Biologicals like anti-tumor necrosis factor (TNF) therapy for Crohn’s disease (CD) are safe and effective but there is a significant rate of primary and secondary nonresponse in the patients. In this study, we applied a computational approach to discover novel drug therapies for anti-TNF refractory CD in silico. We use a transcriptome dataset (GSE100833) for the anti-TNF refractory CD patients from NCBI GEO. After co-expression analysis, we specifically investigated the extent of protein–protein interactions among genes in clusters based on a protein–protein interaction database, STRING. Pathway analysis was performed using the clEnrich function based on KEGG gene sets. Co-expressed genes in cluster 1, 2, 3, 4, up or down-regulated genes and all differentially expressed genes are highly connected. Among them, cluster 1, which is highly enriched for chemokine signaling, also showed enrichment for cytokine–cytokine receptor interaction and identifies several drugs including cyclosporin with known efficacy in CD. Vorinostat, histone deacetylase inhibitors, and piperlongumine, which is known to have inhibitory effect on activity of NF-κB, were also identified. Some alkaloids were also selected as potential therapeutic drugs. These finding suggest that they might serve as a novel therapeutic option for anti-TNF refractory CD and support the use of public molecular data and computational approaches to discover novel therapeutic options for CD. © 2020, The Author(s).