Overcoming the limits of drug discovery models with artificial intelligence and robotics
Current estimations of the cost to bring a new drug to market exceeds 2 billion dollars. A significant part of this cost can be attributed to two factors: the complexity of the discovery phase, which requires considerable investment in terms of time and resources, and the low percentage of candidate molecules going into clinical trials. Progresses in both computer hardware and in silico methods have accelerated and improved various aspects of the drug discovery cycle in design-make-test-analyze (DMTA) pharmaceutical chemistry. AI offers enormous possibilities where it is supported by scientifically correct models (and this has not happened often). We have made a great effort to combine AI and system biology and the results are remarkable. We have achieved considerable advances in accuracy and speed in the design of new molecules by combining semantic analysis, machine learning, and precision chemical synthesis. The new challenge is to integrate AI with robotics to create a DMTA model that turns traditional labs into true factories with high safety and efficiency, also limiting the use of animal model testing.
Stefano Piotto
15
mins
