Identification of telomere maintenance related biomarkers and regulatory mechanisms in chronic obstructive pulmonary disease by machine learning algorithm.

Chronic obstructive pulmonary disease (COPD) is a progressive respiratory disease that accelerates the aging process of the lung. Despite advancements in managing symptoms and preventing acute exacerbations, significant gaps remain in our understanding of the complex mechanisms that drive disease progression and contribute to mortality in COPD. In our work, we have successfully identified a set of five robust biomarkers (including RMI1, RAD51, RAD52, SNRNP70 and CHEK1). These biomarkers effectively distinguish COPD samples from normal samples, with area under the curve (AUC) value greater than 0.65 in the training set and greater than 0.80 in the validation set. Gene set enrichment analysis (GSEA) analysis showed that the main enrichment pathways were Non-alcoholic fatty liver disease, Spliceosome, Oxidative phosphorylation, etc. We also found these five genes had high accuracy in the diagnosis of COPD in both the training and verification sets. Molecular docking showed that the TOP5 small drug molecules acting with CHEK1 were U-0126, KN-62, BX-912, LY-294,002 and AZD-7762. The results of real-time reverse transcriptase-polymerase chain reaction (RT-qPCR) showed that there were significant differences in the expression of SNRNP70 and RAD52 between COPD and control samples (p < 0.05).