PatCID is a collection of molecular structures to speed up discovery of new molecules, automatically and at scale with Deep Search. This large-scale collection was created by processing molecular images in United States, European, Japanese, Korean, and Chinese patents. In this We have applied a graph-based visual recognition model to obtain high-precision conversion of these molecular images (MolGrapher, ICCV2023, accepted). Contact us for early access.
In this example we will demonstrate how to interact with Deep Search for
- Querying molecules by similarity or substructure
- Given a list of patents, show the molecules that are mentioned
- Given a list of molecules, list the patents mentioning them
👉 Run the chemistry_patcid.ipynb notebook.
The content of this notebook requires access to Deep Search capabilities which are not available on the public access system.
Contact us if you are interested in exploring the enterprise-level Deep Search capabilities.