Identification of endocrine-disrupting chemicals targeting key OP-associated genes via bioinformatics and machine learning

Abstract

Osteoporosis (OP), a metabolic disorder predominantly impacting postmenopausal women, has seen considerable progress in diagnosis and treatment over the past few decades. However, the intricate interplay between genetic factors and endocrine disruptors (EDCs) in the pathogenesis of OP remains inadequately elucidated. The objective of this research is to examine the environmental pollutants and their regulatory mechanisms that could potentially influence the pathogenesis of OP, in order to establish a theoretical foundation for the targeted prevention and medical management of individuals with OP. Utilizing CTD and GEO datasets, network toxicology and bioinformatics analyses were conducted to identify target genes from a pool of 98 co-associated genes. Subsequently, a novel prediction model was developed employing a multiple machine learning algorithm. The efficacy of the model was validated based on the area under the receiver operating characteristic curve. Finally, real-time quantitative polymerase chain reaction (qRT-PCR) was used to confirm the expression levels of key genes in clinical samples. We have identified significant genes (FOXO3 and LUM) associated with OP and conducted Gene Ontology, Kyoto Encyclopedia of Genes and Genomes enrichment analysis, immune infiltration analysis, and molecular docking analysis. Through the analysis of these key genes, we have identified 13 EDCs that have the potential to impact OP. Several endocrine disruptors, such as Dexamethasone, Perfluorononanoic acid, genistein, cadmium, and bisphenol A, have been identified as notable environmental pollutants that impact the OP. Molecular docking analysis revealed significant binding affinity of major EDCs to the post-translational protein structures of key genes. This study demonstrates that EDCs, including dexamethasone, perfluorononanoic acid, genistein, cadmium, and bisphenol A, can be identified as important environmental pollutants affecting OP, and that FOXO3 and LUM have the potential to be diagnostic markers for OP. These results elucidate a novel association between EDCs regulated by key genes and the onset of OP.

Keywords

• Osteoporosis
• Endocrine-disrupting chemicals
• Bioinformatics analysis
• Machine learning
• Molecular docking