Transcriptome Analysis Reveals Auto-Inflammatory Genes of Chronic Nonbacterial Osteomyelitis (CNO) Based on The Machine Learning

Abstract

Abstract Objectives: Chronic nonbacterial osteomyelitis (CNO) is an auto-inflammatory bone disorder. Since the origin and development of CNO involve many complex immune processes, resulting in delayed diagnosis and lack of effective treatment. Although bioinformatics analysis has been utilized to seek key genes and pathways of CNO, only a few bioinformatics studies that focus on CNO pathogenesis and mechanisms have been reported. This study aimed to identify key biomarkers that could serve as early diagnostic or therapeutic markers for CNO. Methods: Two RNA-seq datasets (GSE133378 and GSE187429) were obtained from the gene expression omnibus (GEO). Weighted gene co-expression network analysis (WGCNA) and differentially expressed genes (DEGs) analysis were conducted to identify the correlated genes associated with CNO. After that, the auto-inflammatory genes mostly associated with CNO were yielding based on the GeneCards database and the CNO prediction model, which was created by the LASSO machine learning algorithms. According to the receiver operating characteristic (ROC) curves, the accuracy of the model and auto-inflammatory genes was verified by utilizing external datasets (GSE7014). Eventually, we performed clustering analysis by ConsensusClusterPlus. Results: Totally, eighty CNO-related genes were identified, which were significantly enriched in the biological process of regulation of actin filament organization, cell-cell junction organization and gamma-catenin binding. The mainly enriched pathways were Adherens junction, Viral carcinogenesis and Systemic lupus erythematosus. Two auto-inflammatory genes with high expression in CNO samples were identified by combing an optimal machine learning algorithm (LASSO) with GeneCards database. The external validation dataset (GSE187429) was utilized for ROC analysis of prediction model and two genes, and the results have well validation efficiency. Then, we found that the expression of the two genes differed between clusters based on consensus clustering analysis. Finally, the ceRNA network of lncRNA and small molecule compounds of the two auto-inflammatory genes were predicted. Conclusion: Two auto-inflammatory genes, HCG18/has-mir-147a/UTS2/MPO axis and the signal pathways identified in this study can help us understand the molecular mechanism of CNO formation and provide candidate targets for the diagnosis and treatment of CNO.