{"id":368836,"date":"2023-12-05T07:49:33","date_gmt":"2023-12-05T12:49:33","guid":{"rendered":"https:\/\/platohealth.ai\/artidock-from-receptor-ai-next-generation-ai-docking-that-beats-diffdock-and-alphafold-latest\/"},"modified":"2023-12-05T07:49:56","modified_gmt":"2023-12-05T12:49:56","slug":"artidock-from-receptor-ai-next-generation-ai-docking-that-beats-diffdock-and-alphafold-latest","status":"publish","type":"post","link":"https:\/\/platohealth.ai\/artidock-from-receptor-ai-next-generation-ai-docking-that-beats-diffdock-and-alphafold-latest\/","title":{"rendered":"ArtiDock from Receptor.AI: Next-generation AI Docking That Beats DiffDock and AlphaFold-latest","gt_translate_keys":[{"key":"rendered","format":"text"}]},"content":{"rendered":"
Receptor.AI\u00a0has announced ArtiDock, the best-in-class model for \u201cAI docking,” which predicts the binding poses of small molecule ligands in protein binding pockets with unprecedented speed and accuracy.<\/p>\n
The company\u00a0performed a comprehensive comparison of ArtiDock with the best modern AI docking techniques and with the most widely used conventional docking programs, Vina and Gold.<\/p>\n
ArtiDock not only beats the previous best-performer,\u00a0DiffDock<\/a>\u00a0by a large margin but also performs on par with classical docking programs and rivals the recently announced next generation of\u00a0AlphaFold-latest<\/a>\u00a0that is capable of predicting protein-ligand complexes.<\/p>\n The key to success is a proprietary technique of data augmentation combined with a fast and lightweight model architecture. ArtiDock is trained on a mixture of artificial and real complexes and is aware of a much larger set of combinations of intermolecular interactions than are present in the resolved structures. This gives it a dramatic boost in accuracy and predictive power.<\/p>\n Instead of relying on only about 20.000+\u00a0resolved protein-ligand complexes, millions of artificial binding pockets, which closely follow the experimentally observed statistical properties of the real ones, are generated. The algorithmic approach to such generation is currently being published.<\/p>\n The benchmark included the commonly used\u00a0Astex docking dataset<\/a>\u00a0and the novel\u00a0PoseBuster dataset<\/a>, which is dedicated to challenging the quality of AI docking algorithms. The usual RMSD metric was used, as well as an additional PoseBusters-Valid set of metrics. The latter also assesses steric clashes and the conformer quality of the ligand.<\/p>\n The ArtiDock outperforms all its rivals (both AI and docking) with a large margin for the Astex dataset using either RMSD or PB-Valid metrics, as shown in\u00a0Figure 1.\u00a0<\/p>\nHow does it work?<\/h2>\n
Benchmark<\/h2>\n