Amazon has pitched an AI-powered drug discovery platform to the biopharma industry that it says brings together computational methods and wet-lab workflows into a single application and can slash research time for new medicines.
Amazon Web Services (AWS) said the 'lab-in-the-loop' platform – dubbed Amazon Bio Discovery (ABD) – is launching with more than 40 AI biological foundation models, agentic assistants to help select the best models for research projects and interpret results, and a network of contract research organisation (CRO) partners to carry out wet-lab experiments whose results can be fed back into the system and guide the next design stages.
It deploys a four-stage process, starting with the evaluation of models and building of an in silico workflow, and followed by assistant-led experimental design, analysis of computational results, and transfer of the most promising candidates to wet labs operated by Ginkgo Bioworks, Twist Bioscience, or A-Alpha Bio for validation. Users can access the AI model library, upload their own custom models, or use third-party tools.
The launch marks an acceleration of AWS's efforts to provide services and tools for drug developers, building on other initiatives like its HealthOmics analysis platform and HealthLake service for identifying patient cohorts for clinical trials.
It also puts the company in competition with other tech giants offering AI-powered drug discovery, like NVIDIA, Alphabet/Isomorphic Labs, and OpenAI, which has just signed a strategic-level alliance with Novo Nordisk.
One factor that could help AWS break into the category is its established presence in providing enterprise-level cloud services, including to 19 of the top 20 pharma multinationals, on which the ABD platform is built.
Rajiv Chopra, vice president of healthcare AI and life sciences at AWS, told Reuters that Amazon Bio Discovery is already being put through its paces at Bayer, the Broad Institute, and Voyager Therapeutics.
A project at Memorial Sloan Kettering Cancer Center has validated the platform, generating 100,000 antibody candidates that could be sent for wet-lab testing in a matter of weeks, as described in this white paper.
In a blog post, AWS' Ryan Green and Jiwon Kim write that ABD "makes lab-in-the-loop accessible and scalable across your entire research organisation."
They add: "Lab-in-the-loop drug discovery shouldn't take time away from science. It should be accessible to everyone doing the science, whether you're designing the computational workflows or running the experiments that test your hypotheses."
https://pharmaphorum.com/news/amazon-launches-its-ai-drug-discovery-platform
No comments:
Post a Comment
Note: Only a member of this blog may post a comment.