CAS, a division of the American Chemical Society specializing in scientific information solutions, and Molecule.one, a tech-bio specialist in AI-based solutions for pharmaceutical chemistry, have established a strategic collaboration focused on the joint development of computer-aided synthesis design technologies to help accelerate scientific breakthroughs in early-stage drug discovery and aid chemists in the discovery of novel small molecules.
What does the partnership mean for the industry?
Leveraging their existing tech and expertise, including Molecule.one’s proprietary generative deep learning models and synthesis planning platform with chemist-first UI, and CAS’ world-class chemical reactions content collection and deep industry knowledge, the two firms will work together to develop solutions for efficient and innovative chemical synthesis planning.
Beyond their complementary capabilities, the collaboration by the two organizations is primarily fueled by their shared goal to empower scientists and accelerate breakthroughs.
The first joint offering, M1 RetroScore powered by CAS, will be part of the Molecule.one product suite and will replace their current synthetic accessibility scoring tool, RetroSAS. Leveraging the recent advances in generative model development by Molecule.one’s team, the groundbreaking solution uses machine learning models trained on the CAS best-in-class chemical reactions content to predict the likelihood of synthesis for novel molecules.
It provides all users with synthetic accessibility scores along with the corresponding potential top synthesis pathways for given targets. For users of CAS SciFindern, the tool will connect with reference reactions within the CAS SciFindern platform. M1 RetroScore powered by CAS is the first commercially available synthetic accessibility scoring tool trained on CAS content.
Additionally, Molecule.one and CAS are also collaborating to enhance the world-class retrosynthesis capabilities already available in CAS SciFindern. As part of their continued dedication to continuously improve their solutions, CAS will also incorporate Molecule.one’s deep learning models to improve its synthesis predictions and aid scientists in exploring potential synthetic routes for small molecules, thus accelerating scientific breakthroughs.
What does the partnership mean for Molecule.one?
“We are thrilled to bring to market the first generative models for chemistry trained on CAS’ content, as CAS’ strategic partner for synthetic accessibility and retrosynthesis deep learning enhancements. Generative AI has demonstrated amazing feats across multiple fields when trained on large datasets and I believe that there’s a clear need for bringing this to drug discovery,” commented Piotr Byrski, Molecule.one’s co-founder and Chief Executive Officer.
“I’m very proud of both CAS & Molecule.one teams, who have worked diligently over the past year to identify how we can bring together Molecule.one’s unique technology and expertise in artificial intelligence and CAS’ comprehensive chemical dataset,” Byrski said.
“I am excited that in the upcoming months and years, we will witness the outcomes of this collaboration and the significant impact it will have on the pharmaceutical industry.”
What does the partnership mean for CAS?
“Chemical reactions and retrosynthesis capabilities from CAS are unparalleled. By partnering with Molecule.one, we will bring a unique combination of capabilities to the market to solve challenges in synthesis planning – both early-stage screening for synthesizability as well as laboratory and scale-up synthesis design,” says Tim Wahlberg, Chief Product Officer for CAS.
“These solutions will continue to reduce time spent in the discovery process for scientists, empowering them with efficiencies throughout their innovation journey,” Wahlberg said.
M1 RetroScore powered by CAS officially launched on August 14. Learn more here.
