De Novo drug design is a long standing goal of computational drug discovery. Despite more than 20 years of efforts and significant advances, both in methods development and in available hardware, existing approaches are still not widely used in drug discovery. One of the most significant stumbling blocks limiting usability and wider adoption of de novo design approaches is often limited synthetic feasibility of proposed molecules.
Talk will start with introduction to de novo design problem and continue with history and overview of variety of historical and emerging approaches. SWARM - a novel de novo method with emphasis on synthetic accessibility and decoupling between structure generation and scoring function – together with its applications in drug discovery will be described.