Identifying Hub Genes for Glaucoma Based on Bulk RNA Sequencing Data and Multi-machine Learning Models

AIMS

The aims of this study were to determine hub genes in glaucoma through multiple machine learning algorithms.

BACKGROUND

Glaucoma has afflicted many patients for many years, with excessive pressure in the eye continuously damaging the nervous system and leading to severe blindness. An effective molecular diagnostic method is currently lacking.

OBJECTIVE

The present study attempted to reveal the molecular mechanism and gene regulatory network of hub genes in glaucoma, followed by an attempt to reveal the drug-gene-disease network regulated by hub genes.

METHODS

A microarray sequencing dataset (GSE9944) was obtained through the Gene Expression Omnibus database. The differentially expressed genes in Glaucoma were identified. Based on these genes, we constructed three machine learning models for feature training, Random Forest model (RF), Least absolute shrinkage and selection operator regression model (LASSO), and Support Vector Machines model (SVM). Meanwhile, Weighted Gene Co-Expression Network Analysis (WGCNA) was performed for GSE9944 expression profiles to identify Glaucoma-related genes. The overlapping genes in the four groups were considered as hub genes of Glaucoma. Based on these genes, we also constructed a molecular diagnostic model of Glaucoma. In this study, we also performed molecular docking analysis to explore the gene-drug network targeting hub genes. In addition, we evaluated the immune cell infiltration landscape in Glaucoma samples and normal samples by applying CIBERSORT method.

RESULTS

8 hub genes were determined: ATP6V0D1, PLEC, SLC25A1, HRSP12, PKN1, RHOD, TMEM158 and GSN. The diagnostic model showed excellent diagnostic performance (area under the curve=1). GSN might positively regulate T cell CD4 naïve as well as negatively regulate T cell regulation (Tregs). In addition, we constructed gene-drug networks in an attempt to explore novel therapeutic agents for Glaucoma.

CONCLUSION

Our results systematically determined 8 hub genes and established a molecular diagnostic model that allowed the diagnosis of Glaucoma. Our study provided a basis for future systematic studies of Glaucoma pathogenesis.

Identification of hub biomarkers and immune-related pathways participating in the progression of Kawasaki disease by integrated bioinformatics analysis

BACKGROUND

Kawasaki disease (KD) is a systemic vasculitis that commonly affects children and its etiology remains unknown. Growing evidence suggests that immune-mediated inflammation and immune cells in the peripheral blood play crucial roles in the pathophysiology of KD. The objective of this research was to find important biomarkers and immune-related mechanisms implicated in KD, along with their correlation with immune cells in the peripheral blood.

MATERIAL/METHODS

Gene microarray data from the Gene Expression Omnibus (GEO) was utilized in this study. Three datasets, namely GSE63881 (341 samples), GSE73463 (233 samples), and GSE73461 (279 samples), were obtained. To find intersecting genes, we employed differentially expressed genes (DEGs) analysis and weighted gene co-expression network analysis (WGCNA). Subsequently, functional annotation, construction of protein-protein interaction (PPI) networks, and Least Absolute Shrinkage and Selection Operator (LASSO) regression were performed to identify hub genes. The accuracy of these hub genes in identifying KD was evaluated using the receiver operating characteristic curve (ROC). Furthermore, Gene Set Variation Analysis (GSVA) was employed to explore the composition of circulating immune cells within the assessed datasets and their relationship with the hub gene markers.

RESULTS

WGCNA yielded eight co-expression modules, with one hub module (MEblue module) exhibiting the strongest association with acute KD. 425 distinct genes were identified. Integrating WGCNA and DEGs yielded a total of 277 intersecting genes. By conducting LASSO analysis, five hub genes (S100A12, MMP9, TLR2, NLRC4 and ARG1) were identified as potential biomarkers for KD. The diagnostic value of these five hub genes was demonstrated through ROC curve analysis, indicating their high accuracy in diagnosing KD. Analysis of the circulating immune cell composition within the assessed datasets revealed a significant association between KD and various immune cell types, including activated dendritic cells, neutrophils, immature dendritic cells, macrophages, and activated CD8 T cells. Importantly, all five hub genes exhibited strong correlations with immune cells.

CONCLUSION

Activated dendritic cells, neutrophils, and macrophages were closely associated with the pathogenesis of KD. Furthermore, the hub genes (S100A12, MMP9, TLR2, NLRC4, and ARG1) are likely to participate in the pathogenic mechanisms of KD through immune-related signaling pathways.

The driving factors of corporate carbon emissions: an application of the LASSO model with survey data

Corporate carbon performance is a key driver of achieving corporate sustainability. The identification of factors that influence corporate carbon emissions is fundamental to promoting carbon performance. Based on the carbon disclosure project (CDP) database, we integrate the least absolute shrinkage and selection operator (LASSO) regression model and the fixed effects model to identify the determinants of carbon emissions. Furthermore, we rank determining factors according to their importance. We find that Capx enters the models under all carbon contexts. For Scope 1 and Scope 2, financial-level factors play a greater role. For Scope 3, corporate internal incentive policies and emission reduction behaviors are important. Different from absolute carbon emissions, for relative carbon emissions, the financial-level factors' debt-paying ability is a vital reference indicator for the impact of corporate carbon emissions.

Bioinformatic analysis of underlying mechanisms of Kawasaki disease via Weighted Gene Correlation Network Analysis (WGCNA) and the Least Absolute Shrinkage and Selection Operator method (LASSO) regression model

BACKGROUND

Kawasaki disease (KD) is a febrile systemic vasculitis involvingchildren younger than five years old. However, the specific biomarkers and precise mechanisms of this disease are not fully understood, which can delay the best treatment time, hence, this study aimed to detect the potential biomarkers and pathophysiological process of KD through bioinformatic analysis.

METHODS

The Gene Expression Omnibus database (GEO) was the source of the RNA sequencing data from KD patients. Differential expressed genes (DEGs) were screened between KD patients and healthy controls (HCs) with the "limma" R package. Weighted gene correlation network analysis (WGCNA) was performed to discover the most corresponding module and hub genes of KD. The node genes were obtained by the combination of the least absolute shrinkage and selection operator (LASSO) regression model with the top 5 genes from five algorithms in CytoHubba, which were further validated with the receiver operating characteristic curve (ROC curve). CIBERSORTx was employed to discover the constitution of immune cells in KDs and HCs. Functional enrichment analysis was performed to understand the biological implications of the modular genes. Finally, competing endogenous RNAs (ceRNA) networks of node genes were predicted using online databases.

RESULTS

A total of 267 DEGs were analyzed between 153 KD patients and 92 HCs in the training set, spanning two modules according to WGCNA. The turquoise module was identified as the hub module, which was mainly enriched in cell activation involved in immune response, myeloid leukocyte activation, myeloid leukocyte mediated immunity, secretion and leukocyte mediated immunity biological processes; included type II diabetes mellitus, nicotinate and nicotinamide metabolism, O-glycan biosynthesis, glycerolipid and glutathione metabolism pathways. The node genes included ADM, ALPL, HK3, MMP9 and S100A12, and there was good performance in the validation studies. Immune cell infiltration analysis revealed that gamma delta T cells, monocytes, M0 macrophage, activated dendritic cells, activated mast cells and neutrophils were elevated in KD patients. Regarding the ceRNA networks, three intact networks were constructed: NEAT1/NORAD/XIST-hsa-miR-524-5p-ADM, NEAT1/NORAD/XIST-hsa-miR-204-5p-ALPL, NEAT1/NORAD/XIST-hsa-miR-524-5p/hsa-miR-204-5p-MMP9.

CONCLUSION

To conclude, the five-gene signature and three ceRNA networks constructed in our study are of great value in the early diagnosis of KD and might help to elucidate our understanding of KD at the RNA regulatory level.

Robust analysis of novel mRNA-lncRNA cross talk based on ceRNA hypothesis uncovers carcinogenic mechanism and promotes diagnostic accuracy in esophageal cancer

Background

ceRNAs have emerged as pivotal players in the regulation of gene expression and play a crucial role in the physiology and development of various cancers. Nevertheless, the function and underlying mechanisms of ceRNAs in esophageal cancer (EC) are still largely unknown.

Methods

In this study, profiles of DEmRNAs, DElncRNAs, and DEmiRNAs between normal and EC tumor tissue samples were obtained from the Cancer Genome Atlas database using the DESeq package in R by setting the adjusted P<0.05 and |log₂(fold change)|>2 as the cutoff. The ceRNA network (ceRNet) was initially constructed to reveal the interaction of these ceRNAs during carcinogenesis based on the bioinformatics of miRcode, miRDB, miRTarBase, and TargetScan. Then, independent microarray data of GSE6188, GSE89102, and GSE92396 and correlation analysis were used to validate molecular biomarkers in the initial ceRNet. Finally, a least absolute shrinkage and selection operator logistic regression model was built using an oncogenic ceRNet to diagnose EC more accurately.

Results

We successfully constructed an oncogenic ceRNet of EC, crosstalk of hsa-miR372-centered CADM2-ADAMTS9-AS2 and hsa-miR145-centered SERPINE1-PVT1. In addition, the risk-score model -0.0053*log₂(CADM2)+0.0168*log₂(SERPINE1)-0.0073*log₂(ADAMTS9-AS2)+0.0905*log₂(PVT1)+0.0047*log₂(hsa-miR372)-0.0193*log₂(hsa-miR145), (log₂[gene count]) could improve diagnosis of EC with an AUC of 0.988.

Conclusion

We identified two novel pairs of ceRNAs in EC and its role of diagnosis. The pairs of hsa-miR372-centered CADM2-ADAMTS9-AS2 and hsa-miR145-centered SERPINE1-PVT1 were likely potential carcinogenic mechanisms of EC, and their joint detection could improve diagnostic accuracy.