Founded in 2012, the startup Atomwise develops neural networks for the search for new drugs. The creators believe that the methods of deep machine learning can save time and money on the development of drugs and raise pharmaceuticals to a new level. The company has already developed more than 50 programs for the study of molecular interactions.
Atomwise invited specialists on various serious diseases such as Ebola, multiple sclerosis, and Alzheimer’s disease. Scientists described the biochemical features of the diseases in detail and this data was transferred for processing by artificial intelligence.
The deep machine learning system, called AtomNet, analyzes more than 10 million compounds every day. It predicts optimal molecular interactions and helps researchers to develop a structural drug design.
Neural networks allow for the discovery of suitable drugs without using laboratory tests. However, although the compounds found are effective in controlling the disease their impact on the human body is more difficult to predict. Therefore, traditional clinical trials are also conducted for potential AtomNet drugs.
During the first stage of financing, the startup attracted $45 million. The list of investors was headed by Monsanto Growth Ventures, Data Collective (DCVC), and B Capital Group.
The attraction of investments should help Atomwise become one of the most prolific and versatile research groups in the field of pharmaceuticals. The company will spend the bulk of the funds to provide a technical base and build a business infrastructure.
Among Atomwise’s customers are four out of the top ten U.S. pharmaceutical companies, including Merck and Monsanto. Atomwise collaborates with more than 40 major research universities, including Harvard, Stanford, Duke University, and Baylor College of Medicine.