Objective: Screening out potential herbal medicines and herbal ingredients for the treatment of gastric cancer based on transcriptomic analysis of immune infiltration and ferroptosis.Methods: Gene expression profiles of gastric tumour tissues and normal tissue samples were obtained from the GEO database and the samples were analysed for immune cell infiltration condition and differential expressed genes of ferroptosis. Key genes were screened by protein-protein interaction (PPI) and enrichment analysis, and molecular docking was used to predict and preliminary validate potential herbal and traditional Chinese medicine components for gastric cancer based on the key genes. Finally, RT-QPCR was used to validate the prediction results.Results: Immune cell infiltration analysis revealed high levels of infiltration of activated CD4 memory T cells, monocytes, M0 macrophages in gastric tumor tissues, while plasma cells and resting mast cells had higher levels of infiltration in the paraneoplastic tissues. Differential gene expression analysis identified 1,012 upregulated genes and 880 downregulated genes, of which 84 immune related differentially expressed genes such as CTSB, PGF and PLAU and 10 ferroptosis-related differentially expressed genes such as HSF1, NOX4 and NF2 were highly expressed in gastric cancer tissues. The results of enrichment analysis showed that they mainly involve 343 biological processes such as extracellular matrix organization and extracellular structural organization; 37 cellular components such as complexes of collagen trimer and basement membrane; 35 molecular functions such as signal receptor activator activity and receptor ligand activity; 19 regulatory pathways such as cytokine-cytokine receptor interactions and retinol metabolism. Finally, two key genes, TLR4 and KRAS, were selected and 12 herbal medicines such as Radix Salviae liguliobae, Rhizoma Coptidis, Rhizoma Polygoni cuspidati and 27 herbal ingredients such as resveratrol, salvianolic acid b were predicted on the basis of key genes. Molecular docking results showed that KRAS binds tightly to coumarin and magnolol, while TLR4 can bind tightly to resveratrol, curcumin, salvianolic acid b, shikonin. Subsequently, the effect of resveratrol and magnolol was experimentally verified.Conclusion: Herbal medicines such as S. liguliobae, Rhizoma Coptidis, Rhizoma P. cuspidati and herbal ingredients such as resveratrol, curcumin, salvianolic acid b may provide research directions and alternative therapeutic approaches for immunomodulation of TME and ferroptosis of tumour cells in gastric cancer.

Background: Ulcerative colitis (UC) is an idiopathic, chronic disorder characterized by inflammation, injury, and disruption of the colonic mucosa. However, there are still many difficulties in the diagnosis and differential diagnosis of UC. An increasing amount of research has shown a connection between ferroptosis and the etiology of UC. Therefore, our study aimed to identify the key genes related to ferroptosis in UC to provide new ideas for diagnosis UC.Methods: Gene expression profiles of normal and UC samples were extracted from the Gene Expression Omnibus (GEO) database. By combining differentially expressed genes (DEGs), Weighted correlation network analysis (WGCNA) genes, and ferroptosis-related genes, hub genes were identified and then screened using Lasso regression. Based on the key genes, gene ontology (GO) and gene set enrichment analysis (GSEA) analyses were performed. We used NaiveBeyas, Logistic, IBk, and RandomForest algorithms to build a disease diagnosis model using the hub genes. The model was validated using GSE87473 as the validation set.Results: Gene expression matrices of GSE87466 and GSE75214 were downloaded from the GEO database, including 184 UC patients and 43 control samples. A total of 699 DEGs were obtained. From FerrDb, 565 genes related to ferroptosis were identified. The 1,513 genes with the highest absolute correlation coefficient value in the MEblue module were obtained from WGCNA analysis. Five hub genes (LCN2, MUC1, PARP8, PLIN2, and TIMP1) were identified using the Lasso regression algorithm based on the overlapped DEGs, WGCNA-identified genes, and ferroptosis-related genes. GO and GSEA analyses revealed that 5 hub genes were identified as being involved in the negative regulation of transcription by competitive promoter binding, cellular response to citrate cycle_tca_cycle, cytosolic_dna_sensing pathway, UV-A, and beta-alanine metabolism. The logistic algorithm's values of the area under the curve (AUC)were 1.000 and 0.995 for training and validation cohorts, and sensitivity is 0.962, specificity is 1.000, respectively, as determined by comparing various methods. Conclusions:The previously described hub genes were identified as being intimately related to ferroptosis in UC and capable of distinguishing UC patients from controls. By detecting the expression of several genes, this model may aid in diagnosing UC and understanding the etiology and treatment of the disease.