Vanesa Brecher, Vasiliki Androutsopoulou, Serge Sicouri, Basel Ramlawi, Dimitrios V Avgerinos, Thanos Athanasiou, Dimitrios E Magouliotis
Under the random drug-pair split, the model achieved a micro-averaged AUPRC of 0.857 and a macro mean AUPRC of 0.825. In the held-out-drug evaluation, mean AUPRC decreased to 0.341 across 142 evaluable labels. Perturbation of the target features did not significantly change the reported performance metrics. We investigated case studies for interactions with bupropion and ritonavir with integrated gradients and identified molecular regions associated with known CYP-mediated interaction mechanisms.
Cardiovascular diseases remain a leading cause of global morbidity and mortality, driven by the complex interplay of genetic, epigenetic, transcriptomic, proteomic, and metabolic networks. Traditional reductionist approaches have inadequately captured this molecular complexity, motivating the emergence of integrative multi-omics and systems biology as foundational paradigms in cardiovascular research. This review provides a comprehensive, systems-level framework for translational discovery in cardiovascular disease, synthesizing advances in multi-omics technologies, network biology, artificial intelligence, and bioinformatics applied to conditions including thoracic aortic aneurysm, heart failure, valvular disease, and vascular remodeling. We survey the landscape of publicly available omics repositories and examine how transcriptomics, epigenomics, proteomics, and metabolomics are being integrated to decipher disease mechanisms. We outline network-based analytical frameworks encompassing protein-protein interaction networks, gene co-expression networks, hub gene analysis, and multiplex network modeling, highlighting their utility in identifying causal molecular drivers and therapeutically actionable targets. The growing contribution of machine learning, deep learning, and multimodal artificial intelligence to cardiovascular genomics is critically examined alongside challenges of interpretability and clinical validation. Finally, we address the translational interface between systems-level discovery and precision cardiovascular medicine, including polygenic risk stratification, multi-omics biomarker development, and the path from computational prediction to clinical implementation.