Atomwise have created AtomNet, which uses deep learning technology for small molecular discovery. These algorithms allow researchers to analyse simulations of molecules, test compounds and predict their effects on the human body quickly and more accurately. This also happens much sooner in the drug discovery process timeline, meaning less wasted hours and money.


Drug discovery is a costly and long process requiring a multidisciplinary team, with some drugs taking up to 5.5 years to complete and costing billions of dollars. This technology is dramatically increasing the rate of drug delivery, as their systems are capable of screening 10 million compounds each day. They are already working with some of the top research institutes and pharmaceutical companies in the world (including Harvard University and Stanford University), and, the more projects they work on, the more their AI systems learn.


The drug discovery market was valued at $35.2 billion dollars in 2017, estimated to grow the $71 billion in 2025. Atomwise smart technology has the potential to optimise the drug discovery process. Most recently, they discovered a drug candidate that blocked Ebola infectivity from multiple epidemics – a process that could have taken years under normal drug discovery.


Atomwise have raised $51.6 million, with their most recent round closing at $45 million led by Monsanto Growth Ventures, Data Collective (DCVC) and B Capital Group. They have also started to involve academic scientists in their work through their Artificial Intelligence Molecular Screen (AIMS) awards program. Scientists apply and identify the disorder they hope to treat, and successful applicants receive compounds that Atomwise predicts are most likely to work as that drug. This growing community of scientists can then test in their labs, creating a scientific ecosystem working towards creating new drug solutions faster and more effectively.