CAS is the American Chemical Society’s specialized division for scientific information solutions and Molecule.one is the Leading company in biotechnology. The two institutions have entered into a strategic collaboration focused on the joint development of computer-aided synthetic design techniques to accelerate scientific breakthroughs in early-stage drug discovery and to help chemists discover novel small molecules.
Leveraging its existing technologies and expertise, including Molecule.one’s proprietary generative deep learning models and synthesis planning platform with chemistry-first UI, and CAS’ world-class chemical reaction content collection technology and deep industry knowledge, the two The institutions will work together to enhance and develop efficient and innovative chemical synthesis planning solutions. In addition to complementary capabilities, the two institutions’ collaboration is driven by a shared goal of supporting scientists and accelerating breakthrough innovation.
Piotr Byrski, co-founder and CEO of Molecule.one, said: “As a strategic partner of CAS in the areas of synthetic accessibility and retrosynthesis deep learning augmentation technology, we are excited to bring the first batch of chemical generators trained with CAS content to Generative artificial intelligence has achieved amazing results in multiple fields if trained with large data sets, and I think it is an obvious need to apply it to drug discovery. I feel for CAS and the Molecule.one team Proud that they have worked hard over the past year to determine how we can bring together Molecule.one’s unique technology and expertise in artificial intelligence and CAS’ comprehensive chemical datasets. I am excited about the next few months In the years to come, we will see the fruits of this collaboration and its significant impact on the pharmaceutical industry.”
The first product to be launched jointly is the CAS-powered M1 RetroScore, which will become part of the Molecule.one product suite and will replace their current synthetic accessibility scoring tool, RetroSAS. Leveraging the latest advances in generative model development from the Molecule.one team, the solution predicts the likelihood of synthesizing novel molecules using machine learning models trained on best-in-class chemical reaction content provided by CAS. M1 RetroScore powered by CAS provides all users with a synthetic accessibility score, and the corresponding potential top synthetic pathways for a given target. For users of CAS SciFinder n , the tool will link to reference reactions within the CAS SciFinder n platform. M1 RetroScore powered by CAS is the first commercial synthetic accessibility scoring tool trained on CAS content.
Molecule.one and CAS are also collaborating to enhance the existing world-class reverse transcription synthesis capabilities of CAS SciFinder n . As part of its ongoing commitment to improving solutions, CAS will incorporate Molecule.one’s deep learning models to improve its synthesis predictions and help scientists explore potential synthetic routes for small molecules, thereby accelerating scientific breakthroughs.
Tim Wahlberg, Chief Product Officer of CAS, said: “CAS has unparalleled chemical reaction and retrosynthetic capabilities. By partnering with Molecule.one, we will bring to the market a unique combination of capabilities to solve the challenges of synthetic planning, including both From early screening for synthetic capabilities to laboratory and scale-up synthetic design. These solutions will continue to reduce the time scientists spend in the R&D process, making them more efficient throughout the innovation journey.”
The CAS-powered M1 RetroScore will be officially launched on August 14th and will be on display at Molecule.one’s booth #364 at ACS Fall 2023 in San Francisco, August 14-16 .
Source: https://www.businesswire.com/
