SyntheticGestalt and Enamine, both leaders in the fields of AI and chemical building blocks, have joined forces to develop advanced AI models that can generate biologically active compounds for drug discovery. This collaborative effort aims to optimize the properties of these compounds and streamline the drug discovery process.
The Enamine REAL database, which contains an impressive 38 billion molecules, will be integrated into SyntheticGestalt’s Drug Discovery Service. This service utilizes proprietary AI models to assess the physicochemical and ADME/Tox properties of compounds. If any issues arise, the service will provide alternative compounds with improved properties.
Enamine will then synthesize the selected compounds within a remarkably fast period of 3-4 weeks. The company will also conduct in-house pharmacological in vitro profiling tests, shortening the entire discovery cycle. This streamlined process promises to accelerate the development of new therapeutic options.
Additionally, SyntheticGestalt will enhance its pre-trained AI model by incorporating data from Enamine. This will create the largest pre-trained model in the world based on 3D compound structures, significantly improving the predictive accuracy of SyntheticGestalt’s machine learning models.
The resulting models will be made available for joint research to interested parties and will be showcased at NVIDIA’s prestigious event, NVIDIA GTC Japan AI Day, in March 2024. This collaboration represents a major milestone in AI drug discovery, offering immense potential for the development of new and effective drugs.
Iaroslava Kos, Director of Business Development at Enamine, expressed enthusiasm for the collaboration, stating, “The promise provided by AI/ML powered computational designs in the discovery of new drugs cannot be underestimated.” She also emphasized the shared goals and expertise that both companies bring to the table.
Koki Shimada, CEO of SyntheticGestalt, highlighted the significance of developing pre-training models using real-world application data. He believes that their ultra-large pre-trained model will revolutionize AI drug discovery, similar to the impact of large-scale pre-training on Language Models.
This groundbreaking collaboration represents a major leap in AI drug discovery and holds immense promise for the future of pharmaceutical research and development.
The source of the article is from the blog scimag.news