Abstract
Today, numerous international researchers have demonstrated that N(7)-methylguanosine (m(7)G) related long non-coding RNAs (m(7)G-related lncRNAs) are closely linked to the happenings and developments of various human beings' cancers. However, the connection between m(7)G-related lncRNAs and glioma prognosis has not been investigated. We did this study to look for new potential biomarkers and construct an m(7)G-related lncRNA prognostic signature for glioma. We identified those lncRNAs associated with DEGs from glioma tissue sequences as m(7)G-related lncRNAs. First, we used Pearson's correlation analysis to identify 28 DEGs by glioma and normal brain tissue gene sequences and predicated 657 m(7)G-related lncRNAs. Then, eight lncRNAs associated with prognosis were obtained and used to construct the m(7)G risk score model by lasso and Cox regression analysis methods. Furthermore, we used Kaplan-Meier analysis, time-dependent ROC, principal component analysis, clinical variables, independent prognostic analysis, nomograms, calibration curves, and expression levels of lncRNAs to determine the model's accuracy. Importantly, we validated the model with external and internal validation methods and found it has strong predictive power. Finally, we performed functional enrichment analysis (GSEA, aaGSEA enrichment analyses) and analyzed immune checkpoints, associated pathways, and drug sensitivity based on predictors. In conclusion, we successfully constructed the formula of m(7)G-related lncRNAs with powerful predictive functions. Our study provides instructional value for analyzing glioma pathogenesis and offers potential research targets for glioma treatment and scientific research.