Analysis of PANoptosis‑related genes in septic cardiomyopathy by bioinformatics, machine learning and experimental validation

The mechanisms underlying the pathogenesis of septic cardiomyopathy (SCM) are intricate and incompletely understood. PANoptosis is a novel type of programmed cell death, and in the present study, bioinformatics, machine learning and experimental validation were used to identify key PANoptosis‑related genes (PRGs) associated with SCM. Differentially expressed genes were obtained through analysis of the Gene Expression Omnibus dataset, and these genes were intersected with the PRGs to obtain the differentially expressed PRGs. Three machine learning algorithms were used to screen key PRGs; CIBERSORT was used for immune infiltration analysis and the diagnostic value of key PRGs was evaluated by plotting receiver operating characteristic curves. Additionally, a competitive endogenous (ce)RNA regulatory network analysis was conducted, and drug prediction analysis was performed. Finally, the expression of key PRGs was verified via quantitative PCR. A total of 157 differentially expressed genes and 21 differentially expressed PRGs were screened. In addition, two key PRGs (RIPK2 and GADD45B) were screened using least absolute shrinkage and selection operator regression, the support vector machine‑recursive feature elimination algorithm and the random forest algorithm, with both genes demonstrating a high diagnostic value. RIPK2 and GADD45B were positively correlated with neutrophils. The ceRNA regulatory network included two mRNAs, eight microRNAs and 16 long noncoding RNAs and 10 drugs/molecular compounds were predicted. Finally, quantitative PCR results revealed that the expression of both RIPK2 and GADD45B was upregulated in the lipopolysaccharide‑induced HL‑1 cell injury model. In conclusion, the present study identified two key PRGs (RIPK2 and GADD45B) associated with SCM; these findings may lead to the development of novel diagnostic and therapeutic approaches for SCM.