Unleashing machine learning to solve medicinal chemistry

We're transforming drug design with a rich dataset built by screening millions of compounds against thousands of proteins to generate billions of data points, ideal for training advanced machine learning models.

Finding better drugs faster by focusing on what matters: Data

All major machine learning problems have been solved by new data, not new algorithms. The data to solve drug discovery doesn't exist yet.

We’re closing the data gap to find the next blockbuster drugs

Finding better drugs faster by focusing on what matters: Data

All major machine learning problems have been solved by new data, not new algorithms. The data to solve drug discovery doesn't exist yet.

We’re closing the data gap to find the next blockbuster drugs

We’re building a massive, proprietary dataset of protein-molecule interactions

We’re building a massive, proprietary dataset of protein-molecule interactions

A rapid search over massive chemical space

Central to our innovative technology is a dynamic, cyclical engine: we continuously harness data, iterate machine learning, and refine our approach, with each cycle only taking a few months.

Powerful, scalable software enabling us to rapidly design new molecules

Our innovative software can build and refine novel chemical matter much more likely to have the right activities.

We’ve measured billions of interactions and trained hundreds of models to identify hits

17.4B

PROTEIN-MOLECULE INTERACTIONS MEASURED

6.7M

ML-DESIGNED MOLECULES MADE AND TESTED

Where we’re headed:

Rich Dataset

Building our platform to measure 500 protein targets against 20M molecules each in the next year

Internal Programs

Developing new medicines, starting with two targets in oncology

Partnerships

Working with biopharma partners to quickly and efficiently explore new molecule opportunities