In order to improve the result of clinical practice for undergraduate students of sto-matology, the standardized training program was carried out before clinical practice by the teaching and researching Office of Periodontology of Affiliated Stomatological Hospital of Chongqing Medical University, in this program, based on our own realities, the structure and content of teaching was designed to guide the transformation of students to doctors, collective teaching and personalized teach-ing was combined to reform the conventional model of teaching and various student-oriented teaching methods were explored, including mimic forum of disciplines, intractable case discussion, clinical practice summary, course completion test and so on, thus promoting the result of clinical practice. Be-sides, the construction of teaching staff was strengthened, and the evaluation and management system was established to ensure the teaching quality, thus summarizing a set of efficient and effective clinical teaching methods presented in this paper.

Objective:To explore genes related to costimulatory molecule related to the prognosis of bladder cancer,and to construct and evaluate prognosis model based on costimulatory molecule-based signature(CMS).Methods:Gene expression matrix and clinical information of bladder cancer patients were downloaded from TCGA database and GEO database(GSE31684),and costimulatory molecule-related genes were retrieved from the literature.The univariate and multivariate Cox analysis were used to screened prognostic-related genes and constructed prognostic model.Forecast accuracy of model was verified in TCGA training group,TCGA validation data group and GEO group by Kaplan-Meier survival analysis and receiver operating characteristic curve(ROC).Considering risk score and clinical characteristics,we constructed a nomogram and evaluated its performance by consistency analysis and ROC.CIBERSORT algorithm was used to analyze immune cell composition of tumor microenvironment infiltration,and gene set enrichment analysis(GSEA)was performed to explore the potential mechanism.Results:Four prognostic-related CMSs were found:TNFRSF14,CD276,ICOS and TMIGD2,of which three were included in the risk score construction.Multivariate Cox regression results showed that the risk score based on CMS was an independent prognostic factor for bladder cancer patients.Consistency analysis and ROC results showed that the nomogram had ideal prognosis prediction accuracy.Immune infiltration analysis showed that the high risk group was likely to be in immunosuppressive state.GSEA results suggested that genes in high risk group were enriched in extracel-lular matrix(ECM)receptors interaction,cell cycle and other pathways.Conclusion:TNFRSF14,CD276 and ICOS may be potential prognostic biomarkers for bladder cancer patients.CMS-based risk score and nomogram could contribute to early prognosis and choice of personalized treatment.