Grant: Project Grant

As it did in other sectors, artificial intelligence (AI) is expected to bring a technlogical breakthrough in the early stage of drug discovery, where billions of molecules are computationally evaluated to predict which are the best drug candidates. The larger the chemical library, the more chances to find a drug that perfectly matches a given protein therapeutic target. The collection of around 30 billion such molecules available from commercial vendors is immensely vast. But it is a drop in the ocean of molecules that could be synthesized, in part because commercial vendors restrict themselves to a relatively limited set of traditional chemical reactions. Academic chemistry groups in universities across Canada continuously invent novel chemical reactions that could be used to make entirely new types of drug-like molecules. Some of these molecules could unlock the discovery of drugs in challenging therapeutic areas. But the mission of chemists in academia is not to synthesize billions of molecules for experimental screening. As a result, academic chemistry remains closed to the early, high-throughput stage of drug discovery. Here, we will use computational techniques to open academic chemistry to drug discovery. We will transform the chemical reactions invented in universities across Canada into a virtual Canadian chemical library composed of billions of unique molecules, unavailable from other sources. We will use state-of-the-art AI technologies developed in our laboratories to predict which are bioactive for any given therapeutic protein target. We will synthesize the predicted molecules and confirm experimentally their bioactivities. As the project progresses, it will expand in breadth as our network of chemists grows, and in depth, as we refine our AI-based algorithms. We will demonstrate the value of our integrated Canadian platform by discovering chemical starting points towards novel therapies against challenging therapeutic targets.