Machine Learning Identifies FLNA as a Key Molecular Target Regulating Neuronal Apoptosis after Spinal Cord Injury.

Spinal cord injury (SCI), a traumatic type of central nervous system injury, is closely associated with neuronal apoptosis. However, the specific biomarkers and regulatory mechanisms of neuronal apoptosis in SCI patients remain unclear. In this study, we aimed to identify differentially expressed proteins (DEPs) that regulate neuronal apoptosis after SCI and reveal potential diagnostic and therapeutic targets. Spinal cord tissues were collected for LC‒MS/MS analysis at five different time points after injury. Enrichment analysis, WGCNA, random forest, support vector machine recursive feature elimination, and receiver operating characteristic (ROC) curve analysis methods were used to identify proteins and pathways associated with neuronal apoptosis. Validation was performed using a rat model and PC12 cells. A total of 351 DEPs were identified. By integrating DEPs, WGCNA, and machine learning methods, filamin A (FLNA), an apoptosis-related protein, was identified. The reliability of this finding was confirmed in the above three datasets. Spearman correlation analysis was performed to identify the top 100 proteins whose expression correlated with that of FLNA, which were then subjected to enrichment analysis. GO enrichment analysis and KEGG enrichment analysis revealed that expression of these proteins was enriched in mitochondrial oxidative phosphorylation. Western blot and qRT‒PCR analyses confirmed the upregulation of FLNA expression in a rat model of SCI. In vitro experiments revealed that silencing FLNA expression using siRNA reduced H