncRNA-protein interactions (ncRPIs) is key to understanding the function of ncRNAs. Although a number of computational methods for predicting ncRPIs have been developed, the problem of predicting ncRPIs remains challenging. It has always been the focus of ncRPIs research to select suitable feature extraction methods and develop a deep learning architecture with better recognition performance. In this work, we proposed an ensemble deep learning framework, RPI-EDLCN, based on a capsule network (CapsuleNet) to predict ncRPIs. In terms of feature input, we extracted the sequence features, secondary structure sequence features, motif information, and physicochemical properties of ncRNA/protein. The sequence and secondary structure sequence features of ncRNA/protein are encoded by the conjoint k-mer method and then input into an ensemble deep learning model based on CapsuleNet by combining the motif information and physicochemical properties. In this model, the encoding features are processed by convolution neural network (CNN), deep neural network (DNN), and stacked autoencoder (SAE). Then the advanced features obtained from the processing are input into the CapsuleNet for further feature learning. Compared with other state-of-the-art methods under 5-fold cross-validation, the performance of RPI-EDLCN is the best, and the accuracy of RPI-EDLCN on RPI1807, RPI2241, and NPInter v2.0 data sets was 93.8%, 88.2%, and 91.9%, respectively. The results of the independent test indicated that RPI-EDLCN can effectively predict potential ncRPIs in different organisms. In addition, RPI-EDLCN successfully predicted hub ncRNAs and proteins in Mus musculus ncRNA-protein networks. Overall, our model can be used as an effective tool to predict ncRPIs and provides some useful guidance for future biological studies.

GATLGEMF: A graph attention model with line graph embedding multi-complex features for ncRNA-protein interactions prediction

Non-coding RNA (ncRNA) plays an important role in many fundamental biological processes, and it may be closely associated with many complex human diseases. NcRNAs exert their functions by interacting with proteins. Therefore, identifying novel ncRNA-protein interactions (NPIs) is important for understanding the mechanism of ncRNAs role. The computational approach has the advantage of low cost and high efficiency. Machine learning and deep learning have achieved great success by making full use of sequence information and structure information. Graph neural network (GNN) is a deep learning algorithm for complex network link prediction, which can extract and discover features in graph topology data. In this study, we propose a new computational model called GATLGEMF. We used a line graph transformation strategy to obtain the most valuable feature information and input this feature information into the attention network to predict NPIs. The results on four benchmark datasets show that our method achieves superior performance. We further compare GATLGEMF with the state-of-the-art existing methods to evaluate the model performance. GATLGEMF shows the best performance with the area under curve (AUC) of 92.41% and 98.93% on RPI2241 and NPInter v2.0 datasets, respectively. In addition, a case study shows that GATLGEMF has the ability to predict new interactions based on known interactions. The source code is available at https://github.com/JianjunTan-Beijing/GATLGEMF.

The Identification and Comparative Analysis of Non-Coding RNAs in Spores and Mycelia of Penicillium expansum

Penicillium expansum is the most popular post-harvest pathogen and causes blue mold disease in pome fruit and leads to significant economic losses worldwide every year. However, the fundamental regulation mechanisms of growth in P. expansum are unclear. Recently, non-coding RNAs (ncRNAs) have attracted more attention due to critical roles in normalizing gene expression and maintaining cellular genotypes in organisms. However, the research related to ncRNAs in P. expansum have not been reported. Therefore, to provide an overview of ncRNAs on composition, distribution, expression changes, and potential targets in the growth process, a comparative transcriptomic analysis was performed on spores and mycelia of P. expansum in the present study. A total of 2595 novel mRNAs, 3362 long non-coding RNAs (lncRNAs), 10 novel microRNAs (miRNAs), 86 novel small interfering RNAs (siRNAs), and 11,238 circular RNAs (circRNAs) were predicted and quantified. Of these, 1482 novel mRNAs, 5987 known mRNAs, 2047 lncRNAs, 40 miRNAs, 38 novel siRNAs, and 9235 circRNAs were differentially expressed (DE) in response to the different development stages. Afterward, the involved functions and pathways of DE RNAs were revealed via Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) database enrichment analysis. The interaction networks between mRNAs, lncRNAs, and miRNAs were also predicted based on their correlation coefficient of expression profiles. Among them, it was found that miR168 family members may play important roles in fungal growth due to their central location in the network. These findings will contribute to a better understanding on regulation machinery at the RNA level on fungal growth and provide a theoretical basis to develop novel control strategies against P. expansum.

Long non-coding RNA-targeting therapeutics: discovery and development update

INTRODUCTION

lncRNAs are major players in regulatory networks orchestrating multiple cellular functions, such as 3D chromosomal interactions, epigenetic modifications, gene expression and others. Due to progress in the development of nucleic acid-based therapeutics, lncRNAs potentially represent easily accessible therapeutic targets.

AREAS COVERED

Currently, significant efforts are directed at studies that can tap the enormous therapeutic potential of lncRNAs. This review describes recent developments in this field, particularly focusing on clinical applications.

EXPERT OPINION

Extensive druggable target range of lncRNA combined with high specificity and accelerated development process of nucleic acid-based therapeutics open new prospects for treatment in areas of extreme unmet medical need, such as genetic diseases, aggressive cancers, protein deficiencies, and subsets of common diseases caused by known mutations. Although currently wide acceptance of lncRNA-targeting nucleic acid-based therapeutics is impeded by the need for parenteral or direct-to-CNS administration, development of less invasive techniques and orally available/BBB-penetrant nucleic acid-based therapeutics is showing early successes. Recently, mRNA-based COVID-19 vaccines have demonstrated clinical safety of all aspects of nucleic acid-based therapeutic technology, including multiple chemical modifications of nucleic acids and nanoparticle delivery. These trends position lncRNA-targeting drugs as significant players in the future of drug development, especially in the area of personalized medicine.

Identification of potential LncRNAs as papillary thyroid carcinoma biomarkers based on integrated bioinformatics analysis using TCGA and RNA sequencing data

The roles and mechanisms of long non-coding RNAs (lncRNAs) in papillary thyroid cancer (PTC) remain elusive. We obtained RNA sequencing (RNA-seq) data of surgical PTC specimens from patients with thyroid cancer (THCA; n = 20) and identified differentially expressed genes (DEGs) between cancer and cancer-adjacent tissue samples. We identified 2309 DEGs (1372 significantly upregulated and 937 significantly downregulated). We performed Gene Ontology, Kyoto Encyclopedia of Genes and Genomes, gene set enrichment, and protein-protein interaction network analyses and screened for hub lncRNAs. Using the same methods, we analyzed the RNA-seq data from THCA dataset in The Cancer Genome Atlas (TCGA) database to identify differentially expressed lncRNAs. We identified 15 key differentially expressed lncRNAs and pathways that were closely related to PTC. Subsequently, by intersecting the differentially expressed lncRNAs with hub lncRNAs, we identified LINC02407 as the key lncRNA. Assessment of the associated clinical characteristics and prognostic correlations revealed a close correlation between LINC02407 expression and N stage of patients. Furthermore, receiver operating characteristic curve analysis showed that LINC02407 could better distinguish between cancerous and cancer-adjacent tissues in THCA patients. In conclusion, our findings suggest that LINC02407 is a potential biomarker for PTC diagnosis and the prediction of lymph node metastasis.