SyntheticGestalt, a cutting-edge research and development company specializing in applying artificial intelligence (AI) to the life sciences, has announced a groundbreaking collaboration with Enamine, the world’s leading provider of chemical building blocks and integrated drug discovery services. The joint effort aims to create a suite of AI models that will significantly enhance the generation of synthetically accessible biologically active compounds with optimized physicochemical and ADME/Tox (Absorption, Distribution, Metabolism, Excretion/Toxicity) properties.
As part of this collaboration, SyntheticGestalt will leverage its expertise to develop a pre-trained AI model that will facilitate the discovery of synthesizable drug-like hit candidates. The model will utilize an extensive dataset comprising 38 billion compounds from Enamine’s REAL database, a comprehensive repository of make-on-demand compounds.
Enamine’s REAL database, known for its vast selection of 38 billion molecules, will be incorporated into SyntheticGestalt’s Drug Discovery Service. This service employs proprietary AI models to predict the physicochemical and ADME/Tox properties of compounds, offering instant suggestions for improved alternatives in cases where issues arise.
In an effort to streamline the drug discovery cycle, Enamine commits to synthesizing selected compounds within an impressive timeframe of 3-4 weeks. The company will also provide quality pharmacological in vitro profiling data through in-house tests, ensuring a comprehensive understanding of the compounds’ properties.
SyntheticGestalt, in turn, plans to enhance its pre-trained AI model using the data provided by Enamine. This initiative is expected to result in the creation of the world’s largest pre-trained model based on the 3D structures of compounds, with a focus on improving the predictive accuracy of SyntheticGestalt’s machine learning models. The joint research outcomes, including the pre-trained AI model, will be offered to selected parties interested in advancing drug discovery technologies. It could be recalled that AstraZeneca also Embarks on a similar AI-powered antibody-based drug discovery journey with a $247 million collaboration with Absci.
The Director of Business Development at Enamine, Iaroslava Kos, expressed enthusiasm about the collaboration, emphasizing the potential of AI/ML-powered computational designs in drug discovery. Kos highlighted the importance of joining forces with SyntheticGestalt to realize mutual goals in advancing therapeutics.
Koki Shimada, the CEO of SyntheticGestalt, acknowledged the historical challenges faced by machine learning in drug discovery and emphasized the significance of developing pre-training models using real-world application data. Shimada believes that the collaboration with Enamine and the use of the vast REAL database will propel AI drug discovery to new heights, drawing parallels with the revolutionary impact of large-scale pre-training on Language Models.
The joint research findings and the pre-trained AI model’s performance will be unveiled at NVIDIA’s annual event, NVIDIA GTC Japan AI Day, scheduled for March 2024. This collaboration marks a significant stride toward transforming the landscape of drug discovery through the fusion of AI and cutting-edge chemical expertise.