Valence's deep learning platform enables the design of small-molecule drug candidates in novel regions of chemistry, followed by rapid optimization against project-specific potency, selectivity, safety, and pharmacology criteria. The company's prediction tools work in areas where only small amounts of training data are available, thereby overcoming a critical limitation of existing machine-learning technologies in drug discovery.

The combination of Valence's machine-learning technologies with Charles River's end-to-end capabilities will accelerate discovery efforts from hit design through lead optimization, the companies said.

Through the collaboration, Charles River clients will have the option to access Valence's platform to support their drug discovery efforts. Clients can expect increased diversity in chemical matter being investigated, rapid optimization, and improved success rates for drug discovery projects.

Financial details of the partnership were not disclosed.