Machine learning Reveals ATM and CNOT6L as critical factors in Cataract pathogenesis

Objective

Cataract, a common age-related blinding eye disease, has a complex pathogenesis. This study aims to identify key genes and potential mechanisms associated with cataracts, offering new targets and insights for its prevention and treatment.

Methods

Transcriptomic data analysis and machine learning identified ATM serine/threonine kinase (ATM) and CCR4-NOT transcription complex subunit 6 like (CNOT6L) as key differential genes. Their roles in oxidative stress and apoptosis were validated using overexpression experiments in a cataract cell model. Immune-related analyses explored their regulatory effects on the immune microenvironment, and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis revealed potential mechanisms. In addition, in vitro experiments were conducted to evaluate the effects of ATM and CNOT6L overexpression on cell proliferation, oxidative stress, and apoptosis in lens epithelial cells.

Results

We identified 14 aging-associated differentially expressed genes, and ATM and CNOT6L were screened as key genes through machine learning and external dataset validation. KEGG pathway analysis indicated their involvement in base excision repair, ERBB signaling, and fatty acid metabolism pathways. Immune infiltration analysis revealed that ATM and CNOT6L positively correlated with CD8 T cells and B cells, and negatively correlated with regulatory T cells (Tregs), natural killer (NK) cells, and M1 macrophages. In vitro, overexpression of ATM and CNOT6L in cataract cell models promoted cell proliferation, inhibited apoptosis, reduced reactive oxygen species (ROS) and malondialdehyde (MDA) levels, and enhanced glutathione peroxidase (GSH-PX) activity.

Conclusion

ATM and CNOT6L play protective roles in cataract progression by reducing oxidative stress, inhibiting apoptosis, and regulating the immune microenvironment. They represent promising molecular targets for cataract prevention and treatment.