Amazon Web Services (AWS) has launched a new artificial intelligence platform for drug discovery, creating a single workflow for researchers to use without writing code for the first time.
From discovery to testing
The system, Amazon Bio Discovery, enables scientists to generate and assess potential drug candidates for testing by using natural language prompts. Test results are then automatically returned to the platform, creating a continuous feedback loop for users.
In a press release, Rajiv Chopra, Vice President, AWS Healthcare AI and Life Sciences, Amazon Web Services, said: “AI agents make powerful scientific capabilities accessible to all drug researchers, not just those with computational expertise.
This combination of cutting-edge AI and the robust, secure infrastructure allow scientists to accelerate antibody discovery in ways that weren’t possible before.”
The platform also allows scientists to train AI models using their own experimental data without building dedicated machine learning infrastructure. According to AWS, these models remain private and are not used to train other shared systems.
Early adopters
Early adopters of the platform include the Memorial Sloan Kettering Cancer Center, which used the tool to generate around 300,000 potential antibody candidates, 100,000 of which were selected for testing. This reduced the typical timeline from a year to just a few weeks.
Also in the press release, Nai-Kong Cheung, Enid A. Haupt Chair of Pediatric Oncology, Memorial Sloan Kettering Cancer Center, said: “As researchers, we spent 20 years just to prove that the first generation of antibody worked, and then we spent another 13 years getting it into the human form before getting FDA approval.
This path has been very inefficient. Patients come here with a clock. We need results sooner.”
In addition to MSK Cancer Center, Bayer, the Broad Institute, Fred Hutch Cancer Center and Voyager Therapeutics are among the early adopters of AWS’s new AI offering.
Featured image: JHVEPhoto on Adobe Stock
