No abstract

Ulcerative colitis (UC) is a chronic inflammatory disease that is prone to recurrent attacks. It has complex pathogenesis, which is closely related to genetics, constitution, dietary habits, and environmental factors. From the comprehensive Gene Expression Omnibus database (GEO), we retrieved gene expression profiles and classified 197 cases of UC into three subgroups for the purpose of predicting the basic molecular characteristics for different types of ulcerative colitis. As expected, each group showed its own clinical peculiarity and way of presentation. In this article, consensus clustering was used to divide the sample into three. The WGCNA analysis was applied to evaluate specific modules and reveal transcriptional differences among the subgroups. Subsequently, pathway and function enrichment analysis was conducted based on WGCNA. In subgroup Ⅰ, fructose and mannose metabolism pathway and cell cycle 11/36 pathway are up-regulated, which could be an indicator of exacerbation. Furthermore, the hematopoietic cell lineage pathway, which was significantly up-regulated in subgroup Ⅱ, might be indicating a disease correlation. In subgroup Ⅲ, the gene expression pattern of the peroxisome pathway is similar to the normal group, which may indicate an early stage of UC. Although no significant prognostic difference existing among the groups, there were significant differences in their underlying biological characteristics. This suggests that transcriptome classifications also represent risk factors for different disease states and ages. In summary, the bioinformatics techniques used in this study contribute to identifying molecular subtypes for diagnosing human ulcerative colitis. The transcriptome classification of UC cases suggests that each subgroup may have its own gene expression pattern and pathway, providing further personalized treatment guidance for patients with ulcerative colitis.