Identification of hub genes and molecular networks involved in alveolar bone resorption based on bioinformatics analysis.

Periodontitis is the leading cause of tooth loss, with alveolar bone resorption serving as its fundamental pathological feature. Identifying key regulatory genes and molecular mechanisms is essential for enhancing preventive and therapeutic strategies. Nevertheless, reliable biomarkers and comprehensive regulatory networks have yet to be elucidated. This study aims to systematically identify candidate alveolar bone resorption hub genes (ABRHUB) and their regulatory networks through bioinformatics analysis. Our objective is to establish a bioinformatics foundation for generating hypotheses about the mechanisms underlying periodontitis progression and to identify potential targets for future experimental validation. The GSE16134 dataset extracted from the GEO database was utilized to screen differentially expressed genes in patients with periodontitis and explore the significance of alveolar bone resorption in the prevention and treatment of this condition. Furthermore, these genes were intersected with modular genes screened by weighted gene co-expression network analysis (WGCNA) as well as osteoclast-related genes obtained from molecular characterization databases, and co-expression differential genes (co-DEGs) were obtained. The co-DEGs were subjected to in-depth analysis using various machine learning algorithms. The validation set GSE10334 and qRT-PCR analysis were utilized for double validation, ultimately confirming ABRHUB. The diagnostic value and immune relevance of ABRHUB were further evaluated, and the potential miRNAs and lncRNAs associated with these key genes were predicted using relevant databases. A significant correlation exists between alveolar bone resorption and periodontitis. Through differential analysis of multiple databases and the integration of machine learning techniques, we identified four key ABRHUB genes: NEDD9, P2RX5, CSF1R and NPR3. Subsequently, we validated these genes using the GSE10334 validation set and quantitative reverse transcription polymerase chain reaction (qRT-PCR) analyses, and constructed diagnostic and risk models to elucidate the potential utility of ABRHUB in predicting alveolar bone resorption. Furthermore, the calibration curves we established further validated the accuracy of the model predictions. Ultimately, based on these ABRHUB genes, we constructed a lncRNA-miRNA-mRNA molecular regulatory network, providing a significant bioinformatics foundation for future studies. Four genes associated with osteoclast function, namely NEDD9, P2RX5, CSF1R and NPR3, may serve as potential candidate biomarkers for alveolar bone resorption. Notably, the down-regulation of miR-1260b, miR-1224-5p, miR-3156-5p and miR-4286 may contribute to the progression of periodontitis by promoting the expression of NEDD9, P2RX5 and CSF1R.