Genetic diseases, driven by mutations in single or multiple genes, present complex therapeutic challenges due to their heterogeneous molecular mechanisms. Computational systems approaches—including dynamical (ODE/PDE), network-based modeling, statistical inference, and machine learning—provide systematic frameworks to identify drug targets by analyzing disease pathways and predicting intervention outcomes. This review discusses these methodologies, their integration with multi-omics data, and their application in prioritizing therapeutic candidates for monogenic and polygenic disorders. We further examine persistent challenges such as model scalability, experimental validation, and the need for patient-specific adaptations. By critically evaluating the synergies and limitations of these computational tools, this work underscores their potential to guide rational drug design while highlighting gaps for future methodological development in genetic disease research