Computer-Generated Decision Making: How new drug candidates can be identified faster through automated molecular design
The discovery of new drugs and their development as medicines is a long, arduous, costly, and to some extent serendipitous process. Up to 90% of drug candidates fail during clinical development because they lack efficacy or show undesired side-effects. At the same time, good drugs are overlooked because researchers struggle with multi-dimensional optimization in the face of many unknowns. Gisbert Schneider is a professor at ETH Zurich, where he is the Chair for Computer-Assisted Drug Design. Having published over 400 scientific papers and winning several prestigious awards, Gisbert introduced the concept of adaptive, AI-based drug design to medicinal chemistry. The central idea is to partially transfer decision making from the human science manager to a machine intelligence that generates novel molecules with desired properties from scratch. At Falling Walls, Gisbert will talk about the problem of having too many – in fact almost infinite – choices when it comes to finding a treatment for a disease. He will show how a chemistry-savvy AI points out the molecules that are likely to succeed, thus leading to faster and more cost-effective drug discovery through informed decision-making.