VentureBeat presents: AI Unleashed - An exclusive executive event for enterprise data leaders. Network and learn with industry peers. Learn More
Atomwise, a startup using AI to accelerate drug discovery, today secured $123 million in funding. A spokesperson said the funds will enable the startup to scale its technology and team as it expands its portfolio of joint ventures with researchers at the University of Toronto, Duke University School of Medicine, Charles River, Bayer, Eli Lilly, Merck, and others.
Fewer than 12% of all drugs entering clinical trials end up in pharmacies, and it takes at least 10 years for medicines to complete the journey from discovery to the marketplace. Clinical trials alone take six to seven years, on average, putting the cost of R&D at roughly $2.6 billion, according to the Pharmaceutical Research and Manufacturers of America.
Atomwise claims its AtomNet platform can screen 16 billion chemical compounds for potential hits in under two days, expediting a process that would normally take months or years. (Hit identification is often the first step in drug discovery, where the right molecules — hits — that bind to a target protein and modify its function are identified.) AtomNet’s AI algorithms screen for things like potency and the degree to which a drug acts on a given site while guarding against toxicity, and they work on targets in hard-to-reach parts of the body — like the central nervous system — with a hit rate the company claims is 10,000 times better than wet-lab experiments.
In one instance, Atomwise claims it was able to inhibit a protein-protein interaction in the nervous system for a multiple sclerosis project, exploring 8.2 million molecules that were effective in animal trials and have been licensed to a pharmaceutical company in the U.K. In another, the company discovered a drug candidate with no previous antiviral application that blocked Ebola infectivity across virus strains from multiple epidemics.
An exclusive invite-only evening of insights and networking, designed for senior enterprise executives overseeing data stacks and strategies.
Atomwise also assists companies in developing molecular lead libraries. Using the physical or predicted structure of a target protein for initial screening via convolutional neural network (CNN) models, Atomwise draws on data from standard assays to perform counter-screens and narrow the pool of hits for properties like oral bioavailability or blood-brain barrier permeability. (CNNs are a class of deep neural networks most commonly applied to analyzing visual imagery.) Using AtomNet, Atomwise explores the chemical space around a handful of chemical scaffolds identified in hit discovery, optionally generating a list of compounds for scaffolds ready for lead optimization.
Through its Artificial Intelligence Molecular Screen (AIMS) initiative, Atomwise seeks proposals from academic researchers with interesting protein targets implicated in animal health, biotechnology, human biology, medicine, microbiology, plant biology, and virology. Awardees receive customized virtual screens through AtomNet and 72 small molecules predicted to bind to specific target proteins. In addition, they get support from Atomwise’s medicinal chemists and computational biologists and additional small molecules, dependent on the satisfaction of specific criteria.
Atomwise also leads the Chemistry for Academic Drug Discovery Startups (CADDS) ecosystem, a program designed to help universities and academic researchers build startups in preclinical drug discovery. In partnership with Y Combinator, Charles River, and others, Atomwise offers guidance in the areas of drug development, along with free resources and tools.
Atomwise claims its over 285 partnerships have already produced 17 patent-pending applications and several peer-reviewed publications. The company estimates its corporate partner deals are worth $5.5 billion, and Atomwise recently announced 15 research collaborations with universities to explore broad-spectrum therapies for COVID-19, targeting 15 unique and novel mechanisms of action.
Sanabil Investments and returning investor B Capital Group led the series B round, with participation from returning investors DCVC, BV, Tencent, Y Combinator, Dolby Ventures, and AME Cloud Ventures. (This brings the company’s total raised to almost $175 million, following a $45 million round in March 2018.) San Francisco-based Atomwise — a graduate of Nvidia’s Inception accelerator — says it has had more than 1,000 inbound applications over the past 12 months from academic and commercial research teams interested in using AtomNet for drug discovery.
VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact. Discover our Briefings.