Multiple datasets to explore the molecular mechanism of sepsis

BACKGROUND: This study aimed to identify potential biomarkers, by means of bioinformatics, affecting the occurrence and development of septic shock. METHODS: Download GSE131761 septic shock data set from NCBI geo database, including 33 control samples and 81 septic shock samples. GSE131761 and sequencing data were used to identify and analyze differentially expressed genes in septic shock patients and normal subjects. In addition, with sequencing data as training set and GSE131761 as validation set, a diagnostic model was established by lasso regression to identify key genes. ROC curve verified the stability of the model. Finally, immune infiltration analysis, enrichment analysis, transcriptional regulation analysis and correlation analysis of key genes were carried out to understand the potential molecular mechanism of key genes affecting septic shock. RESULTS: A total of 292 differential genes were screened out from the self-test data, 294 differential genes were screened out by GSE131761, Lasso regression was performed on the intersection genes of the two, a diagnostic model was constructed, and 5 genes were identified as biomarkers of septic shock. These 5 genes were SIGLEC10, VSTM1, GYPB, OPTN, and GIMAP7. The five key genes were strongly correlated with immune cells, and the ROC results showed that the five genes had good predictive performance on the occurrence and development of diseases. In addition, the key genes were strongly correlated with immune regulatory genes. CONCLUSION: In this study, a series of algorithms were used to identify five key genes that are associated with septic shock, which may become potential candidate targets for septic shock diagnosis and treatment. TRIAL REGISTRATION: Approval number:2019XE0149-1.