Integrating RNA-seq and machine learning to identify novel biotargets and high-affinity ligands for cardiovascular disease management

Cardiovascular diseases (CVDs) are the leading cause of mortality globally and, due to their heterogeneous nature, present significant clinical challenges. This study aims to identify novel biotargets for CVDs and propose potential inhibitors against them. The study leverages RNA-sequencing data in conjunction with machine learning (ML) techniques to uncover differentially expressed genes (DEGs) as potential biotargets for CVDs. Transcriptomic data was obtained from the Gene Expression Omnibus (GEO) database, and DESeq2 was used to identify DEGs. Machine learning (ML) models, random forest (RF), and support vector machines (SVM) were used to characterize DEGs and to rank top genes as biomarkers. Functional annotation of top hub genes was performed using clusterProfiler and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway and Gene Ontology (GO) analyses. Protein-protein interaction (PPI) networks were constructed using STRING. Molecular docking analyses were conducted using Biovia Discovery Studio and AutoDock Vina, targeting top genes with ligands sourced from the Drug Gene Interaction Database as repurposable targets. Comprehensive analysis of DEGs led to the identification of multiple hub genes and predictive biomarkers for CVD treatment. Using ML algorithms for biomarker feature prediction, we identified the top DEGs, which included interleukin-6 (IL6), tumor necrosis factor (TNF), myosin heavy chain-6 (MYH6), apolipoprotein E (APOE), low-density lipoprotein receptor (LDLR), proprotein convertase subtilisin/kexin type-9 (PCSK9), angiotensin-converting enzyme (ACE), actin alpha-2 (ACTA2), activated protein kinase (AMP)-activated non-catalytic subunit γ-2 (PRKAG2), and cardiac type troponin T2 (TNNT2). Network and PPI analyses further highlighted the significance of the identified DEGs, which were then targeted for discernment of high-affinity binding ligands from clinically approved and relevant drugs using docking studies. Biomarker-guided approaches for the prediction, evaluation, diagnosis, and treatment of CVDs hold substantial promise for clinical application. The identification of clinically approved ligands targeting the top genes from DEGs in CVD patients might facilitate more effective personalized treatment regimens, improving patient outcomes and ultimately transforming CVD management.