Citadel Discovery

Many of the downstream failures in clinical drug development stem from decisions made during the earliest stages of small-molecule drug discovery. In particular, in-vitro molecule discovery techniques yield an insufficient number of molecules, most of which still require complex chemical optimization. Our platform provides an alternative framework for early drug discovery by leveraging our access to DNA-Encoded Libraries (DELs) to rapidly and cost-effectively generate readouts on tens of millions of small molecules that are used to train custom, project-specific AI models. Our clients use these models to efficiently evaluate hundreds of millions to billions of molecules, curated for desirable properties, yielding a far greater volume of potent molecular options. We have recently completed a successful proof of concept using our DEL drug discovery platform to enable the training, designing, and validating of a client’s AI discovery model. Along with biopharma companies, we have an on-going project with the Broad Institute’s Center for Development of Therapeutics to enable DEL-based AI for their discovery platform.