Currently, small-molecule drug discovery is a long and costly process. AI-enabled virtual screening is the latest advance in accelerating the process. Computational screening provides a feasible trajectory to fully explore and interrogate the expansive universe of chemical compounds capable of binding protein targets of clinical interest.
X-37 uses AtomNet, an AI algorithm developed by Atomwise, to identify chemicals with drug-like structures to accelerate the drug discovery process for difficult-to-target proteins -- often characterized as undruggable.
X-37 combines the power of AtomNet with the capabilities of a team of seasoned and successful drug developers to bring promising small molecules into clinical development.
Combining the speed and scale of artificial intelligence with traditional aspects of drug discovery differentiates X-37 from other drug discovery companies. Target selection and identification of the most attractive drug binding sites on each target is carried out by seasoned chemists, data scientists, and biologists.
Target and binding site selection are followed by the application of AtomNet -- an AI algorithm that is tuned to find the most attractive drug hits while filtering out those with unwanted characteristics. This approach accelerates the drug discovery process by interrogating a massive chemical space for the best compounds with the speed of a CPU. The next step reverts to time-tested chemical Structure Activity Relationship (SAR) to select the best compound for IND-enabling studies and clinical trials.