Identification of hub genes in peripheral blood and construction of a diagnostic nomogram model in ulcerative colitis

Ulcerative colitis (UC) is a chronic inflammatory bowel disease that significantly impacts patients' quality of life and incurs substantial healthcare costs. Currently, the diagnosis of UC primarily relies on invasive examinations, highlighting the necessity for rapid, non-invasive, and low-cost diagnostic methods. This study aims to identify a biomarker that can diagnose and assess UC through bioinformatics analysis of databases. We identified 470 differentially expressed genes (DEGs) from the GSE169568 expression matrix, with 225 genes being up-regulated and 245 down-regulated. Functional analyses using Gene Ontology, Kyoto Encyclopedia of Genes and Genomes, and Gene Set Enrichment Analysis demonstrated the enrichment of immune response-related processes. Weighted Gene Co-expression Network Analysis identified a core module associated with UC, comprising 169 genes that exhibited a strong correlation coefficient. By overlapping the UC-related module genes with DEGs, we determined 34 genes. We then assessed the predictive value of UC using the LASSO-Cox method and found that only NDUFB3 and SLIRP were significant. Furthermore, a difference in immune infiltration was observed between the UC and control groups, with macrophages shown to potentially interact with the hub genes. A nomogram model for UC diagnosis was developed based on these two genes, achieving an area under the curve of 0.898. The hub gene model was validated using the GSE94648 dataset, demonstrating good discrimination of the nomogram. Finally, we confirmed that both molecules were expressed at elevated levels in UC patients through clinical peripheral blood specimens. According to our research, NDUFB3 and SLIRP are promising biomarkers for UC and could aid in UC risk prediction. Future research should focus on clinical validation and the exploration of therapeutic interventions targeting the identified pathways and gene modules.