Predicting Prognostic Bidirectional Molecular Signatures Associating Myocardial Infarction and Lung Cancer: An In-Silico Perspective

Dr. Dhruva Nandi,Rajiv Janardhanan,Piyush Agrawal

DOI: https://doi.org/10.1101/2024.11.15.623806

2024-11-18

Abstract:Myocardial Infarction (MI) and lung cancers substantially contributors to morbidity and mortality. They share common risk factors such as smoking, and hypertension. There is a pressing need to identify bidirectional molecular signatures linking MI and lung cancer relationship to create a unique triage to improve the clinical outcomes of the patient. In the current work, we extracted the common differentially expressed genes (DEGs) between MI and lung cancer and identified 2339 upregulated and 2973 downregulated genes in MI datasets; 952 upregulated and 653 downregulated genes in LUAD; 1466 upregulated and 2816 downregulated genes in LUSC. Among these, 27 genes (11 upregulated and 16 downregulated) were common across MI, LUAD, and LUSC. Functional enrichment analysis revealed shared biological processes, such as inflammatory response, and cell differentiation. KEGG pathway analysis highlighted common pathways such as B cell receptor signaling, TNF signaling. A protein-protein interaction study with STRING revealed a variety of interaction partners; miRNA-mRNA network analysis characterizes conserved binding sites for 6 genes while overall survival analysis for lung cancer patients shows a significant association with prognosis. In addition, machine learning models built using the Support Vector Machine and Random Forest algorithms demonstrated high AUROC values of 0.79 and 0.81 respectively on balanced dataset in the classifying MI patients from non-MI patients. Lastly, based on drug repurposing analysis, we proposed FDA-approved drugs such as Venetoclax, Lomitapide, Regorafenib that could potentially target these genes, indicating novel therapeutic options for the co-occurring conditions of MI and lung cancer. Our findings highlight the similarities in molecular makeup between lung cancer and MI, providing information for future investigations and therapeutic approaches.

Bioinformatics

What problem does this paper attempt to address?

The problem that this paper attempts to solve is to identify the bidirectional molecular signatures between myocardial infarction (MI) and lung cancer in order to create unique triage methods and improve the clinical outcomes of patients. Specifically, by analyzing the differentially expressed genes (DEGs) shared by the two diseases, the researchers explored the roles of these genes in biological processes such as inflammatory response and cell differentiation, as well as the signaling pathways in which they are involved, such as the B - cell receptor signaling pathway and the TNF signaling pathway. In addition, the study further verified the roles of these common genes in predicting patient prognosis through methods such as protein - protein interaction (PPI) analysis, miRNA - mRNA network analysis, and survival analysis, and evaluated the potential of these genes in distinguishing patients with myocardial infarction from those without through machine - learning models. Finally, based on drug re - utilization analysis, FDA - approved drugs that may target these genes were proposed, providing new treatment options for patients with both diseases.