Integrating Machine Learning and Multi-omics to Identify Key SUMOylation Molecular Signature in Sarcoma

Background: Sarcoma (SARC) is a rare and heterogeneous cancer originating from mesenchymal tissue. Due to its complex molecular mechanisms and limited treatment options, patients often have poor prognoses. Protein SUMOylation is an important post-translational modification process that plays a key role in regulating cellular functions and is closely related to the onset and progression of various cancers. However, the specific mechanisms by which SUMOylation affects SARC progression are not fully understood. Methods: In this study, comprehensive bioinformatics approaches were utilized to analyze multiple datasets of SARC samples. By screening and identifying SUMOylation-related genes, we further explored the expression patterns of these genes in SARC and their association with prognosis and then constructed a consensus prognostic model. In particular, we focused on the KIAA1586 gene, which has attracted increasing attention in cancer biology, and conducted an in-depth study of its role in SARC. Results: The study revealed that 19 SUMOylation-related genes were significantly correlated with the prognosis of SARC. Subsequently, the consensus prognostic model constructed by ridge regression could accurately predict the survival of patients in multiple data sets. Afterward, we identified KIAA1586 as the key gene, and its expression level was closely related to the prognosis of patients. GSEA enrichment analysis demonstrated that KIAA1586 might affect the progression of SARC by regulating the cell cycle and immune-related pathways, providing new insights into the molecular mechanism of SARC. Conclusion: We have constructed a SUMOylation signature model that can accurately predict the prognosis of SARC patients, and identified KIAA1586 as a key SUMOylation gene that plays a crucial role in the onset and development of tumors by participating in cell cycle regulation and immune suppression.