Bioinformatics and machine learning-based identification of cell cycle-related genes and molecular subtypes in endometrial cancer

Endometrial cancer is a common malignant tumor in women, with rising incidence rates and an unoptimistic prognosis. DSN1 is a kinetochore protein-coding gene that affects centromere assembly and progression in cell cycles, which is associated with adverse predictions for many cancers. However, the role of DSN1 in UCEC has not yet been reported. We identified the UCEC-related gene module and obtained the differential genes. Then we constructed a diagnostic model and identified the subtype of the molecule and its association with predictions. Subsequently, we identified DSN1 as the core gene and predicted its predictive value. Furthermore, using bioinformatics methods, we found DSN1 was associated with certain clinical characteristics and experimentally validated the expression in cancer tissues of DSN1. Pathway enrichment analysis identified DSN1 as a cell cycle-associated protein, which was validated by WB. The protein interaction network also revealed DSN1 was significantly associated with NDC80. Then we explored the correlation of DSN1 and immune cells and immune cell infiltration and found that DSN1 may affect Th2 enrichment by affecting CCL7 and CCL8. Drug susceptibility analysis showed DSN1 was sensitive to cisplatin and resistant to sunitinib. In conclusion, DSN1 was a novel biomarker that contributes to prognosis and treatment.