Purpose: We aimed to construct a competing endogenous RNA (ceRNA) network and explore the potential biomarkers in Crohn's disease (CD) via bioinformatics analysis. Validation of candidate biomarkers in a 2,4,6-trinitrobenzene sulfonic acid (TNBS) induced experimental colitis model and ceRNA network in an HCT116 cell line was also an aim, along with purposing to reveal the pathogenesis of CD. Methods: GSE102134 and GSE67106 datasets were obtained and used to screen the differentially expressed genes. WCGNA was applied to identify the relative model to construct the ceRNA network. Furthermore, the relationship between candidate gene and immune infiltration was investigated. Then, the expression of potential biomarkers was validated via qRT-PCR in a TNBS induced experimental colitis model. Finally, the ceRNA network was confirmed by RNAi experiments in an HCT116 cell line. Results: The ceRNA network, consisting of four lncRNAs, four miRNAs, and eight mRNAs, was constructed and the ROC analysis showed four mRNAs (PTGS2, LPL, STAT1, and TRIB2) had high diagnostic accuracy (AUC>0.9). In addition, upregulated PTGS2 was positively correlated with immune cell infiltration, including Natural killer cells, exhausted T-cells, monocytes, and Dendritic cells. The outcome of this TNBS induced experimental colitis model verified that the expression of PTGS2 and mir-429 was consistent with results of previous bioinformatics analysis. Furthermore, the predicted ceRNA network MIR3142HG/mir-429/PTGS2 were validated via RNA interference. Knockout of MIR3142HG decreased the mRNA level of PTGS2, whereas inhibition of mir-429 increased the mRNA level of PTGS2 in the HCT116 cell line. Conclusion: The exploration of the ceRNA network in this work might contribute to understanding the pathogenesis of CD. The constructed MIR3142HG/mir-429/PTGS2 ceRNA network may play a role in CD, and PTGS2 can be a potential immune-related biomarker in CD.