Integrating Microarray Analysis, Machine Learning, and Molecular Docking to Explore the Mechanism of Doxorubicin-induced Cardiotoxicity

Doxorubicin (DOX) is a chemotherapeutic agent widely used for the treatment of various cancers; however, its clinical use is limited by its cardiotoxicity. However, the underlying molecular mechanisms remain poorly understood, hindering the development of effective preventive and treatment strategies. This study aimed to identify core target genes and explore the mechanisms involved in DOX-induced cardiotoxicity by integrating microarray analysis, machine learning, and molecular docking.

Differential expression analysis was performed using microarray data from DOX-induced cardiotoxic samples and healthy controls. Multiple machine learning algorithms were applied to identify core target genes. The predictive performance of these genes was evaluated using receiver operating characteristic (ROC) curves. Molecular docking was conducted to evaluate the binding affinity of DOX to the target genes. Functional analysis was performed to investigate potential toxic mechanisms.

In total, 276 differentially expressed genes were identified between DOX-induced cardiotoxicity samples and controls. The support vector machine algorithm demonstrated the best performance, leading to the identification of five core target genes: RAP1A, CTLA4, OR2M1P, TRIM53, and LOC149837. The ROC curves confirmed the strong predictive power of these genes, with area under the curve values greater than 0.85. Molecular docking showed stable binding between DOX and the target genes. Functional analysis suggested that the Rap1 signaling pathway and immune system regulation may be involved in DOX-induced cardiotoxicity.

Traditional toxicological studies often rely on limited experimental approaches that do not fully capture the complexity of disease mechanisms. The integration of microarray analysis, machine learning, and molecular docking in this study offers a comprehensive framework for investigating the toxicological pathways of DOX-induced cardiotoxicity, thereby providing insights into therapeutic development and safety regulations.

By combining microarray analysis, machine learning, and molecular docking, we identified five key target genes associated with DOX-induced cardiotoxicity. Functional analysis further suggested the involvement of the Rap1 signaling pathway and immune system regulation in DOX-induced cardiotoxicity. These findings offer insights into the molecular mechanisms underlying DOX-induced cardiotoxicity and have implications for the development of protective strategies and therapeutic interventions.