Identification of diagnostic genes for acute appendicitis and appendiceal cancer based on bioinformatics and machine learning

Abstract Background Appendiceal cancer is a rare malignancy that is often discovered incidentally in acute appendicitis patients. The diagnosis and treatment of this disease are challenging due to its low incidence and nonspecific symptoms. In this study, we used bioinformatics and machine learning to identify novel diagnostic genes for appendiceal cancer in acute appendicitis. Methods To identify differentially expressed genes (DEGs) between acute appendicitis and appendiceal cancer, we analyzed two GEO datasets (GSE9575 and GSE7535). Then, to select the most important differential expressed genes for diagnosis, we used three machine learning methods (LASSO logistic regression, SVM-RFE, and RandomForest). We also performed functional enrichment analysis and CIBERSORT analysis to explore the biological functions and immune cell infiltration of the feature genes. Results We identified 45 DEGs between appendiceal cancer and acute appendicitis, of which 23 were upregulated and 22 were downregulated in appendiceal cancer. The ROC curve analysis showed that the feature genes had an AUC of 1.000 for discriminating appendiceal cancer from acute appendicitis. The functional enrichment analysis revealed that the feature genes were mainly involved in cell cycle regulation, DNA replication, and DNA repair pathways. The CIBERSORT analysis showed that the feature genes were associated with different immune cell types, such as B cells, T cells, macrophages, and dendritic cells. Conclusion We identified novel critical genes involved in the progression of appendiceal cancer and provided potential biomarkers for its diagnosis. Our findings also suggested that the feature genes might play a role in modulating the immune response in appendiceal cancer.