Objective

Identifying new medicines is an extraordinarily hard and complex problem. Chemists have to select and profile molecules from a huge search space to find those that combine activity at a biological target with a suitable property profile for further development. Given the practical scale limitations of in vitro experiments, it is desirable to have computational tools to conduct de-novo drug design. This presentation will provide an introduction to the emergence of artificial intelligence (AI) in our everyday life and the recent advances that have seen AI surpass human performance in many applications. The core of the presentation will describe: 1) research at AstraZeneca to develop autonomous tools to complete retrosynthesis and synthesis planning tasks, 2) our research efforts targeted towards autonomous de-novo compound generation and performance against simplified problems, 3) project applications, and 4) important research in this field from leading academic laboratories. Finally, the presentation will provide a vision for the integration of AI and robotics for autonomous design-make-test cycles from Hit molecules to clinical Candidate Drugs