Drug discovery startup Atomwise recently raised $123 million bringing the total amount of capital the company has raised to date to almost $175 million. Atomwise plans to use the new financing to continue to scale its AI technology platform, grow its internal pipeline of small molecule drug programs, and continue portfolio development of joint venture companies. The company will also expand its work with corporate partners, such as Eli Lilly and Company, Bayer, Hansoh Pharmaceuticals, and Bridge Biotherapeutics.
Atomwise is a member of NVIDIA Inception, a program that supports AI startups with go-to-market support, expertise, and technology. Atomwise developed the AtomNet® model, the first convolutional neural network (CNN) for drug discovery, capable of screening over 16 billion compounds for potential hits in less than two days. AtomNet removes the physical barriers that have traditionally limited the success of small molecule drug discovery by allowing medicinal chemists and research teams to simulate the molecule screening process, condensing a process that would otherwise take months or years. This fundamental tool makes it possible for chemists to pursue hit discovery, lead optimization and toxicity predictions with precision and accuracy.
To scale from 10 million to 10 billion molecules, they used a custom CUDA kernel for pose prediction. Now they can run a 10 billion molecule screen in two days — that’s about 2,000 molecules per second on NVIDIA GPUs on the Amazon Web Services cloud.
To date, Atomwise has worked on 750 research collaborations to model more than 600 unique disease targets, and virtually screen over 16 billion novel small molecules for potential drug-protein interactions. The company’s technology and collaborations have generated 17 pending patent applications and several peer-reviewed publications. Atomwise has 285 active drug discovery partnerships with researchers at top research institutions around the world, including 15 research collaborations to explore broad-spectrum therapies for COVID-19, targeting unique and novel mechanisms of action.
Machine Learning, CUDA, Kubernetes and NVIDIA GPUs help companies like Atomwise lower drug discovery cost, increase access, and improve insights.