Abstract
BACKGROUND: Rape pollen allergy is a common allergic reaction disorder that affects the health and life of patients seriously. The research on ceRNA regulatory network in rape pollen allergy is poor. METHODS: High throughput whole-transcriptome sequencing was conducted on rape pollen allergic samples and non-allergic samples. Differentially expressed microRNAs (DEmiRNAs), circRNAs (DEcircRNAs), long non-coding RNA (DElncRNAs), mRNA (DEmRNAs) were identified and a ceRNA regulatory network was constructed by Cytoscape. Functional enrichment analyses were performed on DEmRNAs in the ceRNA network. Then, the least absolute shrinkage and selection operator (LASSO) regression model was used to identify characteristic genes for rape pollen allergy. The receiver operating characteristic (ROC) curve was used to evaluate the diagnostic ability of characteristic genes. RESULTS: A total of 25 DEmiRNAs, 258 DEcircRNAs, 304 DElncRNAs, and 383 DEmRNAs in the allergic group compared with the non-allergic group were uncovered, respectively. A ceRNA network containing 21 miRNAs, 57 circRNAs, 28 lncRNAs, and 33 mRNAs was generated with 139 nodes and 160 edges. The signal transduction-related processes, immune-related processes, the ion, inorganic substance, and hormone regulation processes were associated with mRNAs in the ceRNA network. The results of pathway enrichment illustrated that mRNAs in the ceRNA were significantly linked to IL-17 signaling pathway, inflammatory mediator regulation of trp channels, GMP-PKG signaling pathway, signaling by GPCR, and GPCR downstream signaling pathway. Then, five characteristic genes (KCNQ3, CCR5, FOSB, CFAP43, and PRKG1) were defined by the LASSO algorithm. The AUC values of these genes indicated that these genes had a powerful discrimination ability in discriminating allergic samples from non-allergic samples. CONCLUSION: Taken together, we revealed the ceRNA regulatory network in rape pollen allergy and excavated five characteristic genes (KCNQ3, CCR5, FOSB, CFAP43, and PRKG1) with the diagnostic value that may be a potential target in diagnosis and treatment.