AI generated molecules
Bob Krijnen
The success rate of searching for new molecules to produce new medication is low. 90% don't make it to the finish line. The amount of money involved is astronomical. Developments in artificial intelligence (AI) can reverse this trend.
Recent developments may possible accelerate and improve the research results. The term AI and the concept of 'deep learning' are not new. One of the advances is in high-performance computing (Computers with excessive computing- and thinking power). Together with the availability of large datasets and new guidelines for the implementation of deep neural networks (DNNs), they are thought to make a difference. DNN’s already outperform people in various areas. Think of facial recognition, autonomous driving and robot surgery.

Worldwide, several companies have sought partnerships with AI startups. Together they have started AI programs to search for new molecules for the development of new medication. What is still lacking is proof of success: the testing of AI generated molecules and medication. This needs to be demonstrated in clinical research. The silent race for molecules and new medication, born from AI, has begun.
Neither did we.