Background: Bladder cancer has become the tenth most diagnosed cancer worldwide. The prognosis has been shown to differ between non-muscle invasive bladder cancer (NMIBC) and muscle invasive bladder cancer (MIBC). We aimed to identify signature genes that are associated with the invasiveness and survival of bladder cancer and to identify potential treatments.Methods: We downloaded gene expression profiles of bladder cancer from the Gene Expression Omnibus database to identify differentially expressed genes and perform weighted gene co-expression network analysis. Functional enrichment was analyzed by GO and KEGG analyses. Hub genes were identified from the significant module. Another dataset was also acquired to verify the expression of hub genes. Univariate and multivariate Cox regression analyses were applied to the dataset downloaded from The Cancer Genome Atlas database. Risk scores were calculated and the effect was evaluated by Kaplan-Meier survival analysis. A nomogram was constructed and validated using training and testing samples, respectively. Analysis of the tumor immune microenvironment was conducted with the CIBERSORT algorithm.Results: In total, 1,245 differentially expressed genes (DEGs) were identified. A distinct module was identified that was significantly correlated to invasiveness. The genes within this module were found to be significantly associated with extracellular exosomes, GTPase activity, metabolic pathways, etc. Three hub genes (VSIG2, PPFIBP2, and DENND2D) were identified as biomarkers of invasiveness; two of these (PPFIBP2 and DENND2D) were closely associated with prognosis. The risk score was regarded as an independent prognostic factor. The nomogram was associated with acceptable accuracy for predicting 1- and 5-year overall survival. The infiltrating levels of resting NK cells, activated natural killer (NK) cells, CD8+ T cells, activated memory CD4+ T cells, and T follicular helper cells, were significantly higher in the group with lower risk scores. The group with higher risk scores showed predominant infiltration by regulatory T cells (Tregs).Conclusion: We successfully identified three signature genes related to invasiveness and constructed a nomogram of bladder cancer with acceptable performance. Differences suggested by risk scores between groups of patients showing diverse patterns of immune cell infiltration may be beneficial for selecting therapeutic approaches and predicting prognosis.
Traditional classification algorithms perform not very well on imbalanced data sets and small sample size. To deal with the problem, a novel method is proposed to change the class distribution through adding virtual samples, which are generated by the windowed regression over-sampling (WRO) method. The proposed method WRO not only reflects the additive effects but also reflects the multiplicative effect between samples. A comparative study between the proposed method and other over-sampling methods such as synthetic minority over-sampling technique (SMOTE) and borderline over-sampling (BOS) on UCI datasets and Fourier transform infrared spectroscopy (FTIR) data set is provided. Experimental results show that the WRO method can achieve better performance than other methods.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.