Machine Learning-Based Integrative Analysis Identifies SUMOylation-Related Genes Underlying the Immune Heterogeneity of Sepsis.

ABSTRACT: Sepsis heterogeneity poses a challenge to accurate diagnosis and treatment. The impact of SUMOylation, a post-translational modification, on sepsis is largely unexplored. We integrated three GEO datasets to construct a large-scale sepsis cohort and applied three machine learning algorithms to screen hub genes from differentially expressed genes (DEGs) associated with SUMOylation in sepsis. Unsupervised consensus clustering was performed to identify sepsis subtypes. Using single-sample gene set enrichment analysis (ssGSEA) and gene set variation analysis (GSVA), we analysed the immunological and functional features of these subtypes. We assembled the regulatory network of hub genes and performed drug prediction analysis. The expression of hub genes was confirmed in a murine caecal ligation

SUBMITTER: Li Z 

PROVIDER: S-EPMC12858257 | biostudies-literature | 2026 Jan-Dec

REPOSITORIES: biostudies-literature