Exploration on Medication Patterns and Mechanisms of National Patented Traditional Chinese Medicine Compound Prescriptions for Ischemic Stroke Based on Data Mining and In Silico Knockdown

• • Nine active compounds were identified from the core herb combination, with kaempferol-PTGS2 and kaempferol-EGFR complexes exhibiting stable binding in molecular dynamics simulations (RMSD < 0.3 nm over 100 ns), suggesting potential for targeting inflammatory pathways in ischemic stroke; this provides a prioritized list for subsequent experimental validation, reducing screening costs by approximately 40% compared to random screening. • • Spatial transcriptomics revealed that candidate target module scores were elevated in the infarct core and penumbra regions, with RCTD deconvolution showing significant spatial concordance (p < 0.01) between high-expression areas and macrophage/astrocyte distribution, indicating that PTGS2, IL-6, and EGFR are spatially associated with the inflammatory microenvironment; this spatial validation enhances target confidence for drug development. • • CellOracle virtual knockdown identified Jun and Fos as the most prominent transcription factors whose perturbation caused substantial cell state shifts (effect size > 0.5), implicating the AP-1 complex in post-ischemic inflammatory responses and repair; this computational prediction offers a mechanistic hypothesis for targeting transcription factor networks, potentially accelerating lead optimization. • • The study acknowledges critical limitations: single-cell data covered only a single time point (24 h post-MCAO), molecular dynamics simulations lacked independent replicates, and no in vivo/in vitro efficacy experiments were performed; these gaps necessitate further validation, as the core herb combination has not been tested as a clinical formula, and database coverage biases may introduce prediction errors of up to 30%.