Prediction of Prognosis in Patients with Sepsis Based on Platelet-Related Genes

The study aimed to develop a risk prognostic model using platelet-related genes (PRGs) to predict sepsis patient outcomes. Sepsis patient data from the Gene Expression Omnibus (GEO) database and PRGs from the Molecular Signatures Database (MSigDB) were analyzed. Differential analysis identified 1139 differentially expressed genes (DEGs) between sepsis and control groups. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses revealed enrichment in functions related to immune cell regulation and pathways associated with immune response and infectious diseases. A risk prognostic model was established using LASSO and Cox regression analyses, incorporating 10 PRGs selected based on their association with sepsis prognosis. The model demonstrated good stratification and prognostic effects, confirmed by survival and receiver operating characteristic (ROC) curve analyses. It served as an independent prognostic factor in sepsis patients. Further analysis using the CIBERSORT algorithm showed higher infiltration of activated natural killer (NK) cells and lower infiltration of CD8 T cells and CD4 T cells naïve in the high-risk group compared to the low-risk group. Additionally, expression levels of human leukocyte antigen (HLA) genes were significantly lower in the high-risk group. In conclusion, the 10-gene risk model based on PRGs accurately predicted sepsis patient prognosis and immune infiltration levels. This study provides valuable insights into the role of platelets in sepsis prognosis and diagnosis, offering potential implications for personalized treatment strategies.