This work investigates the structural properties of anti-cancer drug molecles through graph. Theoretic and computational approaches. Each drug is modeled as a molecular graph, and the metric dimension and edge metric dimension are calculated to analyze vertex and edge resolvability. The formulations and computations were implemented using Python algorithms, enabling efficient extraction of graph invariants. Furthermore, a regression model was applied to the calculated dimensions to identify correlations and predictive patterns among the molecular structures. The results demonstrate that combining graph theory with computational tools and statistical modeling offers valuable insights into the structural complexity of anti- cancer drugs, with potential applications in drug design and cheminformatics.