Computational Discovery of Selective Wee1 Inhibitors

This case study details the computationally-guided discovery of highly selective Wee1 inhibitors for the treatment of solid tumors. The program, driven by the Schrödinger Therapeutics Group, utilized a rigorous free energy perturbation (FEP+) workflow to optimize potency and selectivity against Wee1, a serine/threonine kinase critical for cell cycle regulation. The project commenced with the goal of identifying novel lead series with improved selectivity for Wee1 compared to polo-like kinase 1 (PLK1). Over six million design ideas were evaluated, leading to the identification of multiple novel chemotypes with enhanced selectivity. The study also highlights the use of machine learning and quantum mechanics strategies to optimize drug metabolism and pharmacokinetics (DMPK) properties. The development candidate, SGR-3515, emerged from this process, demonstrating superior selectivity and potency. The findings illustrate the effectiveness of advanced computational techniques in addressing selectivity challenges in drug discovery.