Text Mining and Drug Discovery Analysis: A Comprehensive Approach to Investigate Diabetes-Induced Osteoporosis

Purpose: Osteoporosis (OP) and diabetes are prevalent diseases in orthopedic and endocrinology departments, with OP potentially arising as a complication of diabetes. However, the mechanism underlying diabetes-induced osteoporosis (DOP) remains enigmatic, and drug discovery in this domain is restricted, hindering research into the DOP's etiology and treatment. With the ultimate goal of preventing OP in diabetic patients, the objective of this study is to mine the genes and pathways linked to DOP using bioinformatics and databases. Method: The present study employed text mining as the initial approach to retrieve genes commonly associated with diabetes and OP. Subsequently, functional annotation was conducted to investigate the roles and functionalities. In order to explore the interactions between proteins relevant to DOP, we constructed protein-protein interaction (PPI) networks. Furthermore, to obtain key genes and candidate drugs for DOP treatment, we conducted drug-gene interaction (DGI) analysis, complemented by a thorough examination of transcriptional factors (TFs)-miRNA co-regulation. Results: The results through text mining revealed 110 genes that are commonly associated with both diabetes and OP. Subsequent enrichment analysis narrowed down the list to 95 symbols that were involved in PPI analysis. After DGI analysis, we identified 7 genes targeted by 11 drugs, which represent candidates for treating DOP. Conclusion: This study unveils ANDECALIXIMAB, SILTUXIMAB, OLOKIZUMAB, SECUKINUMAB, and IXEKIZUMAB as promising potential drugs for DOP treatment, demonstrating the significance of utilizing text mining and pathway analysis to investigate disease mechanisms and explore existing therapeutic options.