ObjectiveTo explore the expression of genes related to hepatocellular carcinoma (HCC) stem cells and their prognostic correlation by using weighted gene co-expression network analysis (WGCNA).MethodsFirstly, the transcriptome sequencing (RNA-seq) and clinical data of HCC were downloaded from the public database the Cancer Genome Atlas (TCGA), and the mRNA expression-based stiffness index (mRNAsi) table of cancer stem cells was downloaded and sorted out to analyze the relationship between mRNAsi and pathological grade and prognosis of HCC. The mRNAsi of HCC was downloaded and the prognostic value of mRNAsi was discussed. Then we used WGCNA to screen the key modules related to liver cancer stem cells (LSCS). Gene Ontology and the Kyoto Encyclopedia of Genes and Genomes were used for the functional and pathway enrichment analysis. The online database STRING was used to construct hub genes coding proteins interaction (PPI) network and screen key genes. Finally, the key genes were analyzed for expression differences and expression correlations. The online database Kaplan-Meier plotter was used for survival analysis and verified.ResultsmRNAsi was significantly upregulated in cancer tissues (P<0.001), and increased with the increase of pathological grade of HCC (P=0.001). The mortality rate of the higher mRNAsi group was higher than that of the lower mRNAsi group (P=0.006). GO analysis found that hub genes were mainly involved in biological processes, such as mitosis and DNA replication, and KEGG showed that hub genes were enriched in cell cycle, DNA mismatch repair, oocyte meiosis, and other signaling pathways. We screened 10 key genes (included CCNB1, CDC20, CDCA8, NDC80, KIF20A, TTK, CDC45, KIF15, MCM2, and NCAPG) related to mRNAsi of HCC based on WGCNA. The key genes were highly expressed in the tumor samples compared to the normal samples. In addition, there was a strong interaction between proteins of these key genes (P<0.05), a strong co-expression relationship at the transcriptional level, and all related to prognosis of HCC.ConclusionsmRNAsi plays an important role in the occurrence and development of HCC. Ten key genes related to LSCS were screened, which may act as therapeutic targets for inhibiting the stem cell characteristics of HCC.
ObjectiveTo find the hub genes related to bone metastasis of breast cancer by weighted geneco-expression network analysis (WGCNA) method, and provide theoretical support for the development of new targeted therapeutic drugs.MethodsThe basic clinical features of 286 breast cancer patients and the gene expression information of tumor specimens were downloaded from the GSE2034 dataset from the Gene Expression Omnibus. R software was used to analyze the gene microarray. The WGCNA package embedded in the R software was used for various analysis in weighted correlation network analysis. Cox proportional hazard regression was performed by using SPSS software.ResultsThe top one quarter genes with the greatest variance variability were selected by WGCNA, and a total of 5 000 genes were used for further enrichment analysis. Finally, 15 gene co-expression modules were constructed, and the magenta module (r=0.94, P<0.001) was significantly positively correlated with bone metastasis of breast cancer. It was further found that six hub genes highly associated with bone metastasis in the magenta module were: Ral GTPase-activating protein subunitalpha-1 (RALGAPA1), B-cell antigen receptor complex-associated protein alpha chain (CD79A), immunoglobulin kappa chain C region (IGKC), arrestin beta 2 (ARRB2), differentially expressed in FDCP 6 homolog (DEF6), and immunoglobulin lambda variable 2 (IGLV2).ConclusionWe found that RALGAPA1, CD79A, IGKC, ARRB2. DEF6, and IGLV2 may play an important role in bone metastasis of breast cancer.
Objective To explore the potential mechanism of the occurrence and development of lupus nephritis (LN) and identify key biomarkers and immune-related pathways associated with the progression of LN. Methods We downloaded a dataset from the Gene Expression Omnibus database. By analyzing the differential expression of genes and performing weighted gene co-expression network analysis (WGCNA), as well as Gene Ontology enrichment, Disease Ontology enrichment, and Kyoto Encyclopedia of Genes and Genomes pathway enrichment, we explored the biological functions of differentially expressed genes in LN. Using three machine learning models, namely LASSO regression, support vector machine, and random forest, we identified the hub genes in LN, and constructed a line diagram diagnosis model based on the hub genes. The diagnostic accuracies of the hub genes were evaluated using the receiver operating characteristic curve, and the relationship between known marker gene sets and hub gene expression was analyzed using single sample gene set enrichment analysis. Results We identified a total of 2297 differentially expressed genes. WGCNA generated 7 co-expression modules, among which the cyan module had the highest correlation with LN. We obtained 347 target genes by combining differential genes. Using the three machine learning methods, LASSO regression, support vector machine, and random forest, we identified three hub genes (CLC, ADGRE4P, and CISD2) that could serve as potential biomarkers for LN. The area under the receiver operating characteristic curve (AUC) analysis showed that these three hub genes had significant diagnostic value (AUCCLC=0.718, AUCADGRE4P=0.813, AUCCISD2=0.718). According to single sample gene set enrichment analysis, the hub genes were mainly associated with apoptosis, glycolysis, metabolism, hypoxia, and tumor necrosis factor-α-nuclear factor-κB-related pathways. Conclusions By combining WGCNA and machine learning techniques, three hub genes (CLC, ADGRE4P, and CISD2) that may be involved in the occurrence and development of LN are identified. These genes have the potential to aid in the early clinical diagnosis of LN and provide insight into the mechanisms underlying LN progression.