Exploring the molecular basis of Panax notoginseng mediated inhibition of cervical cancer through machine learning, transcriptomics analysis, network pharmacology, molecular docking, and in-silico simulation approaches
Abstract Motivation Cervical cancer remains a major global health challenge, particularly in low-resource settings where treatment efficacy is limited by drug resistance and toxicity. Although Panax notoginseng (Sanqi) has demonstrated anticancer activity, its molecular mechanisms against cervical cancer remain insufficiently understood. Results An integrated transcriptomic and systems pharmacology approach was employed to investigate the therapeutic mechanisms of Panax notoginseng in cervical cancer. Transcriptomic analysis identified 533 overlapping differentially expressed genes associated with cell cycle regulation, DNA replication, cellular senescence, and p53 signaling. Network pharmacology revealed 291 overlapping targets between P. notoginseng compounds and cervical cancer-related genes. Protein–protein interaction analysis identified tumor necrosis factor, interleukin-6, proto-oncogene, non-receptor tyrosine kinase (SRC), TOP2A, and CDC45 as key hub genes involved in inflammation, apoptosis, and tumor progression. Molecular docking demonstrated strong binding affinities of ginsenoside Re, Panaxadiol, Daucosterol, and Stigmasterol toward core targets, while molecular dynamics simulations confirmed stable protein–ligand interactions. Absorption, Distribution, Metabolism, Excretion, and Toxicity (ADMET) analysis suggested comparable pharmacokinetic properties and low predicted toxicity. Availability and implementation The datasets analyzed in this study are publicly available through the Gene Expression Omnibus database under accession numbers GSE63514 and GSE9750. Additional data supporting the findings of this study are available within the article and its Supplementary Materials.