The CIRCA project (Chemical Information Resources for Cognitive Analytics) is an informatics cloud solution that applies computer curation, automated natural language processing (NLP) and machine learning techniques to authoritative scientific content, with molecular information at the center.  CIRCA’s technical data corpus includes 22M patents, 28M MEDLINE abstracts, 2M full text articles and other public databases.  Each document is annotated to extract chemicals, reactions, polymers, inorganic materials, drugs, species, gene sequences, substances and effects.  We will detail some of the challenges in computer curation of this large data corpus, and the technical approaches we have developed to address them. Data is indexed as structured and full text fields, as well as for chemical structure similarity.  We will show how this database can be a foundation for computational discovery through cognitive applications. We will introduce new features, including search capabilities from mobile devices.<br/>

CIRCA: Your cheminformatics assistant

2019 Fall National Meeting

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.