Sepsis and sepsis-associated organ failure are devastating conditions for which there are no effective therapeutic agent. Several studies have demonstrated the significance of ferroptosis in sepsis. The study aimed to identify ferroptosis-related genes (FRGs) in sepsis, providing potential therapeutic targets.The weighted gene co-expression network analysis (WGCNA) was utilized to screen sepsis-associated genes. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses were used to explore gene functions. Three machine learning methods were employed to identify sepsis-related hub genes. Survival and multivariate Cox regression analysis allowed further screening for the key gene RRM2 associated with prognosis. The immune infiltration analysis of the screened sepsis key genes was performed. Additionally, a cecum ligation and puncture (CLP)-induced mouse sepsis model was constructed to validate the expression of key gene in the sepsis.Six sepsis-associated differentially expressed FRGs (RRM2, RPL7A, HNRNPA1, PEBP1, MYL8B and TXNIP) were screened by WGCNA and three machine learning methods analysis. Survival analysis and multivariate Cox regression analysis showed that RRM2 was a key gene in sepsis and an independent prognostic factor associated with clinicopathological and molecular features of sepsis. Immune cell infiltration analysis demonstrated that RRM2 had a connection to various immune cells, such as CD4 T cells and neutrophils. Furthermore, animal experiment demonstrated that RRM2 was highly expressed in CLP-induced septic mice, and the use of Fer-1 significantly inhibited RRM2 expression, inhibited serum inflammatory factor TNF-α, IL-6 and IL-1β expression, ameliorated intestinal injury and improved survival in septic mice.RRM2 plays an important role in sepsis and may contribute to sepsis through the ferroptosis pathway. This study provides potential therapeutic targets for sepsis.© 2024. The Author(s), under exclusive licence to Springer Nature Switzerland AG.