Better medicines faster by automated design.

The project integrates constructive artificial intelligence (AI) into the pharmaceutical drug development progress. The central theme is to transfer decision making from the human science manager to a machine intelligence that generates novel molecules with desired properties from scratch. We devise suitable algorithms and implement software for fully automated molecular design in “low data” scenarios. The computer-generated molecules are biologically validated in several proof-of-concept applications, partly together with industry partners. From a managerial point of view, this new approach leads to a fundamental shift of the sources of scientific innovation. Innovation management in drug discovery is no longer dependent on the serendipitous identification of potential medicines, but innovation can now be rationalized and engineered.

Tags: drug discovery, artificial intelligence, automation, science management, medicine